Technology Governance Wiki - User contributions [en]

Governance by markets

2026-10-05T09:56:59Z

Sergey Abramov: /* Data as an input to market power (Sergey) */

= Glossary =
= [https://wiki.ivir.nl/index.php?title=Data_as_an_input_to_market_power Data as an input to market power] (Sergey) =
<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals [[https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1]]. '''Data''' is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. '''Market power''' refers to a firm's ability to influence the price, output, or other market conditions in its favour, without being constrained by competitive forces [[https://www.tutor2u.net/economics/reference/what-is-meant-by-market-power 2]].</p>
<p>'''Data as an input to market power''' means that access to information can help a business limit the competitive pressure it faces. Information about searches, clicks or purchases can improve services and advertising. If competitors cannot obtain comparable information at a reasonable cost, they may struggle to offer an attractive alternative. This can allow a platform to raise prices or worsen quality or privacy protections without losing enough business to make doing so unprofitable. This matters even when a service is free, because competition also concerns advertising charges, personal information and the choices available. Data can be used repeatedly and by several businesses without being used up. However, access can be restricted through contracts and technical controls. Nevertheless, holding a large dataset alone does not establish market power.</p>
<p>'''Google Search''' illustrates this mechanism. Google controls over 91% of global search engine traffic [[https://www.statista.com/statistics/1381664/worldwide-all-devices-market-share-of-search-engines/ 3]] and uses information about users’ searches and the results they click to improve the relevance of its rankings. A larger volume of searches gives it more opportunities to learn, including about unusual queries that smaller search engines encounter less often. Better results can attract further use, producing additional information and reinforcing the advantage. A new search engine may therefore struggle to attract users while lacking the information needed to improve its service. Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. </p>
<p>A platform’s control over data access can shape who is able to compete with it. By determining whether, and on what terms, rivals can access that information, a platform can influence the competitive opportunities available to others. This suggests that regulation should examine access to strategically important data alongside prices and market shares. Article 6(11) of the EU Digital Markets Act [[https://eur-lex.europa.eu/eli/reg/2022/1925/oj/eng#:~:text=giving%20their%20consent.-,11.%C2%A0,-The%20gatekeeper%20shall 4]] reflects this approach: certain large search providers must, on request, share specified search data with rival search providers on fair and non-discriminatory terms, with relevant personal data anonymised. The aim is to help rivals compete while protecting users. A data advantage does not, by itself, establish unlawful conduct.</p>
<p>Would access to equivalent data enable a rival to compete effectively? The concept cannot answer this on its own. Additional data may eventually add little value, become outdated, or have substitutes. Firms also need the skills and computing resources to use it. Google’s default positions in browsers and devices can further restrict rivals’ access to users even if data become available. The unresolved issue is how much competitive advantage comes from data itself and how much comes from these other resources and arrangements. </p>

= Two Sided Markets (Ceren) =
<p>In two-sided markets, two sides of a market interact through an intermediary. These markets typically have some externality,<code class="mwt-code">Carl Shapiro and Hal R Varian, ''Information Rules: A Strategic Guide to the Network Economy'' (Harvard Business School Press 2010), 183.</code><sup></sup> where the decisions of the sides affect the other. The intermediary also has multitudes of effects on both sides, mainly identified in the related literature through its pricing.</p><p>An example of a two sided market is payment systems, whereby the credit card providing bank is the intermediary between the customers and merchants. In this example, banks provide credit cards to customers to be utilized in their payments, and merchants accept payment via multiple credit card systems to accommodate payments from customers holding credit cards from such providers[https://doi.org/10.1257/jep.23.3.125 .]</p><p>The main power of the intermediary materializes in their pricing toward the merchant (and the merchant’s bank in interchange fees in case of payment being transferred through different banks). Customers will be expected to be more loyal, holding a few credit cards, whereas merchants accept payment from many banks to accommodate all customers’ payment methods. Banks will provide more subsidies to customers to attract them and compete aggressively on the customer side, making their platform more desirable to work with by merchants, who will in turn be more accepting of the platform’s fees to gain the business of the larger volume of customers. However, dynamic pricing (increasing fees after gaining entry and loyalty) may be a way for these platforms to abuse their power.</p><p>The concept may be derailed from these logical / analytical systems of operation or expected outcomes when public policy interferes with the market. For example, net neutrality rules regulate what internet service providers (the intermediary) can charge content providers (one side) who aim to reach the consumers (the other side). In this case the above considerations and the concept are halted.</p>

= Switching costs and lock-in (Kate) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul)  =
=== Excludability and Rivalrousness ===<p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p><p>'''Excludability''': Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p><p>'''Rivalrousness''': Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p><p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>

=== AI Training and the (non-)excludability of information. ===<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. Conversely, data may be rivalrous not in its absolute depletion by consumption, but rather the benefit or competitive advantage usage of that data may confer: If all AI firms are able to train on the exact same data (all available data), the utility of that data may still increase, but the competitive advantage of possessing that data is exhausted.</p>

=== Free-Rider Problems & Public Data ===<p>As any party can benefit from publicly available data at no cost or low cost, the incentive for private market actors to fund its creation is comparatively weak. Consequently, public bodies (and less frequently open-source foundations and non-profits) often assume the responsibility for the provision of this type of data. A classical example is weather data and its creation by national weather agencies.</p>

= Interoperability (Sisi) =

#What is interoperability?[[https://www.ibm.com/think/topics/interoperability 1]]<p>Interoperability is a “standards-based” approach that enables different IT systems to communicate with minimal end-user interaction to exchange data and “share functionality”. Each system also allows the other system to “read, update, modify, and analyze” data with minimal human interaction. [[https://www.oracle.com/nl/interoperability/ 2]]</p><p>Interoperability works by having different systems share data by employing the same standards for data formats (e.g., Extensible Markup Language or JavaScript Object Notation), exchange, and vocabulary to title data elements. These standards are agreed upon “in advance” by organizations, industry types, and user groups. Then, software developers build the system based on the model agreed upon.</p>

#Concrete example from a digital technology or platform<p>Interoperable systems are vital in healthcare, commerce, and public safety. A notable example includes the integration of Electronic Health Records (HER). This system allows different healthcare providers to securely access and share personal patient information regardless of the specific EHR system used.[[https://data.europa.eu/en/news-events/news/interoperability-solutions-bridging-systems-connected-future 3]] An example of such a system is the GNU Health Hospital Information System, which is a model employed by small private offices and large national public health systems.[[https://interoperable-europe.ec.europa.eu/collection/gnu-health/solution/gnu-health-hospital-information-system 4]]</p><p>For example, a health care professional may access lab results and patient history data from various platforms to enable them to make a diagnosis.</p>

#One governance implication: what does this concept explain about how platforms exercise power?<p>Interoperability is a powerful tool because it helps provide a “seamless user experience” and optimizes data sharing between separate systems. It helps companies and organizations access data from multiple sources and improve efficiency, such as accelerating workflows and reducing the possibility of errors by pooling data from its sources.</p><p>Interoperability standards are mandated by law on the technical side to ensure “critical systems” are compatible, especially in public safety and health. [[https://www.ibm.com/think/topics/interoperability 5]]</p><p>In Europe, interoperability is governed by the European Health Data Space regulation (EHDS); the “first common EU data space dedicated to a specific sector as part of the European strategy for data.”</p><p>This Regulation aims to benefit EU citizens such as patients, researchers, healthcare professionals, and industry players by establishing a common framework for the exchange of electronic health data within the EU. In addition, it enables certain data to be “reused” for public purpose and innovative scientific research. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 6]]</p>

#Where does the concept break down or fail to explain something<p>While the EDHS harmonizes a legal and technical framework for sharing EU data, its shortcoming is its gradual implementation. The EDHS entered into force in 2025; however, most of its obligations will not apply until 2029. This leaves four years for health care companies to use interoperability in a deregulated manner. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 7]]</p><p>Furthermore, there are actual barriers to achieving full interoperability in healthcare, including data silos (which isolate data collections and hinder data exchange), lack of standardization (where one system records blood pressure as free text, another system stores it as a structured numeric value), and deficient infrastructure, which requires significant technological upgrading.</p><p>The burden remains on health companies to decide which standards to adopt and how much to invest in stable technological infrastructure to strengthen their data governance and interoperability. [[https://www.tigahealth.com/how-to-overcome-healthcare-interoperability-challenges/ 8]]</p>

= "Economies of Scale (Jihae)" =
= [[Network Effects (Indirect) - Sebastián]] =
= [[Monopsony - (Atmika)]] =

Governance by markets

2026-10-05T09:56:12Z

Sergey Abramov: /* Data as an input to market power (Sergey) */

= Glossary =
= [[https://wiki.ivir.nl/index.php?title=Data as_an_input_to_market_power|Data as an input to market power]] (Sergey) =
<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals [[https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1]]. '''Data''' is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. '''Market power''' refers to a firm's ability to influence the price, output, or other market conditions in its favour, without being constrained by competitive forces [[https://www.tutor2u.net/economics/reference/what-is-meant-by-market-power 2]].</p>
<p>'''Data as an input to market power''' means that access to information can help a business limit the competitive pressure it faces. Information about searches, clicks or purchases can improve services and advertising. If competitors cannot obtain comparable information at a reasonable cost, they may struggle to offer an attractive alternative. This can allow a platform to raise prices or worsen quality or privacy protections without losing enough business to make doing so unprofitable. This matters even when a service is free, because competition also concerns advertising charges, personal information and the choices available. Data can be used repeatedly and by several businesses without being used up. However, access can be restricted through contracts and technical controls. Nevertheless, holding a large dataset alone does not establish market power.</p>
<p>'''Google Search''' illustrates this mechanism. Google controls over 91% of global search engine traffic [[https://www.statista.com/statistics/1381664/worldwide-all-devices-market-share-of-search-engines/ 3]] and uses information about users’ searches and the results they click to improve the relevance of its rankings. A larger volume of searches gives it more opportunities to learn, including about unusual queries that smaller search engines encounter less often. Better results can attract further use, producing additional information and reinforcing the advantage. A new search engine may therefore struggle to attract users while lacking the information needed to improve its service. Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. </p>
<p>A platform’s control over data access can shape who is able to compete with it. By determining whether, and on what terms, rivals can access that information, a platform can influence the competitive opportunities available to others. This suggests that regulation should examine access to strategically important data alongside prices and market shares. Article 6(11) of the EU Digital Markets Act [[https://eur-lex.europa.eu/eli/reg/2022/1925/oj/eng#:~:text=giving%20their%20consent.-,11.%C2%A0,-The%20gatekeeper%20shall 4]] reflects this approach: certain large search providers must, on request, share specified search data with rival search providers on fair and non-discriminatory terms, with relevant personal data anonymised. The aim is to help rivals compete while protecting users. A data advantage does not, by itself, establish unlawful conduct.</p>
<p>Would access to equivalent data enable a rival to compete effectively? The concept cannot answer this on its own. Additional data may eventually add little value, become outdated, or have substitutes. Firms also need the skills and computing resources to use it. Google’s default positions in browsers and devices can further restrict rivals’ access to users even if data become available. The unresolved issue is how much competitive advantage comes from data itself and how much comes from these other resources and arrangements. </p>

= Two Sided Markets (Ceren) =
<p>In two-sided markets, two sides of a market interact through an intermediary. These markets typically have some externality,<code class="mwt-code">Carl Shapiro and Hal R Varian, ''Information Rules: A Strategic Guide to the Network Economy'' (Harvard Business School Press 2010), 183.</code><sup></sup> where the decisions of the sides affect the other. The intermediary also has multitudes of effects on both sides, mainly identified in the related literature through its pricing.</p><p>An example of a two sided market is payment systems, whereby the credit card providing bank is the intermediary between the customers and merchants. In this example, banks provide credit cards to customers to be utilized in their payments, and merchants accept payment via multiple credit card systems to accommodate payments from customers holding credit cards from such providers[https://doi.org/10.1257/jep.23.3.125 .]</p><p>The main power of the intermediary materializes in their pricing toward the merchant (and the merchant’s bank in interchange fees in case of payment being transferred through different banks). Customers will be expected to be more loyal, holding a few credit cards, whereas merchants accept payment from many banks to accommodate all customers’ payment methods. Banks will provide more subsidies to customers to attract them and compete aggressively on the customer side, making their platform more desirable to work with by merchants, who will in turn be more accepting of the platform’s fees to gain the business of the larger volume of customers. However, dynamic pricing (increasing fees after gaining entry and loyalty) may be a way for these platforms to abuse their power.</p><p>The concept may be derailed from these logical / analytical systems of operation or expected outcomes when public policy interferes with the market. For example, net neutrality rules regulate what internet service providers (the intermediary) can charge content providers (one side) who aim to reach the consumers (the other side). In this case the above considerations and the concept are halted.</p>

= Switching costs and lock-in (Kate) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul)  =
=== Excludability and Rivalrousness ===<p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p><p>'''Excludability''': Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p><p>'''Rivalrousness''': Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p><p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>

=== AI Training and the (non-)excludability of information. ===<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. Conversely, data may be rivalrous not in its absolute depletion by consumption, but rather the benefit or competitive advantage usage of that data may confer: If all AI firms are able to train on the exact same data (all available data), the utility of that data may still increase, but the competitive advantage of possessing that data is exhausted.</p>

=== Free-Rider Problems & Public Data ===<p>As any party can benefit from publicly available data at no cost or low cost, the incentive for private market actors to fund its creation is comparatively weak. Consequently, public bodies (and less frequently open-source foundations and non-profits) often assume the responsibility for the provision of this type of data. A classical example is weather data and its creation by national weather agencies.</p>

= Interoperability (Sisi) =

#What is interoperability?[[https://www.ibm.com/think/topics/interoperability 1]]<p>Interoperability is a “standards-based” approach that enables different IT systems to communicate with minimal end-user interaction to exchange data and “share functionality”. Each system also allows the other system to “read, update, modify, and analyze” data with minimal human interaction. [[https://www.oracle.com/nl/interoperability/ 2]]</p><p>Interoperability works by having different systems share data by employing the same standards for data formats (e.g., Extensible Markup Language or JavaScript Object Notation), exchange, and vocabulary to title data elements. These standards are agreed upon “in advance” by organizations, industry types, and user groups. Then, software developers build the system based on the model agreed upon.</p>

#Concrete example from a digital technology or platform<p>Interoperable systems are vital in healthcare, commerce, and public safety. A notable example includes the integration of Electronic Health Records (HER). This system allows different healthcare providers to securely access and share personal patient information regardless of the specific EHR system used.[[https://data.europa.eu/en/news-events/news/interoperability-solutions-bridging-systems-connected-future 3]] An example of such a system is the GNU Health Hospital Information System, which is a model employed by small private offices and large national public health systems.[[https://interoperable-europe.ec.europa.eu/collection/gnu-health/solution/gnu-health-hospital-information-system 4]]</p><p>For example, a health care professional may access lab results and patient history data from various platforms to enable them to make a diagnosis.</p>

#One governance implication: what does this concept explain about how platforms exercise power?<p>Interoperability is a powerful tool because it helps provide a “seamless user experience” and optimizes data sharing between separate systems. It helps companies and organizations access data from multiple sources and improve efficiency, such as accelerating workflows and reducing the possibility of errors by pooling data from its sources.</p><p>Interoperability standards are mandated by law on the technical side to ensure “critical systems” are compatible, especially in public safety and health. [[https://www.ibm.com/think/topics/interoperability 5]]</p><p>In Europe, interoperability is governed by the European Health Data Space regulation (EHDS); the “first common EU data space dedicated to a specific sector as part of the European strategy for data.”</p><p>This Regulation aims to benefit EU citizens such as patients, researchers, healthcare professionals, and industry players by establishing a common framework for the exchange of electronic health data within the EU. In addition, it enables certain data to be “reused” for public purpose and innovative scientific research. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 6]]</p>

#Where does the concept break down or fail to explain something<p>While the EDHS harmonizes a legal and technical framework for sharing EU data, its shortcoming is its gradual implementation. The EDHS entered into force in 2025; however, most of its obligations will not apply until 2029. This leaves four years for health care companies to use interoperability in a deregulated manner. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 7]]</p><p>Furthermore, there are actual barriers to achieving full interoperability in healthcare, including data silos (which isolate data collections and hinder data exchange), lack of standardization (where one system records blood pressure as free text, another system stores it as a structured numeric value), and deficient infrastructure, which requires significant technological upgrading.</p><p>The burden remains on health companies to decide which standards to adopt and how much to invest in stable technological infrastructure to strengthen their data governance and interoperability. [[https://www.tigahealth.com/how-to-overcome-healthcare-interoperability-challenges/ 8]]</p>

= "Economies of Scale (Jihae)" =
= [[Network Effects (Indirect) - Sebastián]] =
= [[Monopsony - (Atmika)]] =

Data as an input to market power

2026-10-05T09:55:04Z

Sergey Abramov: Created page with "<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals. https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1 '''Data''' is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. '''Market power''' refers to a firm's ability to influence the price, output, or other market con..."

<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals. [[https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1]] '''Data''' is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. '''Market power''' refers to a firm's ability to influence the price, output, or other market conditions in its favour, without being constrained by competitive forces. [[https://www.tutor2u.net/economics/reference/what-is-meant-by-market-power 2]] </p><p>'''Data as an input to market power''' means that access to information can help a business limit the competitive pressure it faces. Information about searches, clicks or purchases can improve services and advertising. If competitors cannot obtain comparable information at a reasonable cost, they may struggle to offer an attractive alternative. This can allow a platform to raise prices or worsen quality or privacy protections without losing enough business to make doing so unprofitable. This matters even when a service is free, because competition also concerns advertising charges, personal information and the choices available. Data can be used repeatedly and by several businesses without being used up. However, access can be restricted through contracts and technical controls. Nevertheless, holding a large dataset alone does not establish market power.</p><p>'''Google Search''' illustrates this mechanism. Google controls over 91% of global search engine traffic [[https://www.statista.com/statistics/1381664/worldwide-all-devices-market-share-of-search-engines/ 3]] and uses information about users’ searches and the results they click to improve the relevance of its rankings. A larger volume of searches gives it more opportunities to learn, including about unusual queries that smaller search engines encounter less often. Better results can attract further use, producing additional information and reinforcing the advantage. A new search engine may therefore struggle to attract users while lacking the information needed to improve its service. Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. </p><p>A platform’s control over data access can shape who is able to compete with it. By determining whether, and on what terms, rivals can access that information, a platform can influence the competitive opportunities available to others. This suggests that regulation should examine access to strategically important data alongside prices and market shares. Article 6(11) of the EU Digital Markets Act [[https://eur-lex.europa.eu/eli/reg/2022/1925/oj/eng#:~:text=giving%20their%20consent.-,11.%C2%A0,-The%20gatekeeper%20shall 4]] reflects this approach: certain large search providers must, on request, share specified search data with rival search providers on fair and non-discriminatory terms, with relevant personal data anonymised. The aim is to help rivals compete while protecting users. A data advantage does not, by itself, establish unlawful conduct.</p><p>Would access to equivalent data enable a rival to compete effectively? The concept cannot answer this on its own. Additional data may eventually add little value, become outdated, or have substitutes. Firms also need the skills and computing resources to use it. Google’s default positions in browsers and devices can further restrict rivals’ access to users even if data become available. The unresolved issue is how much competitive advantage comes from data itself and how much comes from these other resources and arrangements. </p>

Governance by markets

2026-10-05T09:47:38Z

Sergey Abramov: /* Glossary */

= Glossary =
= Data as an input to market power (Sergey) =
<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals [[https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1]]. '''Data''' is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. '''Market power''' refers to a firm's ability to influence the price, output, or other market conditions in its favour, without being constrained by competitive forces [[https://www.tutor2u.net/economics/reference/what-is-meant-by-market-power 2]].</p><p>'''Data as an input to market power''' means that access to information can help a business limit the competitive pressure it faces. Information about searches, clicks or purchases can improve services and advertising. If competitors cannot obtain comparable information at a reasonable cost, they may struggle to offer an attractive alternative. This can allow a platform to raise prices or worsen quality or privacy protections without losing enough business to make doing so unprofitable. This matters even when a service is free, because competition also concerns advertising charges, personal information and the choices available. Data can be used repeatedly and by several businesses without being used up. However, access can be restricted through contracts and technical controls. Nevertheless, holding a large dataset alone does not establish market power.</p><p>'''Google Search''' illustrates this mechanism. Google controls over 91% of global search engine traffic [[https://www.statista.com/statistics/1381664/worldwide-all-devices-market-share-of-search-engines/ 3]] and uses information about users’ searches and the results they click to improve the relevance of its rankings. A larger volume of searches gives it more opportunities to learn, including about unusual queries that smaller search engines encounter less often. Better results can attract further use, producing additional information and reinforcing the advantage. A new search engine may therefore struggle to attract users while lacking the information needed to improve its service. Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. </p><p>A platform’s control over data access can shape who is able to compete with it. By determining whether, and on what terms, rivals can access that information, a platform can influence the competitive opportunities available to others. This suggests that regulation should examine access to strategically important data alongside prices and market shares. Article 6(11) of the EU Digital Markets Act [[https://eur-lex.europa.eu/eli/reg/2022/1925/oj/eng#:~:text=giving%20their%20consent.-,11.%C2%A0,-The%20gatekeeper%20shall 4]] reflects this approach: certain large search providers must, on request, share specified search data with rival search providers on fair and non-discriminatory terms, with relevant personal data anonymised. The aim is to help rivals compete while protecting users. A data advantage does not, by itself, establish unlawful conduct.</p><p>Would access to equivalent data enable a rival to compete effectively? The concept cannot answer this on its own. Additional data may eventually add little value, become outdated, or have substitutes. Firms also need the skills and computing resources to use it. Google’s default positions in browsers and devices can further restrict rivals’ access to users even if data become available. The unresolved issue is how much competitive advantage comes from data itself and how much comes from these other resources and arrangements. </p>
= Two Sided Markets (Ceren) =
<p>In two-sided markets, two sides of a market interact through an intermediary. These markets typically have some externality,<code class="mwt-code">Carl Shapiro and Hal R Varian, ''Information Rules: A Strategic Guide to the Network Economy'' (Harvard Business School Press 2010), 183.</code><sup></sup> where the decisions of the sides affect the other. The intermediary also has multitudes of effects on both sides, mainly identified in the related literature through its pricing.</p><p>An example of a two sided market is payment systems, whereby the credit card providing bank is the intermediary between the customers and merchants. In this example, banks provide credit cards to customers to be utilized in their payments, and merchants accept payment via multiple credit card systems to accommodate payments from customers holding credit cards from such providers[https://doi.org/10.1257/jep.23.3.125 .]</p><p>The main power of the intermediary materializes in their pricing toward the merchant (and the merchant’s bank in interchange fees in case of payment being transferred through different banks). Customers will be expected to be more loyal, holding a few credit cards, whereas merchants accept payment from many banks to accommodate all customers’ payment methods. Banks will provide more subsidies to customers to attract them and compete aggressively on the customer side, making their platform more desirable to work with by merchants, who will in turn be more accepting of the platform’s fees to gain the business of the larger volume of customers. However, dynamic pricing (increasing fees after gaining entry and loyalty) may be a way for these platforms to abuse their power.</p><p>The concept may be derailed from these logical / analytical systems of operation or expected outcomes when public policy interferes with the market. For example, net neutrality rules regulate what internet service providers (the intermediary) can charge content providers (one side) who aim to reach the consumers (the other side). In this case the above considerations and the concept are halted.</p>

= Switching costs and lock-in (Kate) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul)  =
=== Excludability and Rivalrousness ===<p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p><p>'''Excludability''': Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p><p>'''Rivalrousness''': Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p><p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>

=== AI Training and the (non-)excludability of information. ===<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. Conversely, data may be rivalrous not in its absolute depletion by consumption, but rather the benefit or competitive advantage usage of that data may confer: If all AI firms are able to train on the exact same data (all available data), the utility of that data may still increase, but the competitive advantage of possessing that data is exhausted.</p>

=== Free-Rider Problems & Public Data ===<p>As any party can benefit from publicly available data at no cost or low cost, the incentive for private market actors to fund its creation is comparatively weak. Consequently, public bodies (and less frequently open-source foundations and non-profits) often assume the responsibility for the provision of this type of data. A classical example is weather data and its creation by national weather agencies.</p>

= Interoperability (Sisi) =

#What is interoperability?[[https://www.ibm.com/think/topics/interoperability 1]]<p>Interoperability is a “standards-based” approach that enables different IT systems to communicate with minimal end-user interaction to exchange data and “share functionality”. Each system also allows the other system to “read, update, modify, and analyze” data with minimal human interaction. [[https://www.oracle.com/nl/interoperability/ 2]]</p><p>Interoperability works by having different systems share data by employing the same standards for data formats (e.g., Extensible Markup Language or JavaScript Object Notation), exchange, and vocabulary to title data elements. These standards are agreed upon “in advance” by organizations, industry types, and user groups. Then, software developers build the system based on the model agreed upon.</p>

#Concrete example from a digital technology or platform<p>Interoperable systems are vital in healthcare, commerce, and public safety. A notable example includes the integration of Electronic Health Records (HER). This system allows different healthcare providers to securely access and share personal patient information regardless of the specific EHR system used.[[https://data.europa.eu/en/news-events/news/interoperability-solutions-bridging-systems-connected-future 3]] An example of such a system is the GNU Health Hospital Information System, which is a model employed by small private offices and large national public health systems.[[https://interoperable-europe.ec.europa.eu/collection/gnu-health/solution/gnu-health-hospital-information-system 4]]</p><p>For example, a health care professional may access lab results and patient history data from various platforms to enable them to make a diagnosis.</p>

#One governance implication: what does this concept explain about how platforms exercise power?<p>Interoperability is a powerful tool because it helps provide a “seamless user experience” and optimizes data sharing between separate systems. It helps companies and organizations access data from multiple sources and improve efficiency, such as accelerating workflows and reducing the possibility of errors by pooling data from its sources.</p><p>Interoperability standards are mandated by law on the technical side to ensure “critical systems” are compatible, especially in public safety and health. [[https://www.ibm.com/think/topics/interoperability 5]]</p><p>In Europe, interoperability is governed by the European Health Data Space regulation (EHDS); the “first common EU data space dedicated to a specific sector as part of the European strategy for data.”</p><p>This Regulation aims to benefit EU citizens such as patients, researchers, healthcare professionals, and industry players by establishing a common framework for the exchange of electronic health data within the EU. In addition, it enables certain data to be “reused” for public purpose and innovative scientific research. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 6]]</p>

#Where does the concept break down or fail to explain something<p>While the EDHS harmonizes a legal and technical framework for sharing EU data, its shortcoming is its gradual implementation. The EDHS entered into force in 2025; however, most of its obligations will not apply until 2029. This leaves four years for health care companies to use interoperability in a deregulated manner. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 7]]</p><p>Furthermore, there are actual barriers to achieving full interoperability in healthcare, including data silos (which isolate data collections and hinder data exchange), lack of standardization (where one system records blood pressure as free text, another system stores it as a structured numeric value), and deficient infrastructure, which requires significant technological upgrading.</p><p>The burden remains on health companies to decide which standards to adopt and how much to invest in stable technological infrastructure to strengthen their data governance and interoperability. [[https://www.tigahealth.com/how-to-overcome-healthcare-interoperability-challenges/ 8]]</p>

= "Economies of Scale (Jihae)" =
= [[Network Effects (Indirect) - Sebastián]] =
= [[Monopsony - (Atmika)]] =

Governance by markets

2026-10-04T12:07:17Z

Sergey Abramov: /* Data as an input to market power (Sergey) */

= Glossary =
= Data as an input to market power (Sergey) =
<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals [[https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1]]. '''Data''' is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. '''Market power''' refers to a firm's ability to influence the price, output, or other market conditions in its favour, without being constrained by competitive forces [[https://www.tutor2u.net/economics/reference/what-is-meant-by-market-power 2]].</p>
<p>'''Data as an input to market power''' means that access to information can help a business limit the competitive pressure it faces. Information about searches, clicks or purchases can improve services and advertising. If competitors cannot obtain comparable information at a reasonable cost, they may struggle to offer an attractive alternative. This can allow a platform to raise prices or worsen quality or privacy protections without losing enough business to make doing so unprofitable. This matters even when a service is free, because competition also concerns advertising charges, personal information and the choices available. Data can be used repeatedly and by several businesses without being used up. However, access can be restricted through contracts and technical controls. Nevertheless, holding a large dataset alone does not establish market power.</p>
<p>'''Google Search''' illustrates this mechanism. Google controls over 91% of global search engine traffic [[https://www.statista.com/statistics/1381664/worldwide-all-devices-market-share-of-search-engines/ 3]] and uses information about users’ searches and the results they click to improve the relevance of its rankings. A larger volume of searches gives it more opportunities to learn, including about unusual queries that smaller search engines encounter less often. Better results can attract further use, producing additional information and reinforcing the advantage. A new search engine may therefore struggle to attract users while lacking the information needed to improve its service. Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. </p>
<p>A platform’s control over data access can shape who is able to compete with it. By determining whether, and on what terms, rivals can access that information, a platform can influence the competitive opportunities available to others. This suggests that regulation should examine access to strategically important data alongside prices and market shares. Article 6(11) of the EU Digital Markets Act [[https://eur-lex.europa.eu/eli/reg/2022/1925/oj/eng#:~:text=giving%20their%20consent.-,11.%C2%A0,-The%20gatekeeper%20shall 4]] reflects this approach: certain large search providers must, on request, share specified search data with rival search providers on fair and non-discriminatory terms, with relevant personal data anonymised. The aim is to help rivals compete while protecting users. A data advantage does not, by itself, establish unlawful conduct.</p>
<p>Would access to equivalent data enable a rival to compete effectively? The concept cannot answer this on its own. Additional data may eventually add little value, become outdated, or have substitutes. Firms also need the skills and computing resources to use it. Google’s default positions in browsers and devices can further restrict rivals’ access to users even if data become available. The unresolved issue is how much competitive advantage comes from data itself and how much comes from these other resources and arrangements. </p>

= Two Sided Markets (Ceren) =
= Switching costs and lock-in (Kate) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul)  =
=== Excludability and Rivalrousness ===
<p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p>
<p>'''Excludability''': Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p>
<p>'''Rivalrousness''': Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p>
<p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>

=== AI Training and the (non-)excludability of information. ===
<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. Conversely, data may be rivalrous not in its absolute depletion by consumption, but rather the benefit or competitive advantage usage of that data may confer: If all AI firms are able to train on the exact same data (all available data), the utility of that data may still increase, but the competitive advantage of possessing that data is exhausted.</p>

=== Free-Rider Problems & Public Data ===
<p>As any party can benefit from publicly available data at no cost or low cost, the incentive for private market actors to fund its creation is comparatively weak. Consequently, public bodies (and less frequently open-source foundations and non-profits) often assume the responsibility for the provision of this type of data. A classical example is weather data and its creation by national weather agencies.</p>

= Interoperability (Sisi) =

#What is interoperability?[[https://www.ibm.com/think/topics/interoperability 1]]<p>Interoperability is a “standards-based” approach that enables different IT systems to communicate with minimal end-user interaction to exchange data and “share functionality”. Each system also allows the other system to “read, update, modify, and analyze” data with minimal human interaction. [[https://www.oracle.com/nl/interoperability/ 2]]</p>
<p>Interoperability works by having different systems share data by employing the same standards for data formats (e.g., Extensible Markup Language or JavaScript Object Notation), exchange, and vocabulary to title data elements. These standards are agreed upon “in advance” by organizations, industry types, and user groups. Then, software developers build the system based on the model agreed upon.</p>
#Concrete example from a digital technology or platform<p>Interoperable systems are vital in healthcare, commerce, and public safety. A notable example includes the integration of Electronic Health Records (HER). This system allows different healthcare providers to securely access and share personal patient information regardless of the specific EHR system used.[[https://data.europa.eu/en/news-events/news/interoperability-solutions-bridging-systems-connected-future 3]] An example of such a system is the GNU Health Hospital Information System, which is a model employed by small private offices and large national public health systems.[[https://interoperable-europe.ec.europa.eu/collection/gnu-health/solution/gnu-health-hospital-information-system 4]]</p>
<p>For example, a health care professional may access lab results and patient history data from various platforms to enable them to make a diagnosis.</p>
#One governance implication: what does this concept explain about how platforms exercise power?<p>Interoperability is a powerful tool because it helps provide a “seamless user experience” and optimizes data sharing between separate systems. It helps companies and organizations access data from multiple sources and improve efficiency, such as accelerating workflows and reducing the possibility of errors by pooling data from its sources.</p>
<p>Interoperability standards are mandated by law on the technical side to ensure “critical systems” are compatible, especially in public safety and health. [[https://www.ibm.com/think/topics/interoperability 5]]</p>
<p>In Europe, interoperability is governed by the European Health Data Space regulation (EHDS); the “first common EU data space dedicated to a specific sector as part of the European strategy for data.”</p>
<p>This Regulation aims to benefit EU citizens such as patients, researchers, healthcare professionals, and industry players by establishing a common framework for the exchange of electronic health data within the EU. In addition, it enables certain data to be “reused” for public purpose and innovative scientific research. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 6]]</p>
#Where does the concept break down or fail to explain something<p>While the EDHS harmonizes a legal and technical framework for sharing EU data, its shortcoming is its gradual implementation. The EDHS entered into force in 2025; however, most of its obligations will not apply until 2029. This leaves four years for health care companies to use interoperability in a deregulated manner. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 7]]</p>
<p>Furthermore, there are actual barriers to achieving full interoperability in healthcare, including data silos (which isolate data collections and hinder data exchange), lack of standardization (where one system records blood pressure as free text, another system stores it as a structured numeric value), and deficient infrastructure, which requires significant technological upgrading.</p>
<p>The burden remains on health companies to decide which standards to adopt and how much to invest in stable technological infrastructure to strengthen their data governance and interoperability. [[https://www.tigahealth.com/how-to-overcome-healthcare-interoperability-challenges/ 8]]</p>

= "Economies of Scale (Jihae)" =
= [[Network Effects (Indirect) - Sebastián]] =

Governance by markets

2026-10-04T12:01:52Z

Sergey Abramov: /* Data as an input to market power (Sergey) */

= Glossary =
= Data as an input to market power (Sergey) =
<p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals [[https://repository.nls.ac.in/cgi/viewcontent.cgi?article=1478&context=nlsblr 1]]. Data is an input into production, so businesses that control data can have market power and impede the entry of potential competitors. Market power refers to a firm's ability to influence the price, output, or other market conditions in its favour, without being constrained by competitive forces.</p><p>'''Data as an input to market power''' means that access to information can help a business limit the competitive pressure it faces. Information about searches, clicks or purchases can improve services and advertising. If competitors cannot obtain comparable information at a reasonable cost, they may struggle to offer an attractive alternative. This can allow a platform to raise prices or worsen quality or privacy protections without losing enough business to make doing so unprofitable. This matters even when a service is free, because competition also concerns advertising charges, personal information and the choices available. Data can be used repeatedly and by several businesses without being used up. However, access can be restricted through contracts and technical controls. Nevertheless, holding a large dataset alone does not establish market power.</p><p>'''Google Search''' illustrates this mechanism. Google controls over 91% of global search engine traffic [[https://www.statista.com/statistics/1381664/worldwide-all-devices-market-share-of-search-engines/ 2]] and uses information about users’ searches and the results they click to improve the relevance of its rankings. A larger volume of searches gives it more opportunities to learn, including about unusual queries that smaller search engines encounter less often. Better results can attract further use, producing additional information and reinforcing the advantage. A new search engine may therefore struggle to attract users while lacking the information needed to improve its service. Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. </p><p>A platform’s control over data access can shape who is able to compete with it. By determining whether, and on what terms, rivals can access that information, a platform can influence the competitive opportunities available to others. This suggests that regulation should examine access to strategically important data alongside prices and market shares. Article 6(11) of the EU Digital Markets Act [[https://eur-lex.europa.eu/eli/reg/2022/1925/oj/eng#:~:text=giving%20their%20consent.-,11.%C2%A0,-The%20gatekeeper%20shall 3]] reflects this approach: certain large search providers must, on request, share specified search data with rival search providers on fair and non-discriminatory terms, with relevant personal data anonymised. The aim is to help rivals compete while protecting users. A data advantage does not, by itself, establish unlawful conduct.</p><p>Would access to equivalent data enable a rival to compete effectively? The concept cannot answer this on its own. Additional data may eventually add little value, become outdated, or have substitutes. Firms also need the skills and computing resources to use it. Google’s default positions in browsers and devices can further restrict rivals’ access to users even if data become available. The unresolved issue is how much competitive advantage comes from data itself and how much comes from these other resources and arrangements. </p>

= Two Sided Markets (Ceren) =
= Switching costs and lock-in (Kate) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul)  =
=== Excludability and Rivalrousness ===
<p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p>
<p>'''Excludability''': Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p>
<p>'''Rivalrousness''': Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p>
<p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>

=== AI Training and the (non-)excludability of information. ===
<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. Conversely, data may be rivalrous not in its absolute depletion by consumption, but rather the benefit or competitive advantage usage of that data may confer: If all AI firms are able to train on the exact same data (all available data), the utility of that data may still increase, but the competitive advantage of possessing that data is exhausted.</p>

=== Free-Rider Problems & Public Data ===
<p>As any party can benefit from publicly available data at no cost or low cost, the incentive for private market actors to fund its creation is comparatively weak. Consequently, public bodies (and less frequently open-source foundations and non-profits) often assume the responsibility for the provision of this type of data. A classical example is weather data and its creation by national weather agencies.</p>

= Interoperability (Sisi) =

#What is interoperability?[[https://www.ibm.com/think/topics/interoperability 1]]<p>Interoperability is a “standards-based” approach that enables different IT systems to communicate with minimal end-user interaction to exchange data and “share functionality”. Each system also allows the other system to “read, update, modify, and analyze” data with minimal human interaction. [[https://www.oracle.com/nl/interoperability/ 2]]</p>
<p>Interoperability works by having different systems share data by employing the same standards for data formats (e.g., Extensible Markup Language or JavaScript Object Notation), exchange, and vocabulary to title data elements. These standards are agreed upon “in advance” by organizations, industry types, and user groups. Then, software developers build the system based on the model agreed upon.</p>
#Concrete example from a digital technology or platform<p>Interoperable systems are vital in healthcare, commerce, and public safety. A notable example includes the integration of Electronic Health Records (HER). This system allows different healthcare providers to securely access and share personal patient information regardless of the specific EHR system used.[[https://data.europa.eu/en/news-events/news/interoperability-solutions-bridging-systems-connected-future 3]] An example of such a system is the GNU Health Hospital Information System, which is a model employed by small private offices and large national public health systems.[[https://interoperable-europe.ec.europa.eu/collection/gnu-health/solution/gnu-health-hospital-information-system 4]]</p>
<p>For example, a health care professional may access lab results and patient history data from various platforms to enable them to make a diagnosis.</p>
#One governance implication: what does this concept explain about how platforms exercise power?<p>Interoperability is a powerful tool because it helps provide a “seamless user experience” and optimizes data sharing between separate systems. It helps companies and organizations access data from multiple sources and improve efficiency, such as accelerating workflows and reducing the possibility of errors by pooling data from its sources.</p>
<p>Interoperability standards are mandated by law on the technical side to ensure “critical systems” are compatible, especially in public safety and health. [[https://www.ibm.com/think/topics/interoperability 5]]</p>
<p>In Europe, interoperability is governed by the European Health Data Space regulation (EHDS); the “first common EU data space dedicated to a specific sector as part of the European strategy for data.”</p>
<p>This Regulation aims to benefit EU citizens such as patients, researchers, healthcare professionals, and industry players by establishing a common framework for the exchange of electronic health data within the EU. In addition, it enables certain data to be “reused” for public purpose and innovative scientific research. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 6]]</p>
#Where does the concept break down or fail to explain something<p>While the EDHS harmonizes a legal and technical framework for sharing EU data, its shortcoming is its gradual implementation. The EDHS entered into force in 2025; however, most of its obligations will not apply until 2029. This leaves four years for health care companies to use interoperability in a deregulated manner. [[https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en 7]]</p>
<p>Furthermore, there are actual barriers to achieving full interoperability in healthcare, including data silos (which isolate data collections and hinder data exchange), lack of standardization (where one system records blood pressure as free text, another system stores it as a structured numeric value), and deficient infrastructure, which requires significant technological upgrading.</p>
<p>The burden remains on health companies to decide which standards to adopt and how much to invest in stable technological infrastructure to strengthen their data governance and interoperability. [[https://www.tigahealth.com/how-to-overcome-healthcare-interoperability-challenges/ 8]]</p>

= "Economies of Scale (Jihae)" =
= [[Network Effects (Indirect) - Sebastián]] =

Governance by markets

2026-10-02T12:48:43Z

Sergey Abramov: /* Data as an input to market power (Sergey) */

= Glossary =
= Data as an input to market power (Sergey) =
Draft <p>In the present digital economy, data has emerged as a valuable economic commodity that entities leverage to gain a competitive edge over their rivals. Today, all enterprises in the digital economy have data-driven business models; that is, these entities thrive by collecting, processing, and analysing data on a large scale. Due to such a data-centric business model, it is crucial to consider the commercial use of data when assessing competitive concerns in digital markets. Economists point to economies of scale in information and the dominance of large, data-intensive firms as evidence that the unequal accumulation of data is responsible for a decline in competition. As a result, the data-rich firm exerts more market power.  </p><p>Data can be considered as a metric to assess the market power of a digital platform. Market power is a firm’s ability to act independently of its competitors. Data is an input into production, so firms that control data can have market power and impede the entry of would-be competitors. </p><p>Data affect the competitive dynamics of markets in four main ways. First, they serve as an important input for products and services. Second, they may constitute a dimension of quality from the perspective of consumers (both in terms of the amount of information available to consumers, and the amount of personal data collected from consumers). Third, data may themselves constitute a product that leads to the creation of new markets. Finally, they may be used to facilitate consumer decision making.</p><p>Data access can constitute a significant barrier to entry when this type of analytics is necessary to participate in a market, particularly when alternatives are limited. For example, Google offers free search engine services to acquire data and uses this data for targeted online advertisements to generate revenue.</p>

....

= Two Sided Markets (Ceren) =
= Switching costs and lock-in (Kate) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul) =
<p>Excludability and Rivalrousness</p>

<br><p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p>

<br><p>Excludability: Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p>
<p>Rivalrousness: Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p>

<br><p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>
<p>AI Training and the (non-)excludability of information.</p>
<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. </p>

Governance by markets

2026-10-02T11:56:37Z

Sergey Abramov:

= Glossary =
= Data as an input to market power (Sergey) =
= Two Sided Markets (Ceren) =
[[File:Untitled Diagram.drawio (1).png|300px|thumb|Goods Matrix]]

= Excludability and rivalrousness (Paul) =
<p>Excludability and Rivalrousness</p>

<br><p>The terms excludability and rivalrousness both refer to consumption characteristics that can be assigned to- or identified in goods. </p>

<br><p>Excludability: Goods to which economic actors can technically or legally restrict access are considered to have ‘high excludability’. Conversely, goods to which access cannot be restricted are of ‘low excludability’, or, in absolute terms, ‘non-excludable’.</p>
<p>Rivalrousness: Rivalrousness on the other hand refers to the finitude of the good in question. The greater the depletion of the stock, quantity, or amount is on account of one party’s consumption, the more rivalrous a good is. </p>

<br><p>Tabulating the characteristics against one another allows for inference and definition of the type of good an object, service, or resource may constitute (see fig. 1). Excludability and Rivalrousness of goods can therethrough serve as a predictor for certain types of market failures: Goods which are non-excludable or of low excludability, but nonetheless costly to provide, are vulnerable to so-called free rider problem. Goods which are non-excludable, but highly rivalrous (common goods), may lead to the tragedy of the commons, that is over-exploitation.<br></p>
<p>AI Training and the (non-)excludability of information.</p>
<p>The unusual nature of excludability and rivalrousness of data further compromises the applicability of the goods classification built thereon. In traditional economics, a non-excludable and non-rivalrous good would be seen as a public good. This does apply insofar as data is vulnerable to corporate exploitation without commensurate indemnification. Data usage may however actually be anti-rival in the sense that increasing aggregate usage increases the value of the data (better labeling, easier access, previously undrawn conclusions, etc.), rather than diminish the ‘stock’ of data. </p>

Infrastructures: Digital Sovereignty

2026-10-01T13:18:07Z

Sergey Abramov: /* 4.1 Brasil Semicon (Law No. 14.968/2024) */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2142'''''
<div style="font-weight: 400;">
== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">
== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">
== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">
=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">
=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===

<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial ([https://www.gov.br/mcti/pt-br/acompanhe-o-mcti/transformacaodigital/ai-for-the-good-of-all-2025.pdf PBIA]), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;"></div>

== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">
=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===

<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under [https://www2.camara.leg.br/legin/fed/lei/2024/lei-14968-11-setembro-2024-796221-publicacaooriginal-172984-pl.html Law No. 14.968] of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div></div></div>

=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===

<div><p>Published on 16 September 2026, [https://legis.senado.leg.br/norma/43923839/publicacao/43929692 Law No. 15.506/2026] establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div></div>

== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
i) Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
ii) Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>iii)
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
iv) Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
v) Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
vi) Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>
== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

= United States =
=== The Concept of Digital Sovereignty ===
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
=== The Four Registers ===
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p>
<p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p>
<p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p>
<p>'''Procurement''' is quieter and more effective, and illustrated below by the cloud and semiconductor sectors.</p>
<p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>

=== Direction of Travel ===
<p>Policy directionality is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p>
<p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.<p dir="auto">Ironically, the US maintains mechanisms to protect its data from foreign control while monitoring others’, consolidating digital sovereignty by setting the terms of engagement.</p>

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There's a clear divergence in objective these reflect, and it is rooted in the different conception tof human rights, values and norms, related to data protection and privacy both actors have. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 the infrastructural capacity the EU lacks ]. This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act], a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25% tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.
=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals,  and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-10-01T13:12:55Z

Sergey Abramov: /* 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2142'''''
<div style="font-weight: 400;">
== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">
== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">
== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">
=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">
=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===

<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial ([https://www.gov.br/mcti/pt-br/acompanhe-o-mcti/transformacaodigital/ai-for-the-good-of-all-2025.pdf PBIA]), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;"></div>

== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">
=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>
=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===

<div><p>Published on 16 September 2026, [https://legis.senado.leg.br/norma/43923839/publicacao/43929692 Law No. 15.506/2026] establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div></div>

== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
i) Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
ii) Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>iii)
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
iv) Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
v) Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
vi) Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>
== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

= United States =
=== The Concept of Digital Sovereignty ===
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
=== The Four Registers ===
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p>
<p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p>
<p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p>
<p>'''Procurement''' is quieter and more effective, and illustrated below by the cloud and semiconductor sectors.</p>
<p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>

=== Direction of Travel ===
<p>Policy directionality is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p>
<p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.<p dir="auto">Ironically, the US maintains mechanisms to protect its data from foreign control while monitoring others’, consolidating digital sovereignty by setting the terms of engagement.</p>

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There's a clear divergence in objective these reflect, and it is rooted in the different conception tof human rights, values and norms, related to data protection and privacy both actors have. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 the infrastructural capacity the EU lacks ]. This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act], a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25% tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.
=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals,  and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-10-01T13:07:41Z

Sergey Abramov: /* 3.2 Brazilian AI Plan (PBIA, 2024–2028) */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2142'''''
<div style="font-weight: 400;">
== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">
== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">
== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">
=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">
=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===

<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial ([https://www.gov.br/mcti/pt-br/acompanhe-o-mcti/transformacaodigital/ai-for-the-good-of-all-2025.pdf PBIA]), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;"></div>

== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">
=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>
=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===
</div>
<div><p>Published on 16 September 2026, Law No. 15.506/2026 establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div>
== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
i) Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
ii) Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>iii)
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
iv) Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
v) Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
vi) Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>
== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

= United States =
=== The Concept of Digital Sovereignty ===
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
=== The Four Registers ===
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p>
<p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p>
<p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p>
<p>'''Procurement''' is quieter and more effective, and illustrated below by the cloud and semiconductor sectors.</p>
<p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>

=== Direction of Travel ===
<p>Policy directionality is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p>
<p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.<p dir="auto">Ironically, the US maintains mechanisms to protect its data from foreign control while monitoring others’, consolidating digital sovereignty by setting the terms of engagement.</p>

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There's a clear divergence in objective these reflect, and it is rooted in the different conception tof human rights, values and norms, related to data protection and privacy both actors have. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 the infrastructural capacity the EU lacks ]. This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act], a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25% tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.
=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals,  and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-09-25T14:59:12Z

Sergey Abramov: /* Brazil */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2142'''''
<div style="font-weight: 400;">
== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">
== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">
== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">
=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">
=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===
</div>
<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial (PBIA), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;">
== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">
=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>
=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===
</div>
<div><p>Published on 16 September 2026, Law No. 15.506/2026 establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div>
== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
i) Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
ii) Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>iii)
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
iv) Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
v) Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
vi) Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>
== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

= United States =
== The Concept of Digital Sovereignty ==
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
== The Four Registers ==
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p><p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p><p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p><p>'''Procurement''' is quieter and probably more effective, and is illustrated below by the cloud and semiconductor sectors.</p><p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>
=== Direction of Travel ===
<p>The direction of policy is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p><p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.

The United States nonetheless maintains mechanisms to protect itself from equivalent foreign control over its own data, an arrangement not without irony. Through these safeguards, the United States continues to consolidate a form of dominance that might properly be termed digital sovereignty, in that it is the party setting the terms of engagement, monitoring others' data while shielding its own from comparable scrutiny.

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There is a clear divergence in objective these reflect, and it is rooted in the different conception the EU has of human rights, values and norms, related to data protection and privacy in comparison to the US'. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses the infrastructural capacity the EU lacks ([https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 only 13 to 15 percent of the market is EU-owned]). This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard, stalled, per the [https://www.iss.europa.eu/publications/briefs/technical-political-when-cloud-certification-scheme-divides-europe EUISS], by Member States divided over their reliance on US providers; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act] of August 2022, a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25 percent tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.

Each of these examples have the same objective, that federal money spent on the domestic industry should build capacity, not enrich shareholders or, worse, end up financing a rival through some indirect channel. Subsidy, clawback, denial, tariff, ownership and now buyback limits, each is a different tool, but each answers to the same objective, an America made big and sovereign enough to win the race for AI dominance.

=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">The United States' pursuit of AI dominance rests on a chain of physical dependencies: AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals, particularly the neodymium, dysprosium, and gadolinium used in the magnets and components of chip fabrication and data-centre hardware, and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-09-25T14:47:09Z

Sergey Abramov: /* Brazil */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2117'''''
<div style="font-weight: 400;">

== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">

== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">

== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">

=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">

=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===
</div>
<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial (PBIA), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;">

== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">

=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>

=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===
</div>
<div><p>Published on 16 September 2026, Law No. 15.506/2026 establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div>

== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
i) Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
ii) Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>iii)
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
iv) Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
v) Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
vi) Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>

== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

= United States =
== The Concept of Digital Sovereignty ==
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
== The Four Registers ==
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p><p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p><p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p><p>'''Procurement''' is quieter and probably more effective, and is illustrated below by the cloud and semiconductor sectors.</p><p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>
=== Direction of Travel ===
<p>The direction of policy is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p><p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.

The United States nonetheless maintains mechanisms to protect itself from equivalent foreign control over its own data, an arrangement not without irony. Through these safeguards, the United States continues to consolidate a form of dominance that might properly be termed digital sovereignty, in that it is the party setting the terms of engagement, monitoring others' data while shielding its own from comparable scrutiny.

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There is a clear divergence in objective these reflect, and it is rooted in the different conception the EU has of human rights, values and norms, related to data protection and privacy in comparison to the US'. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses the infrastructural capacity the EU lacks ([https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 only 13 to 15 percent of the market is EU-owned]). This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard, stalled, per the [https://www.iss.europa.eu/publications/briefs/technical-political-when-cloud-certification-scheme-divides-europe EUISS], by Member States divided over their reliance on US providers; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act] of August 2022, a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25 percent tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.

Each of these examples have the same objective, that federal money spent on the domestic industry should build capacity, not enrich shareholders or, worse, end up financing a rival through some indirect channel. Subsidy, clawback, denial, tariff, ownership and now buyback limits, each is a different tool, but each answers to the same objective, an America made big and sovereign enough to win the race for AI dominance.

=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">The United States' pursuit of AI dominance rests on a chain of physical dependencies: AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals, particularly the neodymium, dysprosium, and gadolinium used in the magnets and components of chip fabrication and data-centre hardware, and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-09-25T14:45:29Z

Sergey Abramov: /* 5. The EU–Brazil Digital Partnership (2026) */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2110'''''
<div style="font-weight: 400;">
== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">
== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">
== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">
=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">
=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===
</div>
<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial (PBIA), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;">
== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">
=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>
=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===
</div>
<div><p>Published on 16 September 2026, Law No. 15.506/2026 establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div>
== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
i) Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
ii) Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>iii)
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
iv) Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
v) Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
vi) Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>
== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

= United States =
== The Concept of Digital Sovereignty ==
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
== The Four Registers ==
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p><p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p><p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p><p>'''Procurement''' is quieter and probably more effective, and is illustrated below by the cloud and semiconductor sectors.</p><p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>
=== Direction of Travel ===
<p>The direction of policy is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p><p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.

The United States nonetheless maintains mechanisms to protect itself from equivalent foreign control over its own data, an arrangement not without irony. Through these safeguards, the United States continues to consolidate a form of dominance that might properly be termed digital sovereignty, in that it is the party setting the terms of engagement, monitoring others' data while shielding its own from comparable scrutiny.

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There is a clear divergence in objective these reflect, and it is rooted in the different conception the EU has of human rights, values and norms, related to data protection and privacy in comparison to the US'. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses the infrastructural capacity the EU lacks ([https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 only 13 to 15 percent of the market is EU-owned]). This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard, stalled, per the [https://www.iss.europa.eu/publications/briefs/technical-political-when-cloud-certification-scheme-divides-europe EUISS], by Member States divided over their reliance on US providers; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act] of August 2022, a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25 percent tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.

Each of these examples have the same objective, that federal money spent on the domestic industry should build capacity, not enrich shareholders or, worse, end up financing a rival through some indirect channel. Subsidy, clawback, denial, tariff, ownership and now buyback limits, each is a different tool, but each answers to the same objective, an America made big and sovereign enough to win the race for AI dominance.

=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">The United States' pursuit of AI dominance rests on a chain of physical dependencies: AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals, particularly the neodymium, dysprosium, and gadolinium used in the magnets and components of chip fabrication and data-centre hardware, and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-09-25T14:40:30Z

Sergey Abramov: /* Brazil */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
'''''Words: 2110'''''
<div style="font-weight: 400;">
== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">
== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">
== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">
=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">
=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===
</div>
<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial (PBIA), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;">
== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">
=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>
=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===
</div>
<div><p>Published on 16 September 2026, Law No. 15.506/2026 establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div>
== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>
== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

#Historical context:
<p>In light of decolonization and liberalization of the economy around the 1980s, India experienced a new wave of innovation and global economic relevance owing to their educated workforce (1). India has remained an enthusiastic participant in the global market and international political space, following the same trend as other economic powers that once promoted open trade relations, switching towards national interests - settling on a theme of self-reliance.</p>
<p><br></p>
#Digital Sovereignty Ambition:
<p>“Digital sovereignty” is a concept that has been used by scholars and policymakers as an umbrella term to make claims over all aspects of digital infrastructures (2). In order to analyze the instruments used in India to pursue digital sovereignty, we must explore those instruments under the light of different conceptualizations deeply linked to this term. Therefore, we will resort to the following notions: 1. Technological sovereignty; 2. Internet sovereignty; and 3. Data sovereignty.</p>
<p><br></p>
#Technological Sovereignty:
<p>Technological sovereignty “refers to the efforts of nations and/or countries to create technologies that reduce American control over, and surveillance of technological systems, including the Internet” (3). Therefore, in this section we outline the main attempts from the Indian Government directed towards reducing their technological dependencies, especially when it comes to critical infrastructures.</p>
<p>One of the early instruments to gaining a broader control over India’s infrastructure was the Information Technology Act, 2000, (the “IT Act”) as amended by The Information Technology (Amendment) Act 2008. Section 70 of the IT Act (https://indiankanoon.org/doc/1680277/) invests the Government with the authority to define via an official notification certain computer resources as protected systems. The provision takes a further step in elaborating what counts as “Critical Information Infrastructure” (CII), defining it as the computer resource, the incapacitation or destruction of which shall have a debilitating impact on national security, economy, public health or safety.</p>
<p>In the protection of CII, Section 70A (https://indiankanoon.org/doc/106027784/) decrees the establishment of a dedicated national nodal agency, which shall be responsible for several measures including R&D relating to protection of CII. However, despite this regulatory effort, little was achieved during this time to address the topic of technological dependencies.</p>
<p>As a response, the Ministry of Electronics and Information Technology (MeitY) issued the country’s first National Cyber Security Policy (NCSP) (https://www.meity.gov.in/static/uploads/2024/02/National_cyber_security_policy-2013_0.pdf) in 2013. This Policy can be categorized as a soft law instrument aimed at steering the country towards building a secure and resilient cyberspace for citizens, businesses and Government. Among its many objectives, the Policy aimed at developing “suitable indigenous security technologies through frontier technology research, leading to widespread deployment of secure ICT products / processes in general and specifically for addressing National Security requirements”. The NCSP proposed many strategies to enhance security in the cyberspace, including:</p>
*- Strengthening the Regulatory framework.
*- Creating a National Level Computer Emergency Response Team (CERT-In).
*- Operating a 24x7 National Critical Information Infrastructure Protection Centre to function as the nodal agency for critical information infrastructure protection.
*- Reducing supply chain risks by creating and maintaining testing infrastructure and facilities for IT security product evaluation and compliance verification as per global standards and practices.
<p>Many of these strategies were followed through, albeit with different degrees of success when it comes to achieving sovereignty in the technological sphere.</p>
<p>To begin with, the NCSP led to the creation of the National Critical Information Infrastructure Protection Centre (“NCIIPC”). As the commanding authority accountable, the NCIIPC introduced the Information Security Practices and Procedures for Protected System Rules, 2018 (https://www.meity.gov.in/static/uploads/2024/02/NCIIPC-Rules-notification-1.pdf) (the “NCIIPC Rules”) and issued Guidelines for the Protection of National Critical Information Infrastructure, 2015 (https://www.asianlaws.org/gcld/cyberlawdb/IN/guidelines/NCIIPC_Guidelines_V2.pdf) (the “NCIIPC Guidelines”) to safeguard India’s CII from unauthorized access, modification, use, disclosure and disruption to ensure a safe, secure and resilient information infrastructure for critical sectors in the country. Among these rules, certain requirements stand out, such as the mandatory inventories of hardware and software related to CII (Section 3(3-e) of NCIIPC Rules) and the promotion of indigenous Research and Development (R&D) regarding the protection of CII (Section 4.3.7 of NCIIPC Guidelines).</p>
<p>In addition, the India Stack (the “Stack”) was a nationwide government program that recognized societal inequalities enhanced and perpetuated due to lack of access to the internet and technology. As a socialist endeavor, the Government launched platforms to dissipate welfare functions, promoting digital literacy and enhancing electronic governance. These systems were conceived as being mostly indigenous and open-source, allowing India to now export this technology framework to other nations (https://indianmasterminds.com/news/india-signs-mous-23-countries-digital-public-infrastructure-india-stack-upi-global-183393/) looking to build their own digital stack.</p>
<p>Furthermore, under the Public Procurement (Preference to Make in India) Order, 2017 (https://www.dpiit.gov.in/static/uploads/2025/07/a2243b925eacf5e116e0e3f43630a8b2.pdf), MeitY issued a notification designating cyber security as a strategic sector, stating that "preference shall be provided by all procuring entities to domestically manufactured/produced Cyber Security Products" with preference granted to companies incorporated in India whose product IP and licensing revenue accrue domestically, and royalty payments abroad capped at 50% of the product's cost.</p>
<p>Despite the above, technological dependency is still an issue in India. The policy on data centers is a clear example of this, since these facilities have proliferated in India during recent years under a sovereignty discourse. Under India's Consolidated FDI Policy Circular of 2020 (https://www.dpiit.gov.in/static/uploads/2025/07/3ab2ec2a3bdb91c69653b7c34618c14a.pdf ) (DPIIT), data centers have not required prior government or central bank approval. Implying that the nation intends to favor foreign investment in their construction. These incentives mean that the infrastructure for data centers “will be built upon American foundation models and run on American chips” (4), a policy that ultimately benefits hyperscalers without offering much in return for affected populations.</p>
<p>This dilemma showcases one of the main tensions still prevalent in India: developing indigenous infrastructure for critical services in order to prevent dependency on third countries (as shown with the Stack) vs boosting investment by encouraging foreign companies to build digital infrastructures within Indian sovereign territories.</p>
<p><br><br></p>
#Internet Sovereignty:
<p>Internet or network sovereignty is a notion invoked by protectionists or nationalist agendas in an attempt to “control the digital networks and information running through them”(5).</p>
<p>The NCSP provided an adequate advisory and guideline basis which helped protect infrastructures and public services upon which the Digital India Scheme (DIS) was promoted. DIS sought to expand access to the Internet of Things from city to rural areas to facilitate government functions and financial transactions digitally through secured high speed broadband and digital public infrastructures (DPI), a foundational block of the Stack (6).</p>
<p>The India Stack is a three-stage e-government infrastructure project based on the digitalized Aadhaar platform, comprised of the Identity Layer (developed since 2009), the Payment Layer (developed since 2010), and the final Data Layer (building on the previous two layers), designed to combat money laundering and corruption, better monitor financial flows, and provide the Indian population with simplified access to financial services (7).</p>
<p>This is a pivotal policy to analyze for a better grasp on the internet sovereignty sought by the Indian Government. The main achievements of this initiative are the following:</p>
#Aadhar (Identity Layer): Served as a nationwide security and biometric unique identity scheme that created the world’s largest digital public infrastructure.
#Unified Payments Interface (Payment Layer): Enabled simple digital transactions between two mobiles to transition into a cashless economy with additional financial schemes such as demonetization and zero-cost bank accounts.
#DigiLocker(Data Layer): A secure cloud storage space run by the government to secure personal data documents for official purposes without physical paper dependency.
<p>While being praised due to its efficiency and inclusionary potential, the stack’s increasing centrality in governance raises significant concerns. Significant concerns have been flagged as it relates to mass surveillance (https://www.giga-hamburg.de/en/publications/giga-focus/digital-surveillance-and-the-threat-to-civil-liberties-in-india) that consequently arises from the state's control over the network.</p>
<p>In addition, Section 69A of the IT Act (https://www.giga-hamburg.de/en/publications/giga-focus/digital-surveillance-and-the-threat-to-civil-liberties-in-india) bestows the Government powers over content, by ratifying the capacity to block on narrow grounds (sovereignty, security, public order, etc). Since 2023, though, the government has increasingly used a parallel channel, consisting of notice-and-takedown mechanisms through the "Sahyog" portal (operated by the Cyber Crime Coordination Centre of India) under Section 79(3)(b) (https://indiankanoon.org/doc/161852741/). This provision was intended for removal of content used for unlawful acts, and affected entities have claimed that the Government has attempted to extend its reach to override the restrictions built into Section 69A. This shift generated over 2,300 orders across 19 platforms (https://www.barandbench.com/amp/story/columns/section-69a-of-the-information-technology-act-2000-and-the-expanding-architecture-of-digital-censorship-in-india) in just one year (Oct 2024–Oct 2025) and is now being challenged in court (https://indiankanoon.org/doc/7614885/).</p>
<p>The dilemma presented above showcases a new tension within the Indian digital framework: while certain provisions allow the Central Government to maintain an effective control over the network and the contents shared through it, a significant risk for citizens privacy rights and individual freedoms is arising as a result.</p>
<p><br></p>
#Data sovereignty
<p>Finally, data sovereignty refers to “protecting and controlling information on and flowing through digital networks” (8). The Supreme Court's judgment in ''Justice K.S. Puttaswamy v. Union of India (https://indiankanoon.org/doc/127517806/) ''addressed the privacy concerns of the centralization of citizens’ data in Aadhaar, recognizing that the right to privacy would fall under Article 21 of the Constitution (originally reserved to the right to live). There has been much controversy around the societal ramification of a single identification mechanism including the exclusion of nonparticipants from fundamental public services.</p>
<p>The ''Puttaswamy'' decision formed the material foundation (9) of what would later be the Digital Personal Data Protection Act (https://www.meity.gov.in/static/uploads/2024/06/2bf1f0e9f04e6fb4f8fef35e82c42aa5.pdf), issued in 2023. It delineated two groups (the “Data Fiduciaries”):</p>
*- Processors of data in the Indian territory or
*- Processors of data outside of India, in the condition such processing is in connection with any activity related to offering of goods or services to Data Principals within Indian territory.
<p>This is one of the main elements that is worth mentioning about this framework, as the extraterritorial reach of the law attempts to extend India’s sovereignty over its data beyond the borders of the country.</p>
<p>Furthermore, Section 16(1) established that The Central Government may, by notification, restrict the transfer of personal data by a Data Fiduciary for processing to such country or territory outside India as may be so notified.</p>
<p>To complement the Act, additional Rules were issued in 2025 (https://www.meity.gov.in/static/uploads/2025/11/53450e6e5dc0bfa85ebd78686cadad39.pdf) following the objections and suggestions received from the public by the Central Government. According to Section 13 (4), a Data Fiduciary shall undertake measures to ensure that personal data specified by the Central Government, on the basis of the recommendations of a committee constituted by it, is processed subject to the restriction that the personal data and the traffic data pertaining to its flow is not transferred outside the territory of India.</p>
<p>The abovementioned provisions are a clear indicator of India’s attempt to strengthen the control over their citizen’s data. While international transfers are less restricted than under the EU approach, foreign Data Fiduciaries would still be subject to Indian law, and the Central Government is in position of both banning transfers to specific countries as well as establishing data localization requirements for certain data elements. This structure, while not identical to the GDPR, still manages to reinforce the country’s data sovereignty claims.</p>
<p><br></p>
*Comparison with the EU
<p>Although India and the EU both acknowledge technological sovereignty as a key objective, the corresponding frameworks, prioritization and approaches in the direction of achieving this starkly vary.</p>
<p>For instance, as it pertains to the domain of data governance and data protection, the EU has a comprehensive regulatory landscape that gives high-priority to the safeguarding of rights as evidenced with General Data Protection Regulation (“GDPR”) and the Digital Services Act, which institute well-defined rules on data privacy, and cross-border data flows. A distinguishing element of approaches relates to paradigms in place. The EU has adopted the ''‘allowed list’'' model that is rooted in adequacy decisions, while India has a ‘''negative list''’ model, meaning it permits cross-border data transfers unless specifically restricted. Both paradigms present different challenges and opportunities respectively. The European approach encourages third countries towards benchmarking for adequacy decisions, while India encourages strong relations to avoid a ban. This could explain the fact that India lacks an adequacy decision from the EU.</p>
<p>However, these two nations share similarities regarding their digital infrastructure policy goals. In a shared statement on the second meeting of the EU-India Trade and Technology Council (https://www.eeas.europa.eu/delegations/india/joint-statement-second-meeting-eu-india-trade-and-technology-council_en), both countries agreed to “collaborate on working towards interoperability of their respective DPIs that respect human rights and protect personal data, privacy, and intellectual property rights” and “promote the strengthening of the EU and Indian semiconductor ecosystems to enhance technological capabilities and ensure supply chain resilience by developing sustainable, secure and diversified semiconductor production capacities”. Both countries also attempt to extend the effect of their infrastructure regulations beyond its borders, with India’s IT Act applying to any offence or contravention committed outside India if the act or conduct involves a computer, computer system or computer network located in India (Section 75 https://indiankanoon.org/doc/576992/) and the EU requiring non-EU digital infrastructure and service providers offering services in the Union to appoint an EU representative (NIS2 Art. 26(3)).</p>
<p>Their main differences in that regard stem from the robustness of their legislation. While the EU has developed a strong sectorial cyber security framework in the form of the NIS 2 Directive (with the corresponding transpositions across Member States), India’s framework relies on an outdated law that generally applies to IT services (the IT Act) without the same degree of harmonized regulatory density. Furthermore, from a policy point of view, India has attempted to foster foreign innovation in developing digital infrastructures for data centers, favoring the presence of hyperscalers, whereas the EU has attempted to steer into the opposite direction and “support the development of European cloud computing offerings (https://oeil.europarl.europa.eu/oeil/en/document-summary?id=1656816)”. This showcases a major difference in addressing an already identified issue regarding their respective technological dependencies.</p>
<p><br></p>
<p>References:</p>
*(1) Biswajit Maitra and Moutushi Chakraborty, 'Foreign Trade, Human Capital and Economic Growth in India under the Liberalised Trade Regime' (2023) 5(1) Journal of Asian Economic Integration 29, doi:10.1177/26316846221147581.
*(2) Plixavra Vogiatzoglou and Joris van Hoboken, 'The EU's Digital Sovereignty and Quantum Technologies: To What End?' (2026) 18(2) Law, Innovation and Technology, doi:10.1080/17579961.2026.2705118, 2.
*(3) Stéphane Couture and Sophie Toupin, ‘What Does the Concept of “Sovereignty” Mean in Digital, Network and Technological Sovereignty?’ (Social Science Research Network, 22 January 2018) 5 https://doi.org/10.2139/ssrn.3107272
*(4) Vinit Ravishankar, 'Sovereignty After the Boom' (''Disjunctions'') https://disjunctionsmag.com/articles/sovereignty-after-boom/ accessed 24 September 2026.
*(5) Vogiatzoglou and van Hoboken (n 2) 6.
*(6) Snehal A. Bagul, 'Guarding the Digital Gateway: An In-Depth Analysis of Cybersecurity Challenges in India' in Faisal Rehman and others (eds), ''Emerging Trends in Information System Security Using AI & Data Science for Next-Generation Cyber Analytics'' (Springer 2025) 69–71.
*(7) Sezgin Sönmez, 'The India Stack – Between Networked Growth and Territorial Control' (2025) 50(4) Historical Social Research 318, https://www.jstor.org/stable/27426713.
*(8) Vogiatzoglou and van Hoboken (n 2) 6.
*(9) Committee of Experts under the Chairmanship of Justice B N Srikrishna, ''A Free and Fair Digital Economy: Protecting Privacy, Empowering Indians'' (Ministry of Electronics and Information Technology, 27 July 2018) https://www.thehinducentre.com/resources/article24561547.ece/binary/Data_Protection_Committee_Report-comp

= United States =
== The Concept of Digital Sovereignty ==
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
== The Four Registers ==
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p><p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p><p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p><p>'''Procurement''' is quieter and probably more effective, and is illustrated below by the cloud and semiconductor sectors.</p><p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>
=== Direction of Travel ===
<p>The direction of policy is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p><p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.

The United States nonetheless maintains mechanisms to protect itself from equivalent foreign control over its own data, an arrangement not without irony. Through these safeguards, the United States continues to consolidate a form of dominance that might properly be termed digital sovereignty, in that it is the party setting the terms of engagement, monitoring others' data while shielding its own from comparable scrutiny.

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There is a clear divergence in objective these reflect, and it is rooted in the different conception the EU has of human rights, values and norms, related to data protection and privacy in comparison to the US'. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses the infrastructural capacity the EU lacks ([https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 only 13 to 15 percent of the market is EU-owned]). This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard, stalled, per the [https://www.iss.europa.eu/publications/briefs/technical-political-when-cloud-certification-scheme-divides-europe EUISS], by Member States divided over their reliance on US providers; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act] of August 2022, a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25 percent tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.

Each of these examples have the same objective, that federal money spent on the domestic industry should build capacity, not enrich shareholders or, worse, end up financing a rival through some indirect channel. Subsidy, clawback, denial, tariff, ownership and now buyback limits, each is a different tool, but each answers to the same objective, an America made big and sovereign enough to win the race for AI dominance.

=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">The United States' pursuit of AI dominance rests on a chain of physical dependencies: AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals, particularly the neodymium, dysprosium, and gadolinium used in the magnets and components of chip fabrication and data-centre hardware, and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Infrastructures: Digital Sovereignty

2026-09-25T14:38:12Z

Sergey Abramov: /* 5. The EU–Brazil Digital Partnership (2026) */

= Infrastructures: Digital Sovereignty =
Digital Sovereignty in the European Union is a [https://www.tandfonline.com/doi/full/10.1080/17579961.2026.2705118 term used to convey ] a broad array of policy ambitions in relation to digital technology development and use: the urgency to regain control over data, infrastructures, and (critical) technologies, the establishment of strategic autonomy, the conclusion of partnerships with geopolitical allies, and the reduction of critical dependencies. According to the Digital Decade Policy Programme [https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:52023SC0570 Implementation Report] of 2023, digital sovereignty refers to the EU's "ability to act independently in the digital world and therefore it constitutes a crucial means to safeguard [EU] values".

= Brazil =
<div style="font-weight: 400;">

== 1. Introduction: What Is Digital Sovereignty? ==
</div>
<div style="font-weight: 400;"><p>Digital sovereignty refers to the capacity of a state to exercise autonomy, self-determination, and control over data, digital infrastructure, and technologies. This encompasses three distinct but overlapping dimensions: internet sovereignty (control over networks and online content), data sovereignty (governance of data flows, privacy, and cybersecurity), and technological sovereignty (control over critical technologies such as artificial intelligence, semiconductors, and raw materials). The endeavor toward digital sovereignty was accelerated by the Snowden revelations. Brazil has developed its own discourse on digital sovereignty, shaped by its historical concerns about foreign surveillance, its position as a major emerging economy, and its ambition to become a global technology actor. </p></div>
<div style="font-weight: 400;">

== 2. Brazil's National Digital Sovereignty Outlook ==
</div>
<div style="font-weight: 400;"><p>Brazil's approach to digital sovereignty is rooted in a long-standing concern about technological dependence and foreign surveillance. The 2013 revelation that the National Security Agency had intercepted communications of Brazilian President Dilma Rousseff and surveilled Petrobras (Brazil's state oil company) served as a breakpoint moment. This urged the Brazilian political agenda to reduce reliance on foreign digital infrastructure, and prompted Brazil to accelerate its own digital infrastructure projects. The most notable infrastructure project is the EllaLink undersea cables, connecting Brazil to Portugal (and thereby South America to Europe), which was explicitly designed to circumvent internet routing controlled by the US and enhance Brazil's data [https://www.cambridge.org/core/books/digital-sovereignty-in-the-brics-countries/introduction/88A1164945DC955023BB812CA88FEABC sovereignty. ]</p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty discourse is also closely tied to its open-source tradition. Brazil has historically been a strong proponent of open-source software. In the early 2000’s, during the Lula administration, Brazil’s government promoted free and open-source software as a tool for reducing dependence on proprietary foreign technology and lowering [https://www.researchgate.net/publication/316669377_Free_Software_as_Public_Service_in_Brazil_An_Assessment_of_Activism_Policy_and_Technology costs. ]This open-source orientation remains an element of Brazil's digital sovereignty strategy today: the development of national AI models, open data ecosystems, and publicly governed cloud infrastructure all reflect this tradition.  </p></div>
<div style="font-weight: 400;"><p>The Brazilian AI Plan (PBIA) explicitly includes the development of a national AI software stack and open training databases, reflecting the view that open-source tools are a vehicle for technological autonomy rather than a security risk. </p></div>
<div style="font-weight: 400;"><p>Brazil's national digital sovereignty strategy is therefore characterised by two core priorities: (i) reducing dependence on foreign digital infrastructure and proprietary technology and (ii) building domestic capacity in critical technology sectors (AI, semiconductors, critical minerals). </p></div>
<div style="font-weight: 400;">

== 3. Regulatory Framework: Artificial Intelligence ==
</div>
<div style="font-weight: 400;">

=== 3.1 Brazilian AI Strategy (EBIA) ===
</div>
<div style="font-weight: 400;"><p>Brazil's first official national AI strategy, the Estratégia Brasileira de Inteligência Artificial (EBIA), established the country's overarching vision for AI governance. Based on the OECD's AI principles, EBIA is structured around pillars including ethical and responsible AI, data governance, innovation, education and workforce development, public sector use of AI, and international cooperation. It explicitly prioritises human-centred AI, the protection of fundamental rights, transparency, and risk mitigation. However, EBIA is a soft law instrument: it sets principles only, with no binding obligations, no enforcement mechanisms, and no dedicated budget. It functions as a normative framework that articulates values and directions rather than imposing legal duties. </p></div>
<div style="font-weight: 400;">

=== 3.2 Brazilian AI Plan (PBIA, 2024–2028) ===
</div>
<div style="font-weight: 400;"><p>The Plano Brasileiro de Inteligência Artificial (PBIA), launched on 30 July 2024 under the motto "AI for the Good of All," builds on EBIA by translating strategic vision into a concrete investment and infrastructure roadmap. Approved by the National Council for Science and Technology and coordinated by the Ministry of Science, Technology and Innovation, the PBIA is a strategic non-binding soft-law policy framework with a planned investment of R$23.03 billion through 2028. </p></div>
<div style="font-weight: 400;"><p>The PBIA combines immediate-impact initiatives with structural actions organised into five axes. These actions aim to develop the capabilities and skills needed for sustainable AI development in Brazil, while strengthening national sovereignty, technological autonomy, competitiveness and responsible innovation. Approximately 98.1% of planned investment is directed towards these structural actions. AI for Business Innovation (59.9%), Infrastructure and AI Development (25.2%), and AI for Improving Public Services (7.6%) receive the largest shares, followed by Dissemination, Education and Training in AI (5.0%). Support for AI Regulation and Governance receives approximately 0.4%. This distribution indicates an emphasis on building technological and industrial capacity. It is also indicative that Brazil prioritizes developing the technology infrastructure over implementing regulation in the form of hard law. From a digital sovereignty perspective, several elements of the PBIA are particularly significant. The Infrastructure and AI Development axis includes a National AI Infrastructure Program aimed at investing in high-performance computing and autonomous AI hardware to support technological sovereignty. It also includes a Sustainability and Renewable Energy Programme for AI, supporting renewable-energy infrastructure and efficient resource use, and a Programme for Structuring the Data and Software Ecosystem for AI intended to reduce external dependence, improve training datasets and develop a national AI software stack. AI for improving public services axis provides for a Sovereign Cloud, private or community cloud infrastructure managed exclusively by public agencies or state-owned companies, to increase control, security and sovereignty over sensitive data and ensure it is stored on national territory. The AI for Business Innovation axis aims to support sectors within the AI production chain, including information and communication technologies (ICT) and microelectronics, and to position Brazil as a global data centre hub. The PBIA reports that Brazil currently operates only 168 data centres (ranking 13th globally, compared to over 5,000 in the United States). Its AI Value Chain Promotion Programme seeks to strengthen and expand the domestic AI value chain, from supporting startups to developing Brazilian companies supplying foundation models. </p></div>
<div style="font-weight: 400;"><p>Bill No PL 2338/2023, which proposes a comprehensive AI framework, has passed the Senate and remains under consideration in the Chamber of Deputies, the lower house of Brazil’s National Congress. Until it is enacted, Brazil's AI governance framework remains largely in the realm of soft law, creating legal uncertainty for investors, developers, and citizens alike. </p></div>
<div style="font-weight: 400;">

== 4. Regulatory Framework: Semiconductors and Critical Minerals ==
</div>
<div style="font-weight: 400;">

=== 4.1 Brasil Semicon (Law No. 14.968/2024) ===
</div>
<div style="font-weight: 400;"><p>The Brazil Semiconductor Programme (Brasil Semicon), established under Law No. 14.968 of 11 September 2024, constitutes Brazil's principal hard law instrument for technological sovereignty in the semiconductor sector. It establishes a national strategy and guidelines to encourage research, development, innovation, production, and application of semiconductors, displays, and solar panels. Brasil Semicon updates and expands the instruments of PADIS (the Support Program for Technological Development of the Semiconductor Industry), extending tax incentives until the end of 2029 and broadening their scope to include software, specialised services, technical assistance, and other stages of the production chain. Decree No. 13.065 of 16 July 2026 activated the programme, establishing the rules for the semiconductor industry and facilitating access to PADIS tax incentives. The programme aims to attract investments in the national chip supply chain until 2029, directly addressing Brazil's dependence on foreign semiconductor manufacturing. </p></div>
<div style="font-weight: 400;"><div>

=== 4.2 National Policy on Critical and Strategic Minerals (Law No. 15.506/2026) ===
</div>
<div><p>Published on 16 September 2026, Law No. 15.506/2026 establishes a national policy and governance structure for critical and strategic minerals. It aims to promote research, extraction, processing, and transformation of critical and strategic minerals domestically, adding value to supply chains rather than exporting raw materials. The policy explicitly seeks to strengthen Brazil's economic and geopolitical security by ensuring a stable domestic supply of minerals essential to the digital economy — including those required for semiconductors, batteries, and AI hardware. It provides up to R$5 billion in tax credits (covering up to 20% of eligible expenses) from 2030 through 2034 for mineral processing, manufacturing, and recycling projects, and authorises a Mineral Activity Guarantee Fund with up to R$2 billion in federal contributions. This law is significant because Brazil holds some of the world's largest reserves of rare earth minerals and other critical inputs for the digital economy, and the policy represents a strategic decision to leverage these resources for domestic technological development rather than simply exporting them. </p></div>
<div>

== 5. The EU–Brazil Digital Partnership (2026) ==

<div><p>Signed on 12 June 2026, the EU–Brazil Digital Partnership is a soft law instrument establishing a framework for bilateral cooperation on issues that concern digital sovereignty.  </p></div>
<div>
Article 2 commits both parties to fostering meaningful digital development and inclusion while preserving digital sovereignty, democratic principles, and human rights.  </div>
<div>
Article 3 provides for reciprocal recognition of data protection levels — based on the European Parliament's adequacy decision of January 2026 and Brazil's ANPD Resolution 32 — thereby easing cross-border data flows and building trust in bilateral data governance.  </div>
<div>
Article 15 commits both parties to developing national technology supply chains to ensure that global interdependence does not undermine their respective technological sovereignties.  </div>
<div>
Article 17 provides for cooperation on AI development and deployment, referencing the PBIA and comparable EU programmes.  </div>
<div>
Articles 19 and 21 establish information-sharing mechanisms on data governance, data spaces, digital services, and digital markets, with an explicit commitment to preserving regulatory autonomy.  </div>
<div>
Article 23 references Brazil's EllaLink undersea cable project as a concrete expression of its digital sovereignty ambitions. </div>
<div><p>The Digital Partnership is a precursor to harder legal instruments: it is future-facing, flexible, and oriented toward sharing know-how and building mutual trust. Its soft law nature is appropriate given the current rate of innovation in technology. </p></div>
<div></div>

== 6. Comparison with the EU ==
</div>
<div><p>Brazil and the EU share the same fundamental goal: reducing dependence on foreign (primarily US and Chinese) digital infrastructure and technology. However, they pursue it through different regulatory architectures. </p></div>
<div><p>The EU has adopted a comprehensive, multi-layered hard law framework for digital sovereignty, encompassing the GDPR, the NIS 2 Directive, the Cybersecurity Act, the Chips Act, and the Critical Raw Materials Act, alongside the newly proposed EU Tech Sovereignty Package (June 2026) including the Cloud and AI Development Act (CADA) and an EU Open Source Strategy. The package therefore combines legislative proposals with non-binding policy instruments. The EU's approach combines binding regulation with substantial public investment, and it applies to both public and private sector actors. </p></div>
<div><p>Brazil, by contrast, has followed a more strategy-led approach to developing its dedicated AI framework. Its AI governance framework remains largely in soft law (EBIA and PBIA), with binding legislation (PL 2338) still pending. Brazil’s hard law instruments are concentrated in the industrial and resource sectors (Brasil Semicon; the Critical Minerals Law), reflecting a Brazil’s priority being placed on supply chain sovereignty and infrastructure building (rather than regulatory compliance frameworks).  </p></div>
<div><p>In 2026 Brazil began developing the Nuvem Brasileira, a national cloud infrastructure designed to keep critical government data and control of the infrastructure under Brazilian jurisdiction, while favouring open standards and reducing dependence on foreign technology providers. Specifically, the Brazilian government decided that Microsoft and other foreign big tech providers would not be the preferred providers for the Nuvem Brasileira. The government is explicitly assessing dependencies on foreign suppliers and whether the State could continue operating critical services if an external provider became unavailable. Brazil’s model does not require everything to be Brazilian-made, but focuses on control over data, operations and technology.  This shifts sovereignty from simply where data is stored to who controls the infrastructure, keys, software, jurisdiction and ability to operate independently. It is alongside REDATA, which uses datacentre incentives to build domestic technological capacity. Brazil's open-source orientation also distinguishes it from the EU. Both jurisdictions promote open source as a means of strengthening technological autonomy. While the EU has adopted an Open Source Strategy and the CADA references open-source principles, Brazil's commitment to open-source software is deeper and more historically embedded, reflecting a political tradition that views proprietary technology as a vector of foreign dependence. The PBIA's emphasis on a national AI software stack and open training data reflects this tradition and represents a distinctive approach to technological sovereignty that merits attention in comparative analysis. </p></div>
<div><p>Finally, both Brazil and the EU face the same structural challenge: the global digital economy is dominated by a small number of US and Chinese technology companies, and neither jurisdiction has yet succeeded in producing domestic alternatives at comparable scale. Brazil's ambition to become a global data centre hub and to develop sovereign AI models is bold, but the gap between aspiration and capacity remains large. The EU's more developed AI-specific regulatory architecture provides stronger protections for citizens and clearer rules for industry, but has been criticised for potentially hampering innovation. Brazil's softer approach may allow for faster initial development, but risks leaving citizens and the market without adequate legal protection in the interim. </p></div>
<div>

== 7. Conclusion ==
</div>
<div><p>Brazil's approach to digital sovereignty is characterised by a strong national discourse rooted in concerns about foreign surveillance and technological dependence, a soft law-dominant AI governance framework that is still evolving toward binding regulation, and hard law instruments in the semiconductor and critical minerals sectors that reflect a strategic prioritisation of supply-chain sovereignty. Brazil's open-source tradition adds a distinctive dimension to its digital sovereignty strategy, positioning it differently from both the. The EU–Brazil Digital Partnership represents a meaningful step toward bilateral cooperation, but its soft law nature means  its impact will depend on follow-through.  </p></div></div>

= China =
<p><strong style="font-size: 14.4px;">Words: 2306</strong><br></p>
<div><p><br></p></div>
<div><p>'''Introduction'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;">“Digital Sovereignty” is an increasingly important concept for understanding how States exercise political and economic control over digital infrastructures and technologies. Since the introduction of the Internet in 1994, Chinese academic and political discourse has formulated notions of cyber, information, and network sovereignty in direct relation to digital and technological independence.</p></div>
<div style="text-align: justify;"><p>In the early 2000s, Chinese authorities focused primarily on controlling data flows rather than developing national technological champions. Indeed, at the time Google held around the 40% of the Chinese Internet search market and companies like Yahoo maintained substantial equity participation in Chinese firms like Alibaba, all of which ended in 2009 when the Urumqi riots triggered China’s decision to permanently block platforms like Facebook and YouTube, prompting Google’s withdrawal from the Chinese Internet search market.</p></div>
<div style="text-align: justify;"><p>In 2010 “Cyber Sovereignty” was crystallized as an organizing principle of Chinese digital policy following historical concerns about information control and a national determination to resist foreign influence over China. This political and economic landscape cemented the nation’s ambition to achieve full digital sovereignty.</p></div>
<div style="text-align: justify;"><p>Today Chinese digital sovereignty is incorporated into its legal framework, particularly in the Cybersecurity Law and the Data Security Law (both from 2017). These laws aim for China to develop the autonomy and capacity necessary to exercise full control over its information infrastructure and data.</p></div>
<div><p style="text-align: justify;"><br></p></div>
<div><p>'''Cybersecurity Law'''</p></div>
<div><p><br></p></div>
<div><p>The Cybersecurity Law is a comprehensive legal framework that regulates cyberspace and information infrastructure security in China, governing information networks build, operation, and use within its territory (see article 2). The law covers a wide range of issues, including critical information infrastructure management, cybersecurity, and data protection.</p></div>
<div><p style="text-align: justify;" >In the area of critical information infrastructure management, the law applies to “Critical Information Infrastructures” (CIIs) and “Critical Information Infrastructure Operators” (CIIOs) which include information infrastructure and its operators in sectors such as energy, finance, healthcare, telecommunications, transportation, and governmental services. CIIOs have the obligation to adopt enhanced security measures and protocols to safeguard information infrastructure and their data. Additionally, CIIOs must perform regular security assessments and report their results to State agencies, implement protective measures and protocols, and cooperate with State agencies, among others (see articles 23, 36, 37, 38, and 40).</p></div>
<div><p style="text-align: justify;" >In the area of cybersecurity, the law applies to all information infrastructure operators (not only to CIIOs) which have the obligation to adopt comprehensive cybersecurity measures which include monitoring their infrastructure security, safeguarding against data breaches, impeding misuse of their infrastructures for illegal activities, prepare incident response plans, report cybersecurity incidents, and cooperate with State agencies (see articles 23, 27, 30 and 51).</p></div>
<div><p style="text-align: justify;" >Finally, in the area of data protection, the law applies to CIIOs in the case of data localization, and to all information infrastructure operators in the case of personal information. CIIOs have the obligation to store within China all critical data and personal information they collect or generate in the country’s territory. Additionally, CIIOs must pass a security evaluation and get State authorization to transfer data abroad (see article 39). Moreover, all information infrastructure operators have the obligation to follow guidelines for collecting, processing, and storing personal information. In that sense, information infrastructure operators must secure user consent for gathering personal information and disclose it, and warrant its protection by implementing measures to guard against unauthorized or illegal access and misuse including breach notification (see article 42).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to balance cybersecurity and information infrastructure development (see article 4), continuously improve the national cybersecurity strategy (see article 5), and deal with cybersecurity risks and threats (see article 6).</p></div>
<div><p style="text-align: justify;" >In general, the law provides the State extensive powers to access data held by information infrastructure operators functioning within China, including the authority to require assistance and support in matters of national security or criminal investigations.</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions for information infrastructure operators which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a threat to public order or national security, infringements can result in criminal prosecution and career suspension (see articles 61, 63, 64, 65, 66, 68, 69, 71, and 76).</p></div>
<div><p><br></p></div>
<div><p>'''Data Security Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Data Security Law is a comprehensive legal framework that regulates data management for the purposes of national sovereignty, security, and social and economic development.</p></div>
<div><p style="text-align: justify;" >It establishes the obligation of the State to develop a data classification system based on its importance to social and economic development and the damage its alteration, destruction, leakage, or illicit use would cause to national security, public interest, or people or organizations’ rights and interests. Such classification allows the State to apply varying protection levels depending on the data’s sensitivity and relevance, where “central data of the State” or data deemed important to national security, major public interests, critical economic sectors, or important aspects of people livelihood is given the highest level of protection (see article 21). Other categories of data are “important data” and “general data” which are not defined in the law but in complementary regulations.</p></div>
<div><p style="text-align: justify;" >The law applies to data management carried out within China’s territory, and generates legal responsibility for data management outside if it harms national security, public interest, or people or organizations’ rights and interests (see article 2). Data processors of “important data” have the obligation to designate officers responsible for data security, to conduct risk evaluations periodically reporting the results to State authorities, and to report security incidents to users and State authorities, among others (see articles 27, 29 and 30). Moreover, data processors are impeded to transfer data to foreign judicial authorities without authorization (see article 36).</p></div>
<div><p style="text-align: justify;" >Additionally, the law establishes obligations for the State including to encourage the use of data for economic and innovation purposes, specifically for the development of new data-driven industries as long as they comply with the law (see article 1). Also, it has the obligation to coordinate the national data security, and draft and implement the national data security strategy (see article 5).</p></div>
<div><p style="text-align: justify;" >In general the law enables the State to collect data if it is necessary to safeguard national security or investigate crimes, subject to approval procedures (see article 35). Data processors and individuals are obliged to cooperate, and refusal is subject to fines (see article 48).</p></div>
<div><p style="text-align: justify;" >Infringing the law can result in sanctions which include fines, operational suspensions, and business license revocations (individuals may face personal responsibility). In more extreme cases, where there is a crime or a threat to national sovereignty, security, or economic development interests, infringements can result in criminal prosecution (see articles 45 and 52).</p></div>
<div><p><br></p></div>
<div><p>'''Personal Information Protection Law'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Personal Information Protection Law (of 2021) is a comprehensive legal framework that protects personal information similarly to the European General Data Protection Regulation (GDPR). Although the law is not directly related to China’s digital sovereignty, it establishes obligations relevant to it: State agencies, critical infrastructure operators, and data processors that collect, process, or use data above a certain threshold (defined by State authorities) must store within China all personal information they collect or generate in the country’s territory. Additionally, they must pass a security evaluation and get State authorization to transfer data abroad (see articles 36 and 40).</p></div>
<div><p style="text-align: justify;" >Moreover, the State is allowed to blacklist individuals and organizations and deny them access to personal information generated or collected in China if it deems that they acted against Chinese people rights or interests, as well as to adopt reciprocal measures against foreign States if it deems that they discriminated against China regarding the protection of personal information (see articles 42 and 43).</p></div>
<div><p><br></p></div>
<div><p>'''China’s State Centered Concept of Digital Sovereignty'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The Cybersecurity Law and the Data Security Law are integral parts of China’s strategy to strengthen its digital sovereignty. China’s concept of digital sovereignty is characterized for its strong State-centered approach in which the State holds ultimate and supreme authority in the digital space, controlling data flows and content sharing within its borders. In a sense, China’s concept of digital sovereignty is defensive: it is essential to protect China’s social, political, and economic architecture from destabilization attempts by foreign actors, which explains why the Cybersecurity Law establishes the obligation of private actors to ''“abide by the Constitution and laws, observe public order, respect social morality, and (...) not (...) use the Internet to engage in activities that endanger national security, honor, or interests (...), incite subversion of State power or overthrow of the socialist system (...), incite secession or undermine national unity (...), (or) fabricate or disseminate false information to disrupt economic and social order”'' (see article 13), and why the the Data Security Law establishes as an objective to ''“preserve State sovereignty, security, and development interests”'' (see article 1).</p></div>
<div><p style="text-align: justify;" >Such a concept requires the achievement of two policy objectives: territorialization and indigenization. Through territorialization China seeks to replicate its national borders in cyberspace, ensuring that online processes impacting important national interests take place within Chinese boundaries, and unwanted practices or operations can be barred from entering. Through indigenization, in turn, it strives to replace foreign actors and technologies with local equivalents, reducing foreign dependence.</p></div>
<div><p style="text-align: justify;" >In this context, the Cybersecurity Law and the Data Security Law aim to protect China’s digital sovereignty through both policy objectives. In the case of territorialization, they establish the objective of protecting sovereignty in cyberspace and preventing and responding to cybersecurity risks and threats coming from inside and outside China; the obligation of CIIOs to store in China all personal information and important data that they generate or collect inside the country’s borders, and to conduct security evaluations and obtain State permission before transferring them abroad; the obligation of data processors to not transfer data to foreign judicial authorities without State authorization; the responsibility of data processors for activities carried out outside China; and the capacity of the State to block prohibited information, to temporarily restrict communications in specific areas for national security reasons, and to adopt countermeasures against foreign States that discriminate against China in data investment or trade.</p></div>
<div><p style="text-align: justify;" >In the case of indigenization, the laws establish mechanisms of promotion and access control that indirectly privilege national providers. As an example, they establish the obligation of critical infrastructure operators to purchase information network products and services approved in a national security assessment; the obligation of critical information network products providers to offer products that are certified or comply with national standards; the obligation of the State to support Chinese cybersecurity industries and projects, to support Chinese Artificial Intelligence (AI) industries, to support data development and security industrial systems, to support the creation of national data markets, and to develop a national strategy for big data and data infrastructure.</p></div>
<div><p style="text-align: justify;" >These efforts are related to the “Great Firewall”, the Chinese State system for censoring and controlling Internet traffic within its borders which is, arguably, the most visible example of territorialization, drawing an effective Internet frontier between China and the rest of the world.</p></div>
<div><p style="text-align: justify;" >In this way, China's digital sovereignty concept demands asserting control over information infrastructures and data gathered or generated within its borders as well as to protect relevant or sensitive information from foreign interference and cyber-attacks while simultaneously reinforcing its surveillance capacity and its ability to manage information flows for “social stability”.</p></div>
<div><p><br></p></div>
<div><p>'''China Model Versus the European Union Model'''</p></div>
<div><p><br></p></div>
<div><p style="text-align: justify;" >The European Union (EU) and China represent two different models of digital sovereignty. The EU approach contrasts with China’s State-centered and security-driven model because it emerged from a supranational, rights-centered, and democratic context.</p></div>
<div><p style="text-align: justify;" >EU’s efforts to achieve digital sovereignty began in 2013 when Edward Snowden’s revelations of United States National Security Agency (NSA) interception of European communications exposed its dependence on foreign technology and triggered a wave of policy and regulatory efforts. Such efforts led to the adoption of the GDPR in 2016 which enshrined individuals' rights, including the rights of access to and objection to processing of personal data. The EU continued by developing two consumer protection regulations in 2019, in addition to conducting several antitrust proceedings against American hyperscalers such as Alphabet, Amazon, and Apple between 2015 and 2020. In 2024, the EU adopted the AI Act with the purpose of harmonizing AI development and deployment with EU values and fundamental rights. More recently, the EU initiated infrastructure initiatives such as Gaia X for cloud sovereignty and the Chips Act for semiconductor independence.</p></div>
<div><p style="text-align: justify;" >Despite this extensive regulatory framework, China's claim to digital sovereignty may be considered more consolidated than the EU's because it fuses legal codification with direct State control over technology, information, and infrastructure in a way the EU market-correcting model does not yet.</p></div>
<div><p>As European Commission President von der Leyen stated in 2019, ''"it is not too late to achieve technological sovereignty in some critical technology areas"''. The EU has been trying to catching up with China, which spent the last decade tightening its State authority over its digital infrastructure before the EU began treating its own technology dependence as a vulnerability. China's more recent legislative activity has extended this consolidation into the international stage as well as domestically. In 2024, the Shanghai Declaration on Global AI Governance, issued by China's Ministry of Foreign Affairs, committed signatories to developing AI under principles of safety, reliability, and controllability, while asserting each country's right to regulate AI according to its own national conditions. This language simultaneously projects openness to international cooperation while defending from non-foreign interference, the logic underpinning China's digital sovereignty doctrine.</p></div>
<div><p style="text-align: justify;" >This declaration is not merely aspirational: it is anchored in binding domestic law, specifically Article 20 of China's Security Law, which directs the State to support AI research, infrastructure, ethical norms, and risk supervision.</p></div>
<div><p style="text-align: justify;" >Therefore, where the EU’s sovereignty is exercised primarily through regulating outside actors within its internal market, China's sovereignty is exercised through direct State-led development, information control, and an explicit fusion of domestic statutes and international declaratory positioning. This makes China's model arguably more comprehensive compared to the EU.</p></div>
<div><p>''''''</p>
<p>'''References'''</p></div>
<div><p><br></p></div>
<div><p>'''(1)'''<span class="apple-converted-space"> </span>Cybersecurity Law of the People's Republic of China (2025 Amendment).</p></div>
<div><p>'''(2)'''<span class="apple-converted-space"> </span>Data Security Law of the People's Republic of China (2021).</p></div>
<div><p>'''(3)'''<span class="apple-converted-space"> </span>Personal Information Protection Law of the People's Republic of China (2021).</p></div>
<div><p>'''(4)'''<span class="apple-converted-space"> </span>Ministry of Foreign Affairs of the People's Republic of China, "Shanghai Declaration on Global AI Governance" (4 July 2024).</p></div>
<div><p>'''(5)'''<span class="apple-converted-space"> </span>Jiang, M. Models of State Digital Sovereignty From the Global South: Diverging Experiences From China, India and South Africa, Policy & Internet, 15 December 2024, </p>
[https://doi.org/10.1002/poi3.427' https://doi.org/10.1002/poi3.427']</div>
<div><p>'''(6)'''<span class="apple-converted-space"> </span>Rivaton, R., “The Origins and Fate of Digital Sovereignty”, Fondation pour L’Innovation Politique, 21 may 2026, </p>
[https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty https://www.fondapol.org/decryptage/the-origins-and-fate-of-digital-sovereignty]</div>
<div>'''(7) '''Thumfart, J. & Cong. W., “A Chinese Precursor to the Digital Sovereignty Debate: Digital Anti-Colonialism and Authoritarianism from the Post–Cold War Era to the Tunis Agenda”, Global Studies Quarterly (2022) 2, 1–13.</div>
<div></div>

= [[Infrastructures:India|India]] =

#Historical context:
<p>In light of decolonization and liberalization of the economy around the 1980s, India experienced a new wave of innovation and global economic relevance owing to their educated workforce (1). India has remained an enthusiastic participant in the global market and international political space, following the same trend as other economic powers that once promoted open trade relations, switching towards national interests - settling on a theme of self-reliance.</p>
<p><br></p>
#Digital Sovereignty Ambition:
<p>“Digital sovereignty” is a concept that has been used by scholars and policymakers as an umbrella term to make claims over all aspects of digital infrastructures (2). In order to analyze the instruments used in India to pursue digital sovereignty, we must explore those instruments under the light of different conceptualizations deeply linked to this term. Therefore, we will resort to the following notions: 1. Technological sovereignty; 2. Internet sovereignty; and 3. Data sovereignty.</p>
<p><br></p>
#Technological Sovereignty:
<p>Technological sovereignty “refers to the efforts of nations and/or countries to create technologies that reduce American control over, and surveillance of technological systems, including the Internet” (3). Therefore, in this section we outline the main attempts from the Indian Government directed towards reducing their technological dependencies, especially when it comes to critical infrastructures.</p>
<p>One of the early instruments to gaining a broader control over India’s infrastructure was the Information Technology Act, 2000, (the “IT Act”) as amended by The Information Technology (Amendment) Act 2008. Section 70 of the IT Act (https://indiankanoon.org/doc/1680277/) invests the Government with the authority to define via an official notification certain computer resources as protected systems. The provision takes a further step in elaborating what counts as “Critical Information Infrastructure” (CII), defining it as the computer resource, the incapacitation or destruction of which shall have a debilitating impact on national security, economy, public health or safety.</p>
<p>In the protection of CII, Section 70A (https://indiankanoon.org/doc/106027784/) decrees the establishment of a dedicated national nodal agency, which shall be responsible for several measures including R&D relating to protection of CII. However, despite this regulatory effort, little was achieved during this time to address the topic of technological dependencies.</p>
<p>As a response, the Ministry of Electronics and Information Technology (MeitY) issued the country’s first National Cyber Security Policy (NCSP) (https://www.meity.gov.in/static/uploads/2024/02/National_cyber_security_policy-2013_0.pdf) in 2013. This Policy can be categorized as a soft law instrument aimed at steering the country towards building a secure and resilient cyberspace for citizens, businesses and Government. Among its many objectives, the Policy aimed at developing “suitable indigenous security technologies through frontier technology research, leading to widespread deployment of secure ICT products / processes in general and specifically for addressing National Security requirements”. The NCSP proposed many strategies to enhance security in the cyberspace, including:</p>
*- Strengthening the Regulatory framework.
*- Creating a National Level Computer Emergency Response Team (CERT-In).
*- Operating a 24x7 National Critical Information Infrastructure Protection Centre to function as the nodal agency for critical information infrastructure protection.
*- Reducing supply chain risks by creating and maintaining testing infrastructure and facilities for IT security product evaluation and compliance verification as per global standards and practices.
<p>Many of these strategies were followed through, albeit with different degrees of success when it comes to achieving sovereignty in the technological sphere.</p>
<p>To begin with, the NCSP led to the creation of the National Critical Information Infrastructure Protection Centre (“NCIIPC”). As the commanding authority accountable, the NCIIPC introduced the Information Security Practices and Procedures for Protected System Rules, 2018 (https://www.meity.gov.in/static/uploads/2024/02/NCIIPC-Rules-notification-1.pdf) (the “NCIIPC Rules”) and issued Guidelines for the Protection of National Critical Information Infrastructure, 2015 (https://www.asianlaws.org/gcld/cyberlawdb/IN/guidelines/NCIIPC_Guidelines_V2.pdf) (the “NCIIPC Guidelines”) to safeguard India’s CII from unauthorized access, modification, use, disclosure and disruption to ensure a safe, secure and resilient information infrastructure for critical sectors in the country. Among these rules, certain requirements stand out, such as the mandatory inventories of hardware and software related to CII (Section 3(3-e) of NCIIPC Rules) and the promotion of indigenous Research and Development (R&D) regarding the protection of CII (Section 4.3.7 of NCIIPC Guidelines).</p>
<p>In addition, the India Stack (the “Stack”) was a nationwide government program that recognized societal inequalities enhanced and perpetuated due to lack of access to the internet and technology. As a socialist endeavor, the Government launched platforms to dissipate welfare functions, promoting digital literacy and enhancing electronic governance. These systems were conceived as being mostly indigenous and open-source, allowing India to now export this technology framework to other nations (https://indianmasterminds.com/news/india-signs-mous-23-countries-digital-public-infrastructure-india-stack-upi-global-183393/) looking to build their own digital stack.</p>
<p>Furthermore, under the Public Procurement (Preference to Make in India) Order, 2017 (https://www.dpiit.gov.in/static/uploads/2025/07/a2243b925eacf5e116e0e3f43630a8b2.pdf), MeitY issued a notification designating cyber security as a strategic sector, stating that "preference shall be provided by all procuring entities to domestically manufactured/produced Cyber Security Products" with preference granted to companies incorporated in India whose product IP and licensing revenue accrue domestically, and royalty payments abroad capped at 50% of the product's cost.</p>
<p>Despite the above, technological dependency is still an issue in India. The policy on data centers is a clear example of this, since these facilities have proliferated in India during recent years under a sovereignty discourse. Under India's Consolidated FDI Policy Circular of 2020 (https://www.dpiit.gov.in/static/uploads/2025/07/3ab2ec2a3bdb91c69653b7c34618c14a.pdf ) (DPIIT), data centers have not required prior government or central bank approval. Implying that the nation intends to favor foreign investment in their construction. These incentives mean that the infrastructure for data centers “will be built upon American foundation models and run on American chips” (4), a policy that ultimately benefits hyperscalers without offering much in return for affected populations.</p>
<p>This dilemma showcases one of the main tensions still prevalent in India: developing indigenous infrastructure for critical services in order to prevent dependency on third countries (as shown with the Stack) vs boosting investment by encouraging foreign companies to build digital infrastructures within Indian sovereign territories.</p>
<p><br><br></p>
#Internet Sovereignty:
<p>Internet or network sovereignty is a notion invoked by protectionists or nationalist agendas in an attempt to “control the digital networks and information running through them”(5).</p>
<p>The NCSP provided an adequate advisory and guideline basis which helped protect infrastructures and public services upon which the Digital India Scheme (DIS) was promoted. DIS sought to expand access to the Internet of Things from city to rural areas to facilitate government functions and financial transactions digitally through secured high speed broadband and digital public infrastructures (DPI), a foundational block of the Stack (6).</p>
<p>The India Stack is a three-stage e-government infrastructure project based on the digitalized Aadhaar platform, comprised of the Identity Layer (developed since 2009), the Payment Layer (developed since 2010), and the final Data Layer (building on the previous two layers), designed to combat money laundering and corruption, better monitor financial flows, and provide the Indian population with simplified access to financial services (7).</p>
<p>This is a pivotal policy to analyze for a better grasp on the internet sovereignty sought by the Indian Government. The main achievements of this initiative are the following:</p>
#Aadhar (Identity Layer): Served as a nationwide security and biometric unique identity scheme that created the world’s largest digital public infrastructure.
#Unified Payments Interface (Payment Layer): Enabled simple digital transactions between two mobiles to transition into a cashless economy with additional financial schemes such as demonetization and zero-cost bank accounts.
#DigiLocker(Data Layer): A secure cloud storage space run by the government to secure personal data documents for official purposes without physical paper dependency.
<p>While being praised due to its efficiency and inclusionary potential, the stack’s increasing centrality in governance raises significant concerns. Significant concerns have been flagged as it relates to mass surveillance (https://www.giga-hamburg.de/en/publications/giga-focus/digital-surveillance-and-the-threat-to-civil-liberties-in-india) that consequently arises from the state's control over the network.</p>
<p>In addition, Section 69A of the IT Act (https://www.giga-hamburg.de/en/publications/giga-focus/digital-surveillance-and-the-threat-to-civil-liberties-in-india) bestows the Government powers over content, by ratifying the capacity to block on narrow grounds (sovereignty, security, public order, etc). Since 2023, though, the government has increasingly used a parallel channel, consisting of notice-and-takedown mechanisms through the "Sahyog" portal (operated by the Cyber Crime Coordination Centre of India) under Section 79(3)(b) (https://indiankanoon.org/doc/161852741/). This provision was intended for removal of content used for unlawful acts, and affected entities have claimed that the Government has attempted to extend its reach to override the restrictions built into Section 69A. This shift generated over 2,300 orders across 19 platforms (https://www.barandbench.com/amp/story/columns/section-69a-of-the-information-technology-act-2000-and-the-expanding-architecture-of-digital-censorship-in-india) in just one year (Oct 2024–Oct 2025) and is now being challenged in court (https://indiankanoon.org/doc/7614885/).</p>
<p>The dilemma presented above showcases a new tension within the Indian digital framework: while certain provisions allow the Central Government to maintain an effective control over the network and the contents shared through it, a significant risk for citizens privacy rights and individual freedoms is arising as a result.</p>
<p><br></p>
#Data sovereignty
<p>Finally, data sovereignty refers to “protecting and controlling information on and flowing through digital networks” (8). The Supreme Court's judgment in ''Justice K.S. Puttaswamy v. Union of India (https://indiankanoon.org/doc/127517806/) ''addressed the privacy concerns of the centralization of citizens’ data in Aadhaar, recognizing that the right to privacy would fall under Article 21 of the Constitution (originally reserved to the right to live). There has been much controversy around the societal ramification of a single identification mechanism including the exclusion of nonparticipants from fundamental public services.</p>
<p>The ''Puttaswamy'' decision formed the material foundation (9) of what would later be the Digital Personal Data Protection Act (https://www.meity.gov.in/static/uploads/2024/06/2bf1f0e9f04e6fb4f8fef35e82c42aa5.pdf), issued in 2023. It delineated two groups (the “Data Fiduciaries”):</p>
*- Processors of data in the Indian territory or
*- Processors of data outside of India, in the condition such processing is in connection with any activity related to offering of goods or services to Data Principals within Indian territory.
<p>This is one of the main elements that is worth mentioning about this framework, as the extraterritorial reach of the law attempts to extend India’s sovereignty over its data beyond the borders of the country.</p>
<p>Furthermore, Section 16(1) established that The Central Government may, by notification, restrict the transfer of personal data by a Data Fiduciary for processing to such country or territory outside India as may be so notified.</p>
<p>To complement the Act, additional Rules were issued in 2025 (https://www.meity.gov.in/static/uploads/2025/11/53450e6e5dc0bfa85ebd78686cadad39.pdf) following the objections and suggestions received from the public by the Central Government. According to Section 13 (4), a Data Fiduciary shall undertake measures to ensure that personal data specified by the Central Government, on the basis of the recommendations of a committee constituted by it, is processed subject to the restriction that the personal data and the traffic data pertaining to its flow is not transferred outside the territory of India.</p>
<p>The abovementioned provisions are a clear indicator of India’s attempt to strengthen the control over their citizen’s data. While international transfers are less restricted than under the EU approach, foreign Data Fiduciaries would still be subject to Indian law, and the Central Government is in position of both banning transfers to specific countries as well as establishing data localization requirements for certain data elements. This structure, while not identical to the GDPR, still manages to reinforce the country’s data sovereignty claims.</p>
<p><br></p>
*Comparison with the EU
<p>Although India and the EU both acknowledge technological sovereignty as a key objective, the corresponding frameworks, prioritization and approaches in the direction of achieving this starkly vary.</p>
<p>For instance, as it pertains to the domain of data governance and data protection, the EU has a comprehensive regulatory landscape that gives high-priority to the safeguarding of rights as evidenced with General Data Protection Regulation (“GDPR”) and the Digital Services Act, which institute well-defined rules on data privacy, and cross-border data flows. A distinguishing element of approaches relates to paradigms in place. The EU has adopted the ''‘allowed list’'' model that is rooted in adequacy decisions, while India has a ‘''negative list''’ model, meaning it permits cross-border data transfers unless specifically restricted. Both paradigms present different challenges and opportunities respectively. The European approach encourages third countries towards benchmarking for adequacy decisions, while India encourages strong relations to avoid a ban. This could explain the fact that India lacks an adequacy decision from the EU.</p>
<p>However, these two nations share similarities regarding their digital infrastructure policy goals. In a shared statement on the second meeting of the EU-India Trade and Technology Council (https://www.eeas.europa.eu/delegations/india/joint-statement-second-meeting-eu-india-trade-and-technology-council_en), both countries agreed to “collaborate on working towards interoperability of their respective DPIs that respect human rights and protect personal data, privacy, and intellectual property rights” and “promote the strengthening of the EU and Indian semiconductor ecosystems to enhance technological capabilities and ensure supply chain resilience by developing sustainable, secure and diversified semiconductor production capacities”. Both countries also attempt to extend the effect of their infrastructure regulations beyond its borders, with India’s IT Act applying to any offence or contravention committed outside India if the act or conduct involves a computer, computer system or computer network located in India (Section 75 https://indiankanoon.org/doc/576992/) and the EU requiring non-EU digital infrastructure and service providers offering services in the Union to appoint an EU representative (NIS2 Art. 26(3)).</p>
<p>Their main differences in that regard stem from the robustness of their legislation. While the EU has developed a strong sectorial cyber security framework in the form of the NIS 2 Directive (with the corresponding transpositions across Member States), India’s framework relies on an outdated law that generally applies to IT services (the IT Act) without the same degree of harmonized regulatory density. Furthermore, from a policy point of view, India has attempted to foster foreign innovation in developing digital infrastructures for data centers, favoring the presence of hyperscalers, whereas the EU has attempted to steer into the opposite direction and “support the development of European cloud computing offerings (https://oeil.europarl.europa.eu/oeil/en/document-summary?id=1656816)”. This showcases a major difference in addressing an already identified issue regarding their respective technological dependencies.</p>
<p><br></p>
<p>References:</p>
*(1) Biswajit Maitra and Moutushi Chakraborty, 'Foreign Trade, Human Capital and Economic Growth in India under the Liberalised Trade Regime' (2023) 5(1) Journal of Asian Economic Integration 29, doi:10.1177/26316846221147581.
*(2) Plixavra Vogiatzoglou and Joris van Hoboken, 'The EU's Digital Sovereignty and Quantum Technologies: To What End?' (2026) 18(2) Law, Innovation and Technology, doi:10.1080/17579961.2026.2705118, 2.
*(3) Stéphane Couture and Sophie Toupin, ‘What Does the Concept of “Sovereignty” Mean in Digital, Network and Technological Sovereignty?’ (Social Science Research Network, 22 January 2018) 5 https://doi.org/10.2139/ssrn.3107272
*(4) Vinit Ravishankar, 'Sovereignty After the Boom' (''Disjunctions'') https://disjunctionsmag.com/articles/sovereignty-after-boom/ accessed 24 September 2026.
*(5) Vogiatzoglou and van Hoboken (n 2) 6.
*(6) Snehal A. Bagul, 'Guarding the Digital Gateway: An In-Depth Analysis of Cybersecurity Challenges in India' in Faisal Rehman and others (eds), ''Emerging Trends in Information System Security Using AI & Data Science for Next-Generation Cyber Analytics'' (Springer 2025) 69–71.
*(7) Sezgin Sönmez, 'The India Stack – Between Networked Growth and Territorial Control' (2025) 50(4) Historical Social Research 318, https://www.jstor.org/stable/27426713.
*(8) Vogiatzoglou and van Hoboken (n 2) 6.
*(9) Committee of Experts under the Chairmanship of Justice B N Srikrishna, ''A Free and Fair Digital Economy: Protecting Privacy, Empowering Indians'' (Ministry of Electronics and Information Technology, 27 July 2018) https://www.thehinducentre.com/resources/article24561547.ece/binary/Data_Protection_Committee_Report-comp

= United States =
== The Concept of Digital Sovereignty ==
<p>The idea of “digital sovereignty” is not a single concept. There are a number of scholars who attempted to define digital sovereignty. [https://www.academia.edu/99043334/What_does_the_notion_of_sovereignty_mean_when_referring_to_the_digital Couture and Toupin] and [https://www.researchgate.net/publication/343611573_The_Fight_for_Digital_Sovereignty_What_It_Is_and_Why_It_Matters_Especially_for_the_EU Floridi] emphasized the '''control''' over the digital realm, in which Floridi saw a combination of democratic legitimacy with innovative flexibility by a hybrid system of control. On the other hand, [https://techreg.org/article/view/23778?articlesBySimilarityPage=1 Chung] looked into the aspect of '''authority''' by defining digital sovereignty as a state’s ability to define, implement and enforce political and legal decisions across digital systems, platforms and data infrastructures. She also reconceptualized the term by asserting that sovereignty should be understood as governance '''capacity''' rather than exclusive control. There is a variety of perspectives in determining what digital sovereignty means and conflicting definitions based on varying national strategies.</p><p>The definition of digital sovereignty adopted throughout this study is a merge of authority and practical capacity of a state to set, apply and defend its own rules governing the digital infrastructure, data flows and algorithmic systems that shape its legal and political orders, against interference by outside actors, whether public or private, including other levels of government within the same state.</p>
=== The United States Sovereignty Strategy: AI Dominance ===
<p>Unlike the EU who attempts to maintain digital sovereignty under its regulatory control, the idea of securing digital sovereignty is not its priority for the US. Their goal is to ensure geopolitical dominance by entrenching American advantage in the technologies that determine their economic and military standing, against a field of competitors, namely China as the principal one.</p><p>From ''[https://www.whitehouse.gov/wp-content/uploads/2025/07/Americas-AI-Action-Plan.pdf Winning the Race: America’s AI Action Plan]'', Issued on 23 July 2025, the US outlines 103 recommended actions by three pillars:</p>

*Pillar 1: Accelerate AI Innovation – removing “bureaucratic red tapes” which hinder technological advancement by deregulation
*Pillar 2: Build American AI Infrastructure – fostering an environment conducive to AI innovations
*Pillar 3: Lead in International AI Diplomacy and Security – Ensuring its global hegemonic position in AI technology<p>As explicitly stated from the plan, their core interest is in winning the race and securing its geopolitical primacy relative to other competitors rather than enhancing its authority and capacity over its domestic digital governance. The Plan itself has no legal force, yet the actual implementation is played by executive orders and procurements.</p>
== The Four Registers ==
<p>American digital policy has an organising purpose, and it is not regulation. It is primacy: maintaining American advantage in the technologies that determine economic and military standing, against a field of competitors, of which China is the principal one. The framing is stated in the instruments themselves — the order that set the current course is titled Removing Barriers to American Leadership in Artificial Intelligence, and the strategy it produced is titled Winning the Race. Beneath that strategy, the United States steers digital developments through four registers, descending in legal formality and ascending in speed.</p><p>'''Statute''' is the thinnest register. The [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] and the [https://www.congress.gov/bill/118th-congress/house-bill/7521 Protecting Americans from Foreign Adversary Controlled Applications Act] are both narrow and adversary-specific, and general legislation has failed: the Senate stripped a ten-year moratorium on state enforcement of AI laws from the One Big Beautiful Bill Act by a vote of 99 to 1.</p><p>'''Executive orders''' carry the load. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] produced the Data Security Program, an export control keyed to adversary identity rather than to any standard of protection; Executive Order 14320 finances the export of full-stack American AI packages; Executive Order 14365 asserts a national AI framework and directs litigation against inconsistent state laws.</p><p>'''Procurement''' is quieter and probably more effective, and is illustrated below by the cloud and semiconductor sectors.</p><p>'''Announcement''' is the fastest register and the least examined. On 19 January 2025 the President posted on Truth Social that he would issue an order extending the statutory TikTok divestment deadline, and [https://www.theguardian.com/us-news/2025/jan/19/trump-tiktok-ban-executive-order TikTok restored service] in the United States the same day, before any instrument existed. A platform serving some 170 million US users changed its conduct based solely on anticipated legality.</p>
=== Direction of Travel ===
<p>The direction of policy is consistent across all four registers and runs four ways. Outward, the export programme pushes the domestic AI stack into third markets. Inward, the Data Security Program and Team Telecom review screen foreign access based on a counterparty's identity rather than the safeguards it offers. Homeward, subsidy builds domestic capacity. Downward, federal policy presses on the states, and it is working even though preemption ordinarily requires a congressional enactment: Colorado repealed its 2024 AI Act in May 2026 and replaced it with a narrower automated decision-making statute stripped of deployer impact-assessment duties, while California's frontier AI statute (SB 53) contains a federal-deference clause under which compliance with a comparable federal framework discharges the state obligation. States are deferring in advance to a federal standard that does not yet exist.</p><p>Outward and homeward movement build the US position; inward and downward movement defend it against interference, foreign in the former case and domestic in the latter. The same adversary-identity test recurs in an export control, a subsidy condition, and a screening review alike.</p>

== US Infrastructure and Resources efforts for AI dominance ==

=== Supply Chain management and the Cloud ===

The United States aims to maintain its position in the [https://www.cambridge.org/core/journals/international-organization/article/digital-disintegration-technoblocs-and-strategic-sovereignty-in-the-ai-era/DD86C6FD3FDD7FBBADEF100C6935D577 international order, seeking to surpass and constrain rival powers]. Technological dominance constitutes one of the principal means of pursuing this objective, and in recent years this pursuit has centred increasingly on artificial intelligence, a domain heavily dependent on underlying industrial capacity and resource access.

Cloud service dependency describes a form of [https://www.cambridge.org/core/journals/review-of-international-studies/article/susan-strange-meets-the-everyday-the-mundane-sources-of-structural-power/1D23361D622C523F7EB785EA54105858 structural power] exercised not through territorial control but through infrastructural ownership. Because US firms hold roughly [https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 two-thirds of the EU cloud market], a majority of European data, public and private, resides within entities potentially subject to US jurisdiction, irrespective of the server's physical location. This constitutes strategic advantage precisely because it requires no conventional projection of power abroad: dominance is achieved by occupying the infrastructural layer itself, converting market position into jurisdictional reach and, in turn, translating informational access into epistemic and strategic advantage. The [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] grants the United States access to foreign data provided this is held by a US cloud service provider, access guaranteed only through a defined legal process, the issuance of a judicial warrant subject to probable cause. [https://www.law.cornell.edu/uscode/text/50/1881a FISA Section 702], by contrast, authorises access to foreign data held in US clouds for intelligence purposes through compelled provider assistance, and in this instance no comparable judicial procedure is required by law. These divergent standards of authorisation, and the reach they enable, stand in direct tension with the protections EU law seeks to guarantee.

The United States nonetheless maintains mechanisms to protect itself from equivalent foreign control over its own data, an arrangement not without irony. Through these safeguards, the United States continues to consolidate a form of dominance that might properly be termed digital sovereignty, in that it is the party setting the terms of engagement, monitoring others' data while shielding its own from comparable scrutiny.

[https://www.fedramp.gov/ FedRAMP] establishes the conditions under which any cloud provider, including foreign entities, may process US federal government data, asserting sovereign gatekeeping over access to its own infrastructure even as it resists equivalent constraints abroad. [https://www.federalregister.gov/documents/2024/03/01/2024-04573/preventing-access-to-americans-bulk-sensitive-personal-data-and-united-states-government-related Executive Order 14117] and the accompanying [https://www.federalregister.gov/documents/2025/01/08/2024-31486/preventing-access-to-us-sensitive-personal-data-and-government-related-data-by-countries-of-concern DOJ rule (2025)], although criticised for their generalistic scope, restrict the outbound transfer of bulk sensitive data to designated "countries of concern." The asymmetry is instructive when set against the EU's broadly equivalent instruments: both regimes exercise control, but over opposite targets, the United States restricts flows to named adversaries, while the EU restricts flows to any third country whose legal framework it deems inadequate, a criterion of general application rather than one confined to a defined list of hostile states.

The EU's principal instrument of power has consequently been regulatory rather than infrastructural, expressed through the [https://gdpr-info.eu/art-48-gdpr/ GDPR (Art. 48)] and the [https://eur-lex.europa.eu/eli/reg/2023/2854/oj/eng Data Act], mechanisms of constraint and protection rather than instruments of reach comparable to the [https://www.justice.gov/d9/pages/attachments/2019/04/09/cloud_act.pdf CLOUD Act] or [https://www.law.cornell.edu/uscode/text/50/1881a FISA]. There is a clear divergence in objective these reflect, and it is rooted in the different conception the EU has of human rights, values and norms, related to data protection and privacy in comparison to the US'. Worth mentioning, the EU's position is further constrained by a material deficit, limited infrastructural capacity, financial resources.

As of right now, the United States can circumvent the normative safeguards the EU has imposed in some cases (through [https://www.law.cornell.edu/uscode/text/50/1881a FISA]), and it already possesses the infrastructural capacity the EU lacks ([https://www.srgresearch.com/articles/european-cloud-providers-local-market-share-now-holds-steady-at-15 only 13 to 15 percent of the market is EU-owned]). This gives the US vast dominance in the field, which could be equated to what the EU coins as cloud sovereignty. Aware of the risks this dependency poses, both to data access and to the strategic advantage it affords the US, the EU has sought to reduce it through a succession of initiatives. [https://www.enisa.europa.eu/publications/eucs-cloud-service-scheme EUCS] was one such attempt, and remains only a non-legally-binding standard, stalled, per the [https://www.iss.europa.eu/publications/briefs/technical-political-when-cloud-certification-scheme-divides-europe EUISS], by Member States divided over their reliance on US providers; whether the [https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026PC0502 Cloud and AI Development Act] fares better remains to be seen.

=== Supply Chain management and Semiconductors ===
The United States and the European Union are equally invested in developing a domestic semiconductor industry, what separates them is their approach, the risks willing to run, the tools willing to use, and, above all, the capabilities and resources each power has. The US specifially aims to develop a bigger industry in this sector to rival competing hegemonic powers, to then gain AI dominance, strategic advantage and a stronger position in the international order.

The centrepiece is the [https://www.govinfo.gov/content/pkg/PLAW-117publ167/pdf/PLAW-117publ167.pdf CHIPS and Science Act] of August 2022, a statute that funds domestic fabrication and research through grants, loans and awards administered by the Department of Commerce. Its force comes from its scale: $39 billion in manufacturing incentives (including $2 billion reserved for legacy chips) and $11 billion for R&D, a federal pool officials themselves [https://www.reuters.com/technology/us-picks-team-oversee-527-bln-semiconductor-funding-2022-09-20/ described at the time] as $52.7 billion, on top of a Section 48D investment tax credit set from 25 to 35 percent in 2025. This is a single, centralised federal pool, commanded and disbursed by one government. The [https://eur-lex.europa.eu/eli/reg/2023/1781/oj/eng EU Chips Act (Regulation (EU) 2023/1781)] pursues the identical objective, a resilient domestic semiconductor base, but with a wholly different reach: its own budget is capped at a "maximum indicative amount" of just EUR 3.3 billion and decentralised Member State state aid. The US pursues to clash with other powers and surpass them through several measures in this statute, none of which the EU has an equivalent to.

The [https://www.federalregister.gov/documents/2023/09/25/2023-20471/preventing-the-improper-use-of-chips-act-funding CHIPS guardrails rule] binds every recipient of the funds for ten years: the Expansion Clawback bars material expansion, defined as growth of more than 5 percent, of semiconductor manufacturing capacity in a "foreign country of concern". These are protection measures, converting a subsidy into standing control over where a funded firm may grow and with whom it may cooperate, minimising ties to rival powers while building US capacity unilaterally.

The [https://www.federalregister.gov/documents/2022/10/13/2022-21658/implementation-of-additional-export-controls-certain-advanced-computing-and-semiconductor BIS export-control rule] does two things. First, it blocks advanced chips and chip-making equipment from reaching ‘countries of concern’ even when made outside the US, if US' technology, software or tools were used somewhere along the way to make them. Second, it stops any American, wherever in the world they work, from helping build advanced chips for those countries, even in a support role like shipping or servicing. What the US rule shows, in a way cloud policy does not, is active separation: a deliberate policy of isolating adversary states from American inputs, expertise and even American citizens' labour, building domestic capacity at home rather than permitting it to leak abroad through investment, export or cooperation.

The [https://www.federalregister.gov/documents/2026/01/20/2026-01052/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative Section 232 proclamation] of 14 January 2026 shows the same modus operandi differently: a 25 percent tariff on certain advanced chips, and, in August 2025, the [https://newsroom.intel.com/corporate/intel-and-trump-administration-reach-historic-agreement conversion of $8.9 billion owed to Intel into a 9.9 percent equity stake] in the company. Neither is a law of general application; both are one-off executive moves, using a tariff or an ownership stake to get what the government wants directly. The US reaches for leverage and deal-making, not just regulation, a tool the EU does not use.

CHIPS award recipients are also barred by law from using federal funds to pay shareholder dividends or buy back stock, a restriction the [https://www.nist.gov/document/approach-capital-allocation-policies CHIPS Program Office] has sought to extend for five years from the date of an award.

Each of these examples have the same objective, that federal money spent on the domestic industry should build capacity, not enrich shareholders or, worse, end up financing a rival through some indirect channel. Subsidy, clawback, denial, tariff, ownership and now buyback limits, each is a different tool, but each answers to the same objective, an America made big and sovereign enough to win the race for AI dominance.

=== Supply Chain Management and Rare Earth Minerals ===
<p class="mwt-heading">The United States' pursuit of AI dominance rests on a chain of physical dependencies: AI systems require advanced semiconductors, semiconductors require rare earth elements and critical minerals, particularly the neodymium, dysprosium, and gadolinium used in the magnets and components of chip fabrication and data-centre hardware, and the United States currently remains reliant on processing capacity concentrated in China. Where the [https://www.congress.gov/bill/117th-congress/house-bill/4346 CHIPS and Science Act] addresses the fabrication layer, a separate and more fragmented body of law addresses the mineral layer beneath it. Unlike the EU's [https://eur-lex.europa.eu/eli/reg/2024/1252/oj/eng Critical Raw Materials Act], the American framework layers a foundational statute, an executive order, and a narrower defense-procurement statute.<br></p>

==== Energy Act of 2020 ====
Section 7002 of the Energy Act of 2020 (codified at [https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=30-USC-185389929-22931526&term_occur=999&term_src= 30 U.S.C. §1606]) supplies the foundational definitions. It defines a "critical mineral" through a delegated standard: a "byproduct" is a critical mineral whose recovery depends on the production of a host mineral not itself designated critical, while a "critical material" is any non-fuel mineral, element, substance, or material the Secretary of Energy determines carries a high risk of supply-chain disruption. This is primarily an administrative and classificatory act: it does not itself compel production, but it creates the taxonomy (which of the 17 rare earth elements are legally "critical") on which subsequent executive and procurement action depends.

==== Executive Order 14241 (2025) ====
[https://public-inspection.federalregister.gov/2025-05212.pdf Executive Order 14241], entitled "Immediate Measures to Increase American Mineral Production," signed March 20, 2025, operationalizes that taxonomy. Its Section 1 purpose clause is unusually candid about the framing: national and economic security are now threatened by dependence on hostile foreign powers. Operationally, Section 3 directs every federal agency involved in permitting to submit to the National Energy Dominance Council a list of all mineral production projects for which a plan of operations or permit application has been submitted, after which agency heads must identify projects that can be immediately approved and expedite them. The Order invokes [https://www.law.cornell.edu/uscode/text/50/4533 Title III of the Defense Production Act] to mobilize financing, and tasks the NEDC, chaired by the Secretary of the Interior, with coordinating implementation across Defense, Energy, Interior, and Agriculture. Unlike the CHIPS Act, it required no congressional appropriation, and unlike the Energy Act, it does not merely classify inputs but seeks to physically build an alternative domestic supply chain, the mining-and-processing analogue to the CRMA's "strategic projects" mechanism. Executive Order 14241 has since been diluted by [https://www.whitehouse.gov/presidential-actions/2025/04/reinvigorating-americas-beautiful-clean-coal-industry-and-amending-executive-order-14241/ Executive Order 14261], which extends designation of criticality to sectorally unrelated materials such as coal.

==== 10 U.S.C. §4872 and the DFARS ====
The narrowest instrument is [https://www.law.cornell.edu/uscode/text/10/4872 10 U.S.C. §4872], which provides that the Secretary of Defense may not procure any covered material melted or produced in any covered nation, or any end item containing such material. [https://www.law.cornell.edu/definitions/uscode.php?def_id=10-USC-1197798417-780045830&term_src=title:10:subtitle:A:part:V:subpart:I:chapter:385:subchapter:III:section:4872 “Covered nations"] are defined as North Korea, China, Russia, and Iran, covered materials include samarium-cobalt and neodymium-iron-boron magnets. Implemented through [https://www.acquisition.gov/dfars/225.7018-2-restriction. DFARS 225.7018-2], the clause currently bars acquisition of any covered material melted or produced in any covered country, tightening in 2027 to cover the entire chain. This statute does not build supply; it leverages the federal government's own purchasing power to exclude adversary-sourced inputs from the defense industrial base, a tool with no direct EU parallel, since the CRMA's diversification targets operate through incentive rather than exclusion.

Social-Media Governance in the EU - by Sergey Abramov

2026-09-23T05:01:30Z

Sergey Abramov:

<p>'''Social-Media Governance in the EU'''</p>
<p>Social-media platforms in the European Union are governed through overlapping public and private institutions. Four institutions illustrate the resulting distribution of authority and its limitations.</p>
<p>'''The European Commission''', a public institution, derives its supervisory authority from the Digital Services Act (DSA) article 56(2)-(3). Its governing instruments include binding enforcement decisions and fines under articles 73-74. It exclusively supervises the additional obligations imposed on designated very large platforms, including systemic-risk assessment and mitigation under articles 34-35. These obligations protect fundamental rights, public safety and democratic processes. Its capacity weakness is informational dependence: assessing algorithmic risks requires substantial technical knowledge held by regulated companies. [1]</p>
<p>'''The Netherlands Authority for Consumers and Markets (ACM)''' is a public institution designated as the Dutch Digital Services Coordinator under DSA article 49 and Dutch implementing legislation. Its instruments include investigations, compliance orders and fines under article 51, alongside complaint handling under article 53. It protects users’ rights, particularly fair moderation procedures and children’s safety. Its capacity limitation is cross-border dependence: under article 56, supervision generally follows the provider’s main establishment, subject to the Commission’s powers. ACM therefore often depends on other authorities to address Dutch users’ complaints, illustrating Black’s account of regulatory interdependence. [2]</p>
<p>'''Meta''', a non-state corporate institution operating Facebook and Instagram, derives governing authority from user contracts and control over platform infrastructure. Its instruments are terms of service, Community Standards and moderation decisions. It claims to protect safety while enabling expression. DSA article 14(1) and (4) requires transparent rules and proportionate enforcement attentive to fundamental rights; it constrains rather than creates Meta’s private authority. [3] Its legitimacy weakness is that users have limited influence over rules governing their participation. </p>
<p>'''Appeals Centre Europe''', a non-state dispute-resolution body, derives its recognised role from certification by Ireland’s Coimisiún na Meán under DSA article 21(3). Its instruments are procedural rules and decisions reviewing platform moderation. It seeks to protect fair treatment and access to redress. Its principal capacity weakness is explicit in article 21(2): decisions are non-binding, although platforms must engage in good faith. Independent review therefore does not guarantee implementation. [4]</p>
<p>These arrangements reflect Lobel’s view of governance through cooperation between public and private actors. [5] Yet shared responsibility does not guarantee equal influence or accountability. Effective governance requires users to understand decisions, challenge them and obtain meaningful remedies. The DSA supports these goals, but their achievement depends on effective oversight and enforcement.</p>
<p><br></p>
<p>Footnotes</p>
#Regulation (EU) 2022/2065 of the European Parliament and of the Council on a Single Market for Digital Services and amending Directive 2000/31/EC (Digital Services Act) [2022] OJ L277/1 (‘DSA’), arts 34–35, 56(2)–(3), 67 and 73-74 <https://eur-lex.europa.eu/eli/reg/2022/2065/oj/eng> accessed 22 September 2026.
#DSA (n 2), arts 49, 51, 53 and 56; Netherlands Authority for Consumers and Markets, Digital Services Act: Annual Activity Report 2025 (2026) sections 1.1-1.2 and 2.2 <https://www.acm.nl/en/publications/acm-2025-dsa-annual-report> accessed 22 September 2026; Julia Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-Regulatory” World’ (2001) 54 Current Legal Problems 109-12 <https://doi.org/10.1093/clp/54.1.103> accessed 22 September 2026).
#DSA (n 2), art 14(1) and (4); Meta, ‘Safety Policies’ <https://www.meta.com/en-gb/safety/topics/safety-basics/policies/> accessed 22 September 2026.
#DSA (n 2), art 21(2)-(3); Coimisiún na Meán, Information Note re the Decision to Certify the Appeals Centre Europe as an Out-of-Court Dispute Settlement Body pursuant to Article 21 of the Digital Services Act (May 2025) 3-8 <https://www.cnam.ie/app/uploads/2025/05/ACE-Info-Note-final-version-English-for-website.pdf> accessed 22 September 2026; European Commission, ‘Out-of-Court Dispute Settlement Bodies under the Digital Services Act (DSA)’ (11 September 2026) <https://digital-strategy.ec.europa.eu/en/policies/dsa-out-court-dispute-settlement> accessed 22 September 2026.
#Orly Lobel, ‘The Renew Deal: The Fall of Regulation and the Rise of Governance in Contemporary Legal Thought’ (2004) 89 Minnesota Law Review 342, introduction and pt III <https://scholarship.law.umn.edu/mlr/663/> accessed 22 September 2026.

====== <br> ======
<br>'''Comment from Sebastián:''' excelent work Sergey! I would only add that user engagement is one of Meta's greatest interests behind its governance efforts, despite it not being properly disclosed.

Social-Media Governance in the EU - by Sergey Abramov

2026-09-23T04:55:30Z

Sergey Abramov: /* Social-Media Governance in the EU */

<p>'''Social-Media Governance in the EU'''</p><p>Social-media platforms in the European Union are governed through overlapping public and private institutions. Four institutions illustrate the resulting distribution of authority and its limitations.</p><p>'''The European Commission''', a public institution, derives its supervisory authority from the Digital Services Act (DSA) article 56(2)-(3). Its governing instruments include binding enforcement decisions and fines under articles 73-74. It exclusively supervises the additional obligations imposed on designated very large platforms, including systemic-risk assessment and mitigation under articles 34-35. These obligations protect fundamental rights, public safety and democratic processes. Its capacity weakness is informational dependence: assessing algorithmic risks requires substantial technical knowledge held by regulated companies. [1]</p><p>'''The Netherlands Authority for Consumers and Markets (ACM)''' is a public institution designated as the Dutch Digital Services Coordinator under DSA article 49 and Dutch implementing legislation. Its instruments include investigations, compliance orders and fines under article 51, alongside complaint handling under article 53. It protects users’ rights, particularly fair moderation procedures and children’s safety. Its capacity limitation is cross-border dependence: under article 56, supervision generally follows the provider’s main establishment, subject to the Commission’s powers. ACM therefore often depends on other authorities to address Dutch users’ complaints, illustrating Black’s account of regulatory interdependence. [2]</p><p>'''Meta''', a non-state corporate institution operating Facebook and Instagram, derives governing authority from user contracts and control over platform infrastructure. Its instruments are terms of service, Community Standards and moderation decisions. It claims to protect safety while enabling expression. DSA article 14(1) and (4) requires transparent rules and proportionate enforcement attentive to fundamental rights; it constrains rather than creates Meta’s private authority. [3] Its legitimacy weakness is that users have limited influence over rules governing their participation. </p><p>'''Appeals Centre Europe''', a non-state dispute-resolution body, derives its recognised role from certification by Ireland’s Coimisiún na Meán under DSA article 21(3). Its instruments are procedural rules and decisions reviewing platform moderation. It seeks to protect fair treatment and access to redress. Its principal capacity weakness is explicit in article 21(2): decisions are non-binding, although platforms must engage in good faith. Independent review therefore does not guarantee implementation. [4]</p><p>These arrangements reflect Lobel’s view of governance through cooperation between public and private actors. [5] Yet shared responsibility does not guarantee equal influence or accountability. Effective governance requires users to understand decisions, challenge them and obtain meaningful remedies. The DSA supports these goals, but their achievement depends on effective oversight and enforcement.</p><p><br></p><p>Footnotes</p>
#Regulation (EU) 2022/2065 of the European Parliament and of the Council on a Single Market for Digital Services and amending Directive 2000/31/EC (Digital Services Act) [2022] OJ L277/1 (‘DSA’), arts 34–35, 56(2)–(3), 67 and 73-74 <https://eur-lex.europa.eu/eli/reg/2022/2065/oj/eng> accessed 22 September 2026.
#DSA (n 2), arts 49, 51, 53 and 56; Netherlands Authority for Consumers and Markets, Digital Services Act: Annual Activity Report 2025 (2026) sections 1.1-1.2 and 2.2 <https://www.acm.nl/en/publications/acm-2025-dsa-annual-report> accessed 22 September 2026; Julia Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-Regulatory” World’ (2001) 54 Current Legal Problems 109-12 <https://doi.org/10.1093/clp/54.1.103> accessed 22 September 2026).
#DSA (n 2), art 14(1) and (4); Meta, ‘Safety Policies’ <https://www.meta.com/en-gb/safety/topics/safety-basics/policies/> accessed 22 September 2026.
#DSA (n 2), art 21(2)-(3); Coimisiún na Meán, Information Note re the Decision to Certify the Appeals Centre Europe as an Out-of-Court Dispute Settlement Body pursuant to Article 21 of the Digital Services Act (May 2025) 3-8 <https://www.cnam.ie/app/uploads/2025/05/ACE-Info-Note-final-version-English-for-website.pdf> accessed 22 September 2026.
#Orly Lobel, ‘The Renew Deal: The Fall of Regulation and the Rise of Governance in Contemporary Legal Thought’ (2004) 89 Minnesota Law Review 342, introduction and pt III <https://scholarship.law.umn.edu/mlr/663/> accessed 22 September 2026.

====== <br> ======
<br>'''Comment from Sebastián:''' excelent work Sergey! I would only add that user engagement is one of Meta's greatest interests behind its governance efforts, despite it not being properly disclosed.

Governance by history

2026-09-22T20:01:07Z

Sergey Abramov: /* contributions */

[[File:Old internet.jpg|300px|thumb]]

= contributions =
*can
*you 
*please
*create
*individual pages
*from the contributions below
*and link them back here? 
*[[Information Control and Workplace Surveillance - Jihae Park and Mariona Fortuny]]
*[https://wiki.ivir.nl/index.php?title=Centralization_of_Compute_and_Data Centralization of AI Compute and Data - by Sergey Abramov and Paul Stein]

= links things that popped up during class =
*[https://craphound.com/ Cory Doctorow]
*[https://dougengelbart.org/content/view/209/ The mother of All demos] (first mouse, graphical interface, word processor, etc)
*Can a simple sensor be political? [https://gizmodo.com/why-cant-this-soap-dispenser-identify-dark-skin-1797931773 well, this "racist" soap dispenser made headlines a couple of years ago.]
*Podcast series: [https://www.the-santiago-boys.com/ Morozov: the Santiago Boys.]
*Anatomy on AN AI. [[File:Ai-anatomy-map.pdf|check out the pdf here]]

= Workplace surveillance =
== Information Control and Workplace Surveillance - by Jihae Park and Mariona Fortuny Jané ==

==== <span style="font-size: 12pt;">The Current Governance Issue</span> ====
<span style="font-size: 10pt;">There are numerous current governance problems arising from technological innovations, yet here we will specifically focus on the information control and workplace surveillance. Whether intended or unintended, there were certain historical incidents which resulted in social and political consequences. As noticed from the Control Revolution, the dramatic evolution in mass media and telecommunications technologies has led to the reinforcement of the economic and political control during the Industrial Revolution. ([https://archive.org/details/beniger-james-the-control-revolution Beniger, 1986]) By applying this logic to a modern example, this is clearly how many corporations are getting in control of their employees through surveillance technologies. Along with the emergence of remote and hybrid work after Covid-19, employers are now relying on “bossware” to analyze the emails, chats, and messages of their employees on platforms such as Teams or Slack. ([https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263 Ollier-Malaterre et al., 2026, p.237]) Some companies have strict restrictions on transferring data both internally and externally in protection of confidential information and consistently monitor their workers to the certain extent which could slow down their efficiency. This is problematic as it’s not only violating the privacy of their employees but also depriving employees of rights they should be naturally be able to exercise.</span>

<span style="font-size: 12pt;"></span>

==== <span style="font-size: 12pt;">Historical Lineage</span> ====
<span style="font-size: 12pt;"><span style="font-size: 10pt;">In relation to the current issue of the information control and workplace surveillance, Max Weber’s argument of bureaucracy and rationalization in the middle and late nineteenth century are good historical examples which are still frequently visible nowadays. Weber stated how the range of control could extend not only by expanding the capability to process information but also by limiting the amount of information to be processed. Here, the former is directly related to the control implemented by bureaucratization (in the mid 19th century) and computerization (in the late 20th century), whereas the latter addresses rationalization. An example from bureaucracy is the expansion of paperwork in response to the growth in information. We can still note these traces of bureaucracy from numerous procedures that we have to go through for a decision-making to take place in corporations. He brings up the example of standardized time zones for rationalization which is simply an example for an omission or a breakdown of information to facilitate its processing. ([https://archive.org/details/beniger-james-the-control-revolution Beniger, 1986, p.15-16]) This is also commonly seen from nowadays as some companies may use internal terminologies for the sake of both confidentiality and efficiency.</span></span>

==== <span style="font-size: 12pt;">Mechanism of Path Dependence</span> ====<p>Following along the line with the historical lineage, there are certain mechanisms at work of path dependence for information control and workplace surveillance. Bureaucracy is closely related with organizational routines. Many companies have already settled down with a fixed workflow and decision-making process, which makes it more costly for one to change from its conventional way of working. This means switching costs from an old system to a new one is clearly high considering its time and capital, not necessarily will guarantee increasing returns. Hence, many firms end up maintaining its old strategy, reluctant to make any modifications or changes.<br></p>

==== <span style="font-size: 12pt;">Politics and Distribution</span> ====

As businesses grew too large for managers to oversee everything directly, companies replaced human judgment with formal rule-systems, as we see explained by Beniger's 'intellectual technology' ([https://archive.org/details/beniger-james-the-control-revolution/page/65/mode/2up Beniger, 1986, pp. 17, 21]). Algorithmic management continues this: a call once made by a supervisor's judgment is now a rule built into software that runs without one, extending a logic Beniger traces as far back as to Taylorism and the so called 'assembly line'. Winner ([https://www.jstor.org/stable/20024652 1980]) asks whether such systems must concentrate power this way or were driven there, when other designs were equally possible; nothing required shutting workers out of the decision, so leaving them out was a political choice, and not necessarily a technical necessity. Barbrook and Cameron ([https://doi.org/10.1080/09505439609526455 1996]) explain the reasoning why some can defend this 'domination': they show how 1990s tech culture sold new technologies as freedom even as it tightened control, so the rhetoric could become one of closer oversight as flexibility and autonomy. Together, authority accrues to employers, who in some systems can set and revise these oversight rules unilaterally and face less resistance because the arrangement is framed as freedom; workers bear the cost, losing the informal give-and-take they had with a human supervisor for oversight that is harder to see, question, or contest, while being told it grants them more freedom, not less.

==== <span style="font-size: 12pt;">Contingency and Alternatives</span> ====

A plausible alternative happened on U.S. post WWII: in some workplaces, labor and its pacing was demanded to be set through collective bargaining and enforced jointly, by both management and union stewards monitoring compliance on the line ([https://doi.org/10.1093/icc/dtaf038 Breznitz et al., 2025, p. 211]). So, in this context, workplace surveillance could have developed as a bargained, dual-administered practice rather than a management monopoly. The 1950 [https://en.wikipedia.org/wiki/Reuther%27s_Treaty_of_Detroit Reuther's Treaty of Detroit ] cut that possibility off: unions won guaranteed wages and benefits in exchange for surrendering any general claim over how work was supervised ([https://doi.org/10.1093/icc/dtaf038 Breznitz et al., 2025, pp. 210–211]). Workers depended on wages to survive, and employers controlled the wage relationship. That asymmetry meant employers, given their position of power, were able to arrange things so that, at the very least, workers would have far less capacity to sustain a fight over control itself. Thus, the "trade" wasn't a fair exchange between equals. This reflects which side could afford to hold out on which issue. Hypothetically speaking, had that joint administration model prevailed instead, this could have greatly impacted norms internationally and good practices with worker-union agreed surveillance would have become a widely accepted standard practice.

==== <span style="font-size: 12pt;">Governance Implication</span> ====

An interesting proposition for present-day governance that wants to fill a gap in workplace surveillance is what [https://doi.org/10.1016/j.digbus.2025.100120 Kayas, Ong and Belal (2025)] are proposing: They argue that EU law, through the [https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng GDPR] and its required [https://en.wikipedia.org/wiki/Data_protection_impact_assessment data protection impact assessments (DPIAs)], already offers meaningful protection for workers in terms of lawful high-risk data collection and data processing. Nevertheless, these DPIAs are built around data and privacy, so they leave out several dimensions of monitoring itself: its purpose, how intrusive it is, how visible it is to the people being watched, and its effects on workers' autonomy, trust, and standing relative to management. These authors propose a "surveillance impact assessment" scoring a system against these characteristics directly, before any surveillance system is put in place. Although this mechanism "is constrained by its reliance on existing literature and the lack of empirical validation," it feels like a genuine alternative for two reasons: 1. it targets the actual mechanism at issue, and 2. at the same time it builds upon a mechanism of active participatory approach where employees have a direct say in how surveillance is designed ([https://doi.org/10.1016/j.digbus.2025.100120 Kayas et al., 2025, p. 11]), which, combined with the actual assessment, would put workers' own judgment at the centre of the one decision that affects them. It's important to note that this alternative is proposed as worth piloting, not as a proven fix, and it is fair to ask whether another mandatory assessment stacked on top of DPIAs would function as intended or simply read to workers as more compliance paperwork.

==== References ====
<br>Barbrook, R., & Cameron, A. (1996). The Californian Ideology. ''Science as Culture'', 6(1), 44–72. [https://doi.org/10.1080/09505439609526455 https://doi.org/10.1080/09505439609526455]<br>Beniger, J. R. (1986). ''The Control Revolution: Technological and Economic Origins of the Information Society''. Harvard University Press. [https://archive.org/details/beniger-james-the-control-revolution https://archive.org/details/beniger-james-the-control-revolution] <br>Breznitz, D., Levy, K., Lipartito, K., & Zehavi, A. (2025). An equity-focused research agenda for workplace surveillance. ''Industrial and Corporate Change'', 35(1), 206–221. [https://doi.org/10.1093/icc/dtaf038 https://doi.org/10.1093/icc/dtaf038]<br>Kayas, O. G., Ong, C. E., & Belal, H. M. (2025). From humans to algorithms: A sociotechnical framework of workplace surveillance. ''Digital Business'', 5, 100120. [https://doi.org/10.1016/j.digbus.2025.100120 https://doi.org/10.1016/j.digbus.2025.100120]<br>Ollier-Malaterre, A., Parent-Rocheleau, X., Benkrid, K., & Provost Savard, Y. (2026). Making Sense of Workplace Surveillance: How Workers and Union Representatives Experience, Imagine, and Resist Surveillance Exposure. ''Surveillance & Society'', 24(3), 236–254. [https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263 https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263]<br>Winner, L. (1980). Do Artifacts Have Politics? ''Daedalus'', 109(1), 121–136. [https://www.jstor.org/stable/20024652 https://www.jstor.org/stable/20024652]

<br>
== [[Algorithmic management and workplace surveillance - by Sebastián Muller and Kate Davey''' ''']] ==

=== 1. The current issue. ===
<br><p>Algorithmic management and workplace surveillance can be defined as the use of computer-programmed procedures to coordinate labor input in an organization ([https://joint-research-centre.ec.europa.eu/scientific-activities/employment/algorithmic-management-and-digital-monitoring-work_en 1]). Simply put, it is a way to control workers, and workers generally have no say in how the system that tracks them is created. As Winner says "different people are differently situated and possess unequal degrees of power as well as unequal levels of awareness" ([https://www.jstor.org/stable/20024652?seq=7 2]).</p>
<p>The population affected is workers; both “white collar” and “blue collar” workers are impacted. Whether it’s Amazon tracking warehouse workers’ “Time Off Task” with handheld scanners ([https://www.wired.com/story/amazon-worker-tracking-details-revealed 3]) or Meta employees having their every click tracked ([https://www.techpolicy.press/metas-worker-surveillance-tests-eu-rules-on-ai-and-labor/ 4]) to train AI, employers have the power to watch and control their employees in an unprecedented way.</p>
<p>It matters legally because workers have very few rights to fight back against these surveillance programs; even the The Electronic Communications Privacy Act of 1986 allows exceptions to surveillance of employees ([https://bja.ojp.gov/program/it/privacy-civil-liberties/authorities/statutes/1285 5]).<br></p>
=== <br>2. Historical lineage. ===<p><br>2.1 Panopticon: it is an architectural apparatus developed by Jeremy Bentham in 1791. The design of this building effectively generated the sense of constant monitoring, which in turn motivated the prisoners to behave as if they were being watched at all times. As human labor became commodified, capitalists began implementing measures to ensure the maximization of productivity, including the replication of this design ([https://doi.org/10.14452/MR-070-09-2019-02_2 6]). While unused today, the panopticon is widely considered as both a theoretical and practical inspiration of other forms of surveillance in the workplace.</p>
<p><br></p>
<p>2.2 Scientific management: this theory attempts to analyze and standardize workflows to improve labor productivity. Frederick Winslow Taylor, its main pioneer, used early forms of data analysis during the XX century to decrease walking distances, eliminating unnecessary movements and preventing deliberate slow-walking among workers. ([https://doi.org/10.14452/MR-070-09-2019-02_2 7]). There are many modern forms of data-driven decision-making processes in the workforce. For instance, Amazon warehouses track individual workers' scan rates in real time through handheld scanners, generating data feeds to track and evaluate performance ([https://www.theverge.com/2019/4/25/18516004/amazon-warehouse-fulfillment-centers-productivity-firing-terminations 8]).</p>
<p><br></p>
<p>2.3 PC in the workplace: John Blankenbaker designed the first commercially available personal computer in 1971 ([https://www.computerhistory.org/revolution/personal-computers/17/297 9]), however, it wasn’t until the mid-1980s that the use of personal computers in the workplace became an industry standard. This achievement gave rise to the digital monitoring of work, which in turn allowed employers to monitor employees in real-time, effectively replacing the human eye with a more efficient surveillance mechanism. This represents one of the key aspects of modern-day workplace monitoring.<br><br></p>
<p><br></p>

=== 3. Mechanism of path dependence. ===
'''<br>'''When analyzing the historic lineage of work surveillance, it is possible to point out two main path dependencies mechanisms in place.<p>Firstly, it is clear that many of the technological developments that arose through these events have one major goal, which is to increase wealth within a hierarchical ecosystem. Employers resort to several mechanisms to maximize profits and cut inefficiencies at the expense of workers. </p>
<p>In addition, standardization can also be singled out as a key mechanism at play. Once one corporation implements any of these developments (such as the digital monitoring of the workplace), the market incentives would force the rest to do the same in order to be profitable and remain competitive.</p>

=== 4. Politics and distribution. ===
<br>One simple answer to this question would be that capitalists implement and leverage these mechanisms to increase their wealth while the working class remains powerless and unable to withdraw itself from the system in force. <p>However, if we resort to Winner's framework, we see that the technical elements of these technologies are not intrinsically capitalist, since workplace surveillance can also be observed in alternative regimes (including chattel slavery and the Soviet regime). It seems that, rather than being exclusively tied to capitalistic values, the methods in question respond to strong hierarchical structures.</p>
<p>In addition, even under a libertarian framework, structural inequalities are still present, allowing a privileged class to enjoy individual freedoms while segregated groups keep performing low-income jobs ([https://www.metamute.org/editorial/articles/californian-ideology 10]), arguably still subject to surveillance. Thus, not only are these technologies compatible with hierarchical top-down forms of government, but also with structures that allow such hierarchies to form without state intervention.</p>
<p>That said, this form of management seems less compatible with democratic societies, which tend not only to value individual liberties but to turn them into enforceable rights (including workers' privacy, health and integrity), thus establishing boundaries for workplace surveillance. This could explain why studies of American business leaders have found them unhappy with democratic scruples ([https://www.jstor.org/stable/20024652 11]).</p>
<p><br></p>
=== 5. Contingency and alternatives. ===<p><br></p>
<p>Bush writes about the memex, a machine that could extend a person’s memory and that the memory would belong to the person who owned it. As Bush expands, “For mature thought there is no mechanical substitute" ([https://dl.acm.org/doi/pdf/10.1145/227181.227186 12]). His idea was that a machine could take over the repetitive nature of someone’s job, while the person could record their own thoughts for their own use. Bush’s memex was never created, but the idea inspired products like Microsoft’s Productivity Score that were made for employers (the buyers) not for the person whose memory it used (the worker).</p>
<p>This immediately brings to mind Meta's Model Capability Initiative, which was done to train AI agents the company hopes to sell to other employers as well as watch their employees ([https://www.reuters.com/sustainability/boards-policy-regulation/meta-start-capturing-employee-mouse-movements-keystrokes-ai-training-data-2026-04-21/ 13]). If this was Bush’s memex, it would belong to the employee using it, not Meta. </p>

=== 6. Governance implication.<br> ===
Democratic oversight of the use of algorithms as a form of surveillance is essential to ensure that workers’ rights are not overstepped in the name of efficiency. As such we propose: all workers should have a legal right to be consulted and approve a surveillance program before a company installs and uses it. A tracking system should be a group decision, not an inevitable, unilateral one. While the United States has very little protections for workers in terms of surveillance, the EU leads the way with the GDPR, specifically Article 22 ([http://data.europa.eu/eli/reg/2016/679/oj 14]), and the AI Act Article 5 ([http://data.europa.eu/eli/reg/2024/1689/oj 15]). It’s also worth noting that in December 2026, the Platform Work Directive ([https://eur-lex.europa.eu/eli/dir/2024/2831/oj/eng 16]) will come into effect which will ban automated termination.

= [[Platform Dependence in Public Communication (News) - Ceren Isiklar & Rodrigue Kayembe|Platform Dependence in Public Communication (News)]] Ceren Isiklar & Rodrigue Kayembe =

= Centralization of AI Compute and Data =

== Centralization of AI Compute and Data - by Sisi Hammoudi & Juliette Chamot ==
=== Governance issue ===
<span class="s8">Today, several path-dependence phenomena are emerging across advanced computing technologies. This assignment addresses one of them: the </span><span class="s8">centralisation</span><span class="s8"> of AI compute and data. Training and running frontier AI systems now depends on a handful of hyperscale providers, namely Amazon, Microsoft, Google, and Nvidia, which dominates the GPU market. In other words, near monopolies exist throughout the supply chain. This concentration creates a governance problem affecting public and private actors alike, both legally and institutionally. On the institutional side, it prevents public administrations, and states other than the US or China, from reaching sovereign AI capacity. It also raises competition-law concerns, since dominant market positions can prevent the market from delivering the best products at the best prices to consumers. Finally, AI developers, like states, are often forced into contracts with hyperscale providers, creating a structural dependence on them.</span>

<span class="s8"></span>
=== <span class="s8">Historical lineage</span> ===
==== <span class="s8"><span class="s8">1945</span></span> ====
<span class="s8"><span class="s8">Wartime scientific instruments should have been redirected to extend human command over knowledge, helping establish the postwar model of sustained federal funding for computing research (Vannevar Bush's, "As We May Think », 1945).</span></span>

<span class="s8"><span class="s8"></span></span>
==== <span class="s8"><span class="s8">1990s</span></span> ====
<span class="s8"><span class="s8">The privatisation of the NSFNET established the idea that critical digital infrastructure, even though publicly subsidised, could legitimately be privately owned. The EU Commission followed the US's lead, recommending privatisation and deregulation of telecommunications and media as the best way to build the "infobahn" though EU members could instead have chosen nationalisation of the infrastructure (Barbrook & Cameron, "The Californian Ideology," 1996).</span></span>

<span class="s8"><span class="s8"></span></span>
==== <span class="s8"><span class="s8">2006</span></span> ====
<span class="s8"><span class="s8"></span></span><div class="s13">Hyperscale cloud computing turned massive fixed capital investment in data centresinto a rentable service, favouring firms able to invest at scale.</div>
<div class="s13">Same period: Nvidia's proprietary CUDA software, released alongside its GPUs, became the de facto standard for machine learning, locking in a decade of tooling and expertise around one firm's hardware.</div>
<div class="s13">

==== <span class="s8"><span class="s8"><span class="s8"><span class="s8">From 2022</span></span></span></span> ====
<span class="s8"><span class="s8"><span class="s8"><span class="s8"><span class="s8"><span class="s8">Semiconductor export controls and the CHIPS Act turned compute into an explicit instrument of geopolitical strategy.</span></span></span></span></span></span></div>

<span class="s8"></span>

=== <span class="s8">Mechanisms of path dependance</span><span class="s8"></span> ===
<span class="s8"><span class="s8">Several mechanisms drive this </span><span class="s8">centralisation</span><span class="s8">, but path-dependence ones dominate. Because AI compute is a capital-intensive industry, new entrants struggle to compete with </span><span class="s8">hyperscalers</span><span class="s8"> that have already sunk huge sums into infrastructure, a barrier that produces monopoly-like situations. This dynamic, known as increasing returns, generates system-scale economies that harden into a near-irreversibility of investment (David, "Clio and the Economics of QWERTY," 1985). It overlaps with a second mechanism, </span><span class="s8">standardisation</span><span class="s8">, itself a form of technical interrelatedness: researchers' investment in Nvidia's CUDA-based tooling locks the field into that ecosystem regardless of whether better alternatives exist. A further mechanism is legal and contractual entrenchment. For example, Microsoft's stake in OpenAI binds developers to a specific provider by agreement rather than technical necessity.</span></span>=== Politics and distribution ===<p class="s16" style="text-align: justify;"><span class="s8">Technology does not have a history of distributing power equally. Rather, its development has strengthened the interests and authority of powerful groups that already possessed economic and political resource</span><span class="s8">s, and it is yet to be determined whether AI compute and data will follow suit</span><span class="s8">. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Author Langdon Winner illustrates </span><span class="s8">the historical con</span><span class="s8">s</span><span class="s8">traints of technological development</span><span class="s8"> with the case of New York from </span><span class="s8">the </span><span class="s8">1920s to the 1970s. He argued that the city’s roads, bridges, and parks’ physical arrangement contained “explicit and implicit political purpose</span><span class="s8">.</span><span class="s8">”</span><span class="s8"> </span><span class="s8">These infrastructures privileged wealthy and white communities while restricting poor and Black populations’ choice of mobility. Similarly, the development of industrial machinery increased productivity, simultaneously weakening union and non-union worker</span><span class="s8">s’</span><span class="s8"> bargaining power.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">In the twentieth century, f</span><span class="s8">ollowing the Industrial Revolution, a Control Revolution emerged from the increasing need to manage </span><span class="s8">expanding production, transportation, and information flows requiring strengthened organizational bureaucratic control.</span><span class="s8"> This was heavily shaped by the political context of post-World War II warfare.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Therefore, the bearer of the initial cost of technology is the person putting in the capital to erect it</span><span class="s8">,</span><span class="s8"> but ultimately, the person bearing the long-term cost of technology, whose choices diminish</span><span class="s8">,</span><span class="s8"> </span><span class="s8">is</span><span class="s8"> not the rich and privileged, but the everyday person subject to technologies. Forced to adapt, to reshape their professional skills to adapt to an environment trying to make them obsolete.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">AI compute and data </span><span class="s8">are</span><span class="s8"> at risk of creating the same effect. I</span><span class="s8">n their critique of dotcom neoliberalism</span><span class="s8">,</span><span class="s8"> authors Barbrook and Cameron exposed a similar socio-economic tension. The Californian Ideology claims digital technology empowers individuals</span><span class="s8">;</span><span class="s8"> however, perpetual unequal access to it creates </span><span class="s8">an apartheid between </span><span class="s8">“information-rich” and “information-poor” populations.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Finally, the development of AI compute and data depends on expensive and scar</span><span class="s8">c</span><span class="s8">e computing infrastructure and </span><span class="s8">a </span><span class="s8">highly concentrated semiconductor supply chain.</span><span class="s8"> Today, this is one of the most concentrated industries</span><span class="s8">,</span><span class="s8"> with single companies having near</span><span class="s8">-</span><span class="s8">total monopolies over the production of chips necessary </span><span class="s8">for</span><span class="s8"> the functioning of AI. This monopoly model threatens to reinforce the existing economic hierarchy.</span></p>

<span class="s8"></span>=== <span class="s8">Contingency and alternatives</span> ===<p class="s16" style="text-align: justify;"><span class="s8">If a different technological trajectory was plausible, the most apt time for it to have occurred would be during the transition from the Industrial Revolution to </span><span class="s8">the </span><span class="s8">“Control Revolution”. When industrial production accelerated, our society experienced a “crisis of control”, creating opportunities to organize information and communication differently.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Instead of allowing control technologies to </span><span class="s8">be monopolized within businesses and states, our societies could have pursued a more democratic and decentralized system. Barbrook and Cameron provide such an alternative path through their idea of the “electronic agora</span><span class="s8">,</span><span class="s8">” where digital networks could encourage community participation </span><span class="s8">and </span><span class="s8">direct democracy</span><span class="s8">,</span><span class="s8"> where government could actively engage with the people to shape digital infrastructure and expand access.</span><span class="s8"> These alternatives did not actually prevail because corporat</span><span class="s8">e</span><span class="s8"> and state market participants possessed greater resources and authority to shape technological development at the end of the twentieth century. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Inherently, technological development is not predetermined</span><span class="s8">;</span><span class="s8"> it can support either authoritarian or democratic arrangements.</span><span class="s8"> Today, monopolies control chip production needed for AI compute and data, but other technologies like blockchain permit democratic participation.</span></p>

<span class="s8"></span>=== <span class="s8">Governance implication</span> ===<p class="s16" style="text-align: justify;"><span class="s8">History always threatens to repeat itself. As Barbrook discussed, digital societies are shaped by the political context i</span><span class="s8">n</span><span class="s8"> </span><span class="s8">which</span><span class="s8"> </span><span class="s8">they develop</span><span class="s8">. It is a valid concern, then, that the development of information technology, through AI computing, would follow the same trajectory</span><span class="s8">-</span><span class="s8"> that is, be constrained to benefit the wealthy who funded its development, and ultimately harm the middle and lower classes enabling its functioning. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">However, AI’s place in our society has not yet crystallized. Technology governance does not have to adhere to this system. Such a seemingly sustainable approach to AI compute and data governance has already been broached. In 2022, the Notre Dame Journal on Emerging Technologies published an article in </span><span class="s8">its</span><span class="s8"> issue on Artificial Intelligence where the authors proposed an AI compute system based on a “dual investment strategy”</span><span class="s8">. Firstly, the model would be launched by “subsidizing and negotiating” cloud computing for </span><span class="s17">human </span><span class="s8">AI researchers through “existing” cloud vendors</span><span class="s8">,</span><span class="s8"> which would expand on existing AI compute initiatives. Secondly, the compute system would be specifically invested in as a public</span><span class="s8">ly</span><span class="s8"> owned infrastructure that would </span><span class="s8">b</span><span class="s8">e built</span><span class="s8"> and operate similarly to national libraries which own “academic supercomputing facilities”. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">The pilot program would have a “base level of compute access”</span><span class="s8"> which would keep costs of processing </span><span class="s8">down and</span><span class="s8"> enable it to meet the compute need</span><span class="s8">s</span><span class="s8"> of </span><span class="s8">a majority of</span><span class="s8"> academic researchers. Absent any strengthened legislation on the use of AI compute and data, the hardware would be accompanied by an AI ethics framework to enable sustainable technological governance. In addition to an ethical staff, the human resources needed to keep an AI compute model running would include data officers, scientists, data center and software engineers, research developers, </span><span class="s8">and </span><span class="s8">system administrators. These positions require specialized training, allowing new types of professional</span><span class="s8">s</span><span class="s8">, from any background</span><span class="s8">,</span><span class="s8"> to have access to professional prosperity in the field of AI. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">By implementing these frameworks, never neglecting the human element, and the economic and intellectual impact AI has on our society, we move from a historical determinist perspective to one acknowledging historical constraints and </span><span class="s8">learning</span><span class="s8"> from them.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8"></span></p>

=== <span class="s8">Bibliography</span> ===
<span class="s8"></span><div><div><p>[1] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 124, (1980)</p></div>
<div><p>[2] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 124, (1980)</p></div>
<div><p>[3] James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Harvard University Press, p. 443, 450 (1986)</p></div>
<div><p>[4] Richard Barbrook, Andy Cameron, The Californian Ideology, ''Science as Culture'', ''6''(1), 44–72, 61 (1996)</p></div>
<div><p>[5] Anton Korinek, Jai Vipra, Concentrating Intelligence: Scaling and Market Structure in Artificial Intelligence, National Bureau of Economic Research, 10.3386/w33139, p. 11, 13-15 (2024)</p></div></div>
<div><div><p>[6] James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Harvard University Press, p. 443, 442 (1986)</p></div>
<div><p>[7] Richard Barbrook, Andy Cameron, The Californian Ideology, ''Science as Culture'', ''6''(1), 44–72 (1996)</p></div>
<div><p>[8] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 130 (1980)</p></div>
<div><p>[9] Daniel E. Ho, Jennifer King, Russell C. Wald & Christopher Wan, The Centrality of Data and Compute for AI Innovation: A Blueprint for the National Research Cloud, 3 Notre Dame J. on Emerging Tech. 71, 81(2022)</p></div>
<div><p>[10] Daniel E. Ho, Jennifer King, Russell C. Wald & Christopher Wan, The Centrality of Data and Compute for AI Innovation: A Blueprint for the National Research Cloud, 3 Notre Dame J. on Emerging Tech. 71, 82 (2022)</p></div></div>

== Centralization of AI compute and Data - by Atmika Dechamma & Pablo Latorre ==
=== Governance Issue ===
AI computing capacity and data presents the governance problem of an extreme concentration of technological infrastructure and information in a restricted number of companies that privately and globally dominate the market. This affects public and private actors - such as small and medium enterprises, potential new competitors, and Global South countries - that increasingly depend on AI systems outside their control and/or are deterred from competing with established companies because of impenetrable entry barriers - most notably initial investments and operational costs (1). The institutional and legal relevance of this is that it consolidates, reproduces and amplifies asymmetries of power in favor of the “Big Tech”, reverting the technological trend from digital decentralization to physical infrastructure dependency and curbing innovation and market efficiency.

=== Historical Lineage ===
==== Limitation of Machine Learning ====<p style="text-align: justify;">The AI revolution was built on the limitation of human capabilities to process large information and derive efficient results - where before there existed only facilitative tools to allow for human-enabled processes, AI now offered derivations and cognitive reasoning in moments and at large scales.</p>

<br>==== Revert to Physical Infrastructure Dependency in the rush to be the best ====<p style="text-align: justify;">The present AI model heavily relies on hardware in order to perform complex data processing for deep learning. The data centre models followed require larger model sizes (which require higher resource, information and capital investment) to be able to yield better results. Due to the limitation of resources and a rush to establish the best working model of an AI agent, data centres are built on a rudimentary model heavily dependent on massive amounts of electricity and fresh water (2).</p>

<br>==== Regulation and Decentralization of Information ====<p style="text-align: justify;">Regulation and influence of sovereigns and institutions has resulted in data sharing and information exchange structurally through open source and cloud computing. But, while AI models claims their selling point to be accessibility for all, their information and derivations remain in a closed server, stored for the purpose of training their AI agents to be the best (3). Big AI contenders attract higher consumer and information flows overall resulting in a better developed AI model, proving disadvantageous in any market where early contenders have an advantage and accumulation of power.</p>
=== Mechanisms of Path Dependence ===<p style="text-align: justify;">Early on in the industry there is an expectation of human and capital investment from all stakeholders involved, this high cost for entry into the market and expected capital to sustain and develop a model by both consumer and enterprise makes for a high switching cost. For the AI model to run and scale there is a presumption that the rate of growth is dependent on the size of data fed, resulting in a better model and hence, more consumer usage. A closed centralised data loop is created but the requirement for storage and processing far surpasses current infrastructures capabilities - heavily relying on early market entry advantage and sustained capital investments taking away from smaller aspirants and hoarding of power within closed circles. Despite its physical dependency, AI data processing has unique requirements beyond legacy infrastructures (4) - larger demand and production of GPUs and designed to handle immense thermal heat generated while consuming electricity beyond current capabilities. </p>

<br><p style="text-align: justify;">The AI space largely remains unregulated due to its undefined form and fluid nature, however, this has given way for unethical and morality issues regarding data privacy, exploitation of labour and environmental harms.</p>

=== Politics and Distribution ===
Concentration strengthens the values, interests, capacities, and authority of the “Big Tech” over other actors of public and private nature. In this context, Barbrook and Cameron (5) illuminate the ethos behind AI technology development while Winner (6) makes observable the power relations surrounding it: concentration is the consequence of an AI development anchored in the belief that it must be free from all institutional and normative restriction since it is inherently liberating and non-political which, in turn, provides a political/moral justification for the concentration (if technology is inherently liberating and non-political then there is no power relations to observe or question) that translates itself into power dynamics where the AI developer not only owns the AI technology but also decides how it is trained and deployed and which models the user can or cannot access.

=== Contingency and Alternatives ===
Concentration is due, first and foremost, to the fact that AI computing capacity requires massive and cutting-edge physical infrastructure - such as data centers and microprocessors - which, in turn, requires upfront investments that are unaffordable to most companies. In David’s (7) language, companies that could finance the infrastructure and position themselves first obtained self-reinforcing advantages and a “lock-in” effect in the market. In other words, concentration is the consequence of the development of AI physical infrastructure privately with no mixed or public alternatives as a public service. Therefore, there was a space for an alternative future at the end of the 1980s or the beginning of the 1990s when the privatization model of the technological infrastructure was consolidated in line with the “Californian Ideology”. In that moment governments had the chance to treat technological infrastructure as a mixed or public good rendering it universally accessible as it was the case of the Internet; this would have permitted companies of all sizes to profit from economies of scale, reducing entry barriers and fostering competition. However, this alternative path did not prevail since the Californian Ideology provided the moral/political justification to exclude governments from the development of technological infrastructure in the exact moment in which such a decision would produce structural consequences. If that path had been followed it could have contributed to collective access to economies of scale, lower entry barriers, and a more competitive AI market.

=== Governance Implication ===
Considering the above, public or mixed investment on AI physical infrastructure is a necessary governance measure for market decentralization and collective access to AI computing capacity and data. From an historical perspective, the proposed measure is not merely a corrective action for a market failure but for a contingent and ideological decision made between the late 1980s and early 1990s that produced a structural “path dependence” that now is necessary to change.
==== References ====
(1) The Quiet Re-Centralization of Tech: Why AI Is Reversing 20 Years of Platform Decentralization. (2026, January 10). Retrieved on 8 September 2026, from Medium website : [https://medium.com/@sumitbhardwaj1357/the-quiet-re-centralization-of-tech-why-ai-is-reversing-20-years-of-platform-decentralization-ce15962b6a13 https://medium.com/@sumitbhardwaj1357/the-quiet-re-centralization-of-tech-why-ai-is-reversing-20-years-of-platform-decentralization-ce15962b6a13]

(2) [https://www.mdpi.com/1996-1073/19/1/137 Ginzburg-Ganz, E., Lifshits, P., Machlev, R., Belikov, J., Krieger, Z., & Levron, Y. (2025). Technical challenges of AI data center integration into power grids—A survey. ''Energies'', ''19''(1), 137.]

(3) [https://decai-resources.s3.us-east-1.amazonaws.com/decentralized_ML_perspective-16.pdf Singh, A., Lu, C., Gupta, G., Chopra, A., Blanc, J., Klinghoffer, T., ... & Raskar, R. (2024). A perspective on decentralizing AI.]

(4) The Rise of AI Data Center Models and the Decline of the General-Purpose Data Center. (Circa 2026). Retrieved on 8 September 2026, from SubZero Engineering website : [https://www.subzeroeng.com/the-decline-of-the-general-purpose-data-center/ https://www.subzeroeng.com/the-decline-of-the-general-purpose-data-center/]

(5) [https://git.xpub.nl/onebigear/SI_09_reference_library/raw/commit/22bf61e1dcfed35d58fba23808212bd00584be47/Richard_Barbrook_Andy_Cameron_Californian_ideology.pdf Barbrook, R., & Cameron, A. (1996). The Californian Ideology. ''Science as Culture'', ''6''(1), 44-72].

(6) [https://d1wqtxts1xzle7.cloudfront.net/87708502/langdon_winner_do_artifacts_have_politics-libre.pdf?1655588533=&response-content-disposition=inline%3B+filename%3DDo_Artifacts_Have_Politics.pdf&Expires=1788890463&Signature=UBBiMqkip5MfYZbkbLNAN6aE2hyZdKkER-HFcmyxXkwIZuTnMKdOGoao8ddFr-rQ5T55A8NWUOlnZkGYmi66e7XR9h99DgTELMCrh48sXjOb8KYgA48r~SUUymt1IW~R~aOKJzitMIUHcrc6yEGKK1OuaA~mrEIVOXpUjQMYjqOd26Xmk9YJEglTkqaxeR47VCCth5FvozNQIowdfjzeu8aNHnI8V7Yur1IiXY~PGvYRcN6wBtcgpVMHZUtNaI8NjmYbp2p5TBYrPMF9MCLtvpGhPlsCfsvA8Bdm6XulGwRlksZBKWvVUQ8D~M8oLu12V39YaiETIObyJqbE7ctIcQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA Winner, L. (2017). Do artifacts have politics?. In ''Computer ethics''(pp. 177-192). Routledge].

(7) [https://sjbae.pbworks.com/f/David+1985.pdf David, P. A. (1985). Clio and the Economics of QWERTY. ''The American economic review'', ''75''(2), 332-337].

== Centralization of AI compute and Data - by Sergey Abramov and Paul Stein ==<div>
=== The current governance issue ===
<p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>
<br>
=== Historical lineage ===
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p>
<br><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p>
<br><p>Soon after Amazon launched AWS, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p>
<br><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>
<br>
=== Mechanism of path dependence ===
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers<sup>4</sup>.</p>
<br><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>
<br>
=== Politics and distribution ===
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities. The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
<br>
=== Contingency and Alternatives ===
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p>
<br>
=== Governance implication ===
<p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p>

=== References & Notes ===
1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from 'https://epoch.ai/data-insights/hyperscalers-control-most-compute' [online resource]. Accessed 7 Sept 2026.</p>2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026) https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world accessed 7 September 2026.

3. Amazon Web Services (AWS), a cloud computing platform

4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025) https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf accessed 7 September 2026.

5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.

6. Ibid. 123.

7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.

8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025) https://texasroyaltybrokers.com/ai-clusters-by-state/ accessed 7 September 2026.

9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.

10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.

11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026) https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 accessed 7 September 2026.

12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'') https://www.nsf.gov/impacts/internet accessed 7 September 2026.

13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'') https://casrai.org/dictionary/term/national-ai-research-resource-nairr accessed 7 September 2026.

14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.

15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024) https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en accessed 7 September 2026.</div>

Governance by institutions

2026-09-22T19:51:57Z

Sergey Abramov: /* cloud service (Ceren) */

[https://wiki.ivir.nl/index.php?title=Antropic%27s_Fable_5.1_(Kate_Davey) Anthropic's Fable 5.1 (Kate Davey)]

*<span style="font-weight: normal;">[[social-media platform]] (mariona)[[]]</span>
**********
= <span style="font-weight: normal;">[[cloud service (Ceren)]]</span> =
*<span style="font-weight: normal;"> app store</span>
*<span style="font-weight: normal;">digital identity system</span>
*<span style="font-weight: normal;">online-payment system</span>
*<span style="font-weight: normal;">AI model</span>
*<span style="font-weight: normal;">public algorithmic system.</span>
[https://wiki.ivir.nl/index.php?title=Decentralized_digital_currency:_BITCOIN_(Sisi) Decentralized Digital Currency: Bitcoin (Sisi)]

[[Submarine Internet Cables - By Pablo Latorre]]

[[Artificial General Intelligence - By Sebastián Muller]]

[https://wiki.ivir.nl/index.php?title=Social-Media_Governance_in_the_EU_-_by_Sergey_Abramov Social-Media Governance - by Sergey Abramov]

= [[Predicitive Policing - Europol (Atmika)]] =
== [[App Store Governance - Google Play (Paul)]] ==

Social-Media Governance in the EU - by Sergey Abramov

2026-09-22T19:50:12Z

Sergey Abramov: Created page with "====== Social-Media Governance in the EU ====== <p>Social-media platforms in the European Union are governed through overlapping public and private institutions. Four institutions illustrate the resulting distribution of authority and its limitations.</p><p>'''The European Commission''', a public institution, derives its supervisory authority from the Digital Services Act (DSA) article 56(2)-(3). Its governing instruments include binding enforcement decisions and fines u..."

====== Social-Media Governance in the EU ======
<p>Social-media platforms in the European Union are governed through overlapping public and private institutions. Four institutions illustrate the resulting distribution of authority and its limitations.</p><p>'''The European Commission''', a public institution, derives its supervisory authority from the Digital Services Act (DSA) article 56(2)-(3). Its governing instruments include binding enforcement decisions and fines under articles 73-74. It exclusively supervises the additional obligations imposed on designated very large platforms, including systemic-risk assessment and mitigation under articles 34-35. These obligations protect fundamental rights, public safety and democratic processes. Its capacity weakness is informational dependence: assessing algorithmic risks requires substantial technical knowledge held by regulated companies. Information-gathering powers reduce, but cannot automatically eliminate, this imbalance. [1]</p><p>'''The Netherlands Authority for Consumers and Markets (ACM)''' is a public institution designated as the Dutch Digital Services Coordinator under DSA article 49 and Dutch implementing legislation. Its instruments include investigations, compliance orders and fines under article 51, alongside complaint handling under article 53. It protects users’ rights, particularly fair moderation procedures and children’s safety. Its capacity limitation is cross-border dependence: under article 56, supervision generally follows the provider’s main establishment, subject to the Commission’s powers. ACM therefore often depends on other authorities to address Dutch users’ complaints, illustrating Black’s account of regulatory interdependence. [2]</p><p>'''Meta''', a non-state corporate institution operating Facebook and Instagram, derives governing authority from user contracts and control over platform infrastructure. Its instruments are terms of service, Community Standards and moderation decisions. It claims to protect safety while enabling expression. DSA article 14(1) and (4) requires transparent rules and proportionate enforcement attentive to fundamental rights; it constrains rather than creates Meta’s private authority. [3] Its legitimacy weakness is that users have limited influence over rules governing their participation. Klonick’s analysis identifies precisely this tension between sophisticated private governance and weak direct accountability to users. [4]</p><p>'''Appeals Centre Europe''', a non-state dispute-resolution body, derives its recognised role from certification by Ireland’s Coimisiún na Meán under DSA article 21(3). Its instruments are procedural rules and decisions reviewing platform moderation. It seeks to protect fair treatment and access to redress. Its principal capacity weakness is explicit in article 21(2): decisions are non-binding, although platforms must engage in good faith. Independent review therefore does not guarantee implementation. [5]</p><p>These arrangements reflect Lobel’s view of governance through cooperation between public and private actors. [6] Yet shared responsibility does not guarantee equal influence or accountability. Effective governance requires users to understand decisions, challenge them and obtain meaningful remedies. The DSA supports these goals, but their achievement depends on effective oversight and enforcement.</p><p><br></p><p>'''Footnotes'''</p>
#Regulation (EU) 2022/2065 of the European Parliament and of the Council on a Single Market for Digital Services and amending Directive 2000/31/EC (Digital Services Act) [2022] OJ L277/1 (‘DSA’), arts 34–35, 56(2)–(3), 67 and 73-74 <https://eur-lex.europa.eu/eli/reg/2022/2065/oj/eng> accessed 22 September 2026; Julia Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-Regulatory” World’ (2001) 54 Current Legal Problems 107–10 <https://doi.org/10.1093/clp/54.1.103> accessed 22 September 2026.
#DSA (n 2), arts 49, 51, 53 and 56; Netherlands Authority for Consumers and Markets, Digital Services Act: Annual Activity Report 2025 (2026) sections 1.1-1.2 and 2.2 <https://www.acm.nl/en/publications/acm-2025-dsa-annual-report> accessed 22 September 2026; Black (n 1) 109-12.
#DSA (n 2), art 14(1) and (4); Meta, ‘Safety Policies’ <https://www.meta.com/en-gb/safety/topics/safety-basics/policies/> accessed 22 September 2026.
#Kate Klonick, ‘The New Governors: The People, Rules, and Processes Governing Online Speech’ (2018) 131 Harvard Law Review 1598, 1666-69 <https://harvardlawreview.org/print/vol-131/the-new-governors-the-people-rules-and-processes-governing-online-speech/> accessed 22 September 2026.
#DSA (n 2), art 21(2)-(3); Coimisiún na Meán, Information Note re the Decision to Certify the Appeals Centre Europe as an Out-of-Court Dispute Settlement Body pursuant to Article 21 of the Digital Services Act (May 2025) 3-8 <https://www.cnam.ie/app/uploads/2025/05/ACE-Info-Note-final-version-English-for-website.pdf> accessed 22 September 2026.
#Orly Lobel, ‘The Renew Deal: The Fall of Regulation and the Rise of Governance in Contemporary Legal Thought’ (2004) 89 Minnesota Law Review 342, introduction and pt III <https://scholarship.law.umn.edu/mlr/663/> accessed 22 September 2026.
<br><br>

Assignment 1

2026-09-16T08:54:11Z

Sergey Abramov: /* References and Notes */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn<sup>[1]</sup> in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List <sup>[2]</sup>. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power <sup>[3]</sup>. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution <sup>[4]</sup>. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos <sup>[5]</sup>. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis <sup>[6]</sup>.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p>
<p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p>
<p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo <sup>[7]</sup> argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p>
<p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>

====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References and Notes'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 4) 275 </p>
<p>6. Hao (n 4) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10. Bhattacharya (n 8)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Talk:Algorithmic management and workplace surveillance - by Sebastián Muller and Kate Davey''' '''

2026-09-15T20:16:11Z

Sergey Abramov:

<p>Sergey's comments:</p>
<p><nowiki>I agree with the authors that the fact that workers have very few rights to fight back against surveillance programs represents a current governance issue. From my perspective, workplace surveillance technology undermines democratic values, particularly privacy of employees. It represents a form of exploitation of employees by company owners caused by an obvious imbalance of power: on the one hand, employees are in a weaker position, struggling to survive and fulfill their obligations; on the other hand, business owners benefit from the technologies designed to exercise control and collect data about employees’ behaviour.</nowiki> </p>
<p><nowiki>It reflects a contradiction within the Californian Ideology: the virtual class respects democratic values but benefits from exploitation. As Barbrook and Cameron noted in their work The Californian Ideology, the fear of the rebellious 'underclass' has now corrupted the most fundamental tenet of the Californian Ideology: its belief in the emancipatory potentiality of the new information technologies. The technologies of freedom are turning into the machines of dominance.</nowiki></p>
<p><nowiki>I also support the authors’ general idea that democratic oversight of the use of algorithms as a form of surveillance is essential to ensure that workers’ rights are not overstepped in the name of efficiency. The introduction of a tracking system should be a group decision, not an inevitable, unilateral one. This resonates with Barbrook and Cameron’s notion in The Californian Ideology that if only some people have access to the new information technologies, 'Jeffersonian democracy' can become a hi-tech version of the plantation economy of the Old South. Reflecting its deep ambiguity, the Californian Ideology's technological determinism is not simply optimistic and emancipatory. It is simultaneously a deeply pessimistic and repressive vision of the future.</nowiki></p>
<p><nowiki>However, when it comes to the authors’ proposed implementation of an idea under which all workers should have a legal right to be consulted and approve a surveillance program before a company installs and uses it, it seems rather unrealistic in practice. The reason is that employees have little to no choice because corporations implement these control technologies mainly through internal policies, which are usually obligatory for all employees as long as they do not contradict the law. If employees do not agree to being tracked by a surveillance system, their choices are very limited: successful candidates have to either decline an offer of employment after completing multiple rounds of interviews and assignments, or current employees have to decide to leave the company if they do not agree with amendments to its internal policies. Thus, a possible middle-ground governance scenario incorporating elements of democratic oversight might be the mandatory formation of specific workers’ unions representing collective interests of employees. The assessment and protection of the employees’ privacy and other fundamental rights by these special unions or other intermediaries representing interests of all or a part of employees could become a prerequisite for implementing any new internal policy introducing workplace surveillance.</nowiki> </p>

Talk:Algorithmic management and workplace surveillance - by Sebastián Muller and Kate Davey''' '''

2026-09-15T20:13:59Z

Sergey Abramov: Created page with "<p>I agree with the authors that the fact that workers have very few rights to fight back against surveillance programs represents a current governance issue. From my perspective, workplace surveillance technology undermines democratic values, particularly privacy of employees. It represents a form of exploitation of employees by company owners caused by an obvious imbalance of power: on the one hand, employees are in a weaker position, struggling to survive and fulfill..."

<p>I agree with the authors that the fact that workers have very few rights to fight back against surveillance programs represents a current governance issue. From my perspective, workplace surveillance technology undermines democratic values, particularly privacy of employees. It represents a form of exploitation of employees by company owners caused by an obvious imbalance of power: on the one hand, employees are in a weaker position, struggling to survive and fulfill their obligations; on the other hand, business owners benefit from the technologies designed to exercise control and collect data about employees’ behaviour. </p><p>It reflects a contradiction within the Californian Ideology: the virtual class respects democratic values but benefits from exploitation. As Barbrook and Cameron noted in their work The Californian Ideology, the fear of the rebellious 'underclass' has now corrupted the most fundamental tenet of the Californian Ideology: its belief in the emancipatory potentiality of the new information technologies. The technologies of freedom are turning into the machines of dominance.</p><p>I also support the authors’ general idea that democratic oversight of the use of algorithms as a form of surveillance is essential to ensure that workers’ rights are not overstepped in the name of efficiency. The introduction of a tracking system should be a group decision, not an inevitable, unilateral one. This resonates with Barbrook and Cameron’s notion in The Californian Ideology that if only some people have access to the new information technologies, 'Jeffersonian democracy' can become a hi-tech version of the plantation economy of the Old South. Reflecting its deep ambiguity, the Californian Ideology's technological determinism is not simply optimistic and emancipatory. It is simultaneously a deeply pessimistic and repressive vision of the future.</p><p>However, when it comes to the authors’ proposed implementation of an idea under which all workers should have a legal right to be consulted and approve a surveillance program before a company installs and uses it, it seems rather unrealistic in practice. The reason is that employees have little to no choice because corporations implement these control technologies mainly through internal policies, which are usually obligatory for all employees as long as they do not contradict the law. If employees do not agree to being tracked by a surveillance system, their choices are very limited: successful candidates have to either decline an offer of employment after completing multiple rounds of interviews and assignments, or current employees have to decide to leave the company if they do not agree with amendments to its internal policies. Thus, a possible middle-ground governance scenario incorporating elements of democratic oversight might be the mandatory formation of specific workers’ unions representing collective interests of employees. The assessment and protection of the employees’ privacy and other fundamental rights by these special unions or other intermediaries representing interests of all or a part of employees could become a prerequisite for implementing any new internal policy introducing workplace surveillance. </p>

Assignment 1

2026-09-15T12:09:28Z

Sergey Abramov: /* How does this text shifts/alters the context? What is the reaction to the text? */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn<sup>[1]</sup> in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List <sup>[2]</sup>. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power <sup>[3]</sup>. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution <sup>[4]</sup>. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos <sup>[5]</sup>. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis <sup>[6]</sup>.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p>
<p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p>
<p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo <sup>[7]</sup> argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p>
<p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>

====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References and Notes'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Assignment 1

2026-09-15T12:08:22Z

Sergey Abramov: /* What is missing from the perspective? What is a blind spot? What is intentionally left out? */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn<sup>[1]</sup> in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List <sup>[2]</sup>. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power <sup>[3]</sup>. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution <sup>[4]</sup>. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos <sup>[5]</sup>. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis <sup>[6]</sup>.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References and Notes'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Assignment 1

2026-09-15T12:06:42Z

Sergey Abramov: /* What is missing from the perspective? What is a blind spot? What is intentionally left out? */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn<sup>[1]</sup> in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List <sup>[2]</sup>. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power <sup>[3]</sup>. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution <sup>[4]</sup>. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos <sup>[5]</sup>. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References and Notes'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Assignment 1

2026-09-15T12:05:23Z

Sergey Abramov: /* Who is speaking? What is the speakers position, in terms of political, economic, technological power? */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn<sup>[1]</sup> in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List <sup>[2]</sup>. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power <sup>[3]</sup>. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References and Notes'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Assignment 1

2026-09-15T11:12:29Z

Sergey Abramov: /* References */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References and Notes'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Assignment 1

2026-09-15T11:12:00Z

Sergey Abramov: /* References */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. Meta, ‘Introducing Muse Spark: Scaling Towards Personal Superintelligence’ (Meta AI, 8 April 2026) https://ai.meta.com/blog/introducing-muse-spark-msl/ accessed 1 September 2026.</p>
<p>12. Wesley Muller, ‘Meta Data Center Pollution Permit Doesn’t Account for All Its Emissions, Sierra Club Says’ (Louisiana Illuminator, 4 August 2026) https://lailluminator.com/2026/08/04/meta-data-center-pollution-permit-doesnt-account-for-all-its-emissions-sierra-club-says/ accessed 1 September 2026</p>
<p>13. Wesley Muller, ‘Meta Can Keep Its Data Center Details Private, PSC Decides’ (Louisiana Illuminator, 13 August 2026) https://lailluminator.com/2026/08/13/meta-data-center-private/ accessed 1 September 2026. </p>

Assignment 1

2026-09-15T10:41:45Z

Sergey Abramov: /* References */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>
<p>3. Illinois State Treasurer Michael Frerichs and Schroders International Selection Fund, ‘Notice of Exempt Solicitation: Meta Platforms, Inc’ (US Securities and Exchange Commission, 2025) https://www.sec.gov/Archives/edgar/data/1326801/000121465925007147/r57250px14a6g.htm accessed 1 September 2026.</p>
<p>4. Karen Hao, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (Penguin Press 2025) 274–275.</p>
<p>5. Hao (n 5) 275 </p>
<p>6. Hao (n 5) 275 </p>
<p>7. Chinasa T Okolo is the founder and scientific director of the policy incubator Technecultura and the inaugural policy lead of the research organisation Black in AI.</p>
<p>8. Ananya Bhattacharya, ‘“It’s Laughable”: Global AI Experts Challenge Zuckerberg’s “AI for Everyone”’ (Rest of World, 20 August 2026) https://restofworld.org/2026/mark-zuckerberg-meta-ai-for-everyone-manifesto-global-critique/ accessed 1 September 2026.</p>
<p>9. Abeba Birhane is the founder of Trinity College Dublin’s AI Accountability Lab. </p>
<p>10.  Bhattacharya (n 9)</p>
<p>11. </p>

Assignment 1

2026-09-15T10:21:20Z

Sergey Abramov: /* References */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>
<p>2. Forbes, ‘The World’s Real-Time Billionaires’ (Forbes) https://www.forbes.com/real-time-billionaires/ accessed 1 September 2026.</p>

Assignment 1

2026-09-15T10:20:38Z

Sergey Abramov: /* References */

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References'' ======
<p>1. Katherine Hays, ‘Facebook-owner Meta to nearly double AI spending’ (BBC News, 29 January 2026) https://www.bbc.com/news/articles/cn8jkyk78gno accessed 1 September 2026.</p>

Governance by history

2026-09-13T21:32:15Z

Sergey Abramov: Undid revision 123814 by Sergey Abramov (talk)

[[File:Old internet.jpg|300px|thumb]]

= contributions =
*can
*you 
*please
*create
*individual pages
*from the contributions below
*and link them back here?

= links things that popped up during class =
*[https://craphound.com/ Cory Doctorow]
*[https://dougengelbart.org/content/view/209/ The mother of All demos] (first mouse, graphical interface, word processor, etc)
*Can a simple sensor be political? [https://gizmodo.com/why-cant-this-soap-dispenser-identify-dark-skin-1797931773 well, this "racist" soap dispenser made headlines a couple of years ago.]
*Podcast series: [https://www.the-santiago-boys.com/ Morozov: the Santiago Boys.]
*Anatomy on AN AI. [[File:Ai-anatomy-map.pdf|check out the pdf here]]

= Workplace surveillance =
== Information Control and Workplace Surveillance - by Jihae Park and Mariona Fortuny Jané ==

==== <span style="font-size: 12pt;">The Current Governance Issue</span> ====
<span style="font-size: 10pt;">There are numerous current governance problems arising from technological innovations, yet here we will specifically focus on the information control and workplace surveillance. Whether intended or unintended, there were certain historical incidents which resulted in social and political consequences. As noticed from the Control Revolution, the dramatic evolution in mass media and telecommunications technologies has led to the reinforcement of the economic and political control during the Industrial Revolution. ([https://archive.org/details/beniger-james-the-control-revolution Beniger, 1986]) By applying this logic to a modern example, this is clearly how many corporations are getting in control of their employees through surveillance technologies. Along with the emergence of remote and hybrid work after Covid-19, employers are now relying on “bossware” to analyze the emails, chats, and messages of their employees on platforms such as Teams or Slack. ([https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263 Ollier-Malaterre et al., 2026, p.237]) Some companies have strict restrictions on transferring data both internally and externally in protection of confidential information and consistently monitor their workers to the certain extent which could slow down their efficiency. This is problematic as it’s not only violating the privacy of their employees but also depriving employees of rights they should be naturally be able to exercise.</span>

<span style="font-size: 12pt;"></span>

==== <span style="font-size: 12pt;">Historical Lineage</span> ====
<span style="font-size: 12pt;"><span style="font-size: 10pt;">In relation to the current issue of the information control and workplace surveillance, Max Weber’s argument of bureaucracy and rationalization in the middle and late nineteenth century are good historical examples which are still frequently visible nowadays. Weber stated how the range of control could extend not only by expanding the capability to process information but also by limiting the amount of information to be processed. Here, the former is directly related to the control implemented by bureaucratization (in the mid 19th century) and computerization (in the late 20th century), whereas the latter addresses rationalization. An example from bureaucracy is the expansion of paperwork in response to the growth in information. We can still note these traces of bureaucracy from numerous procedures that we have to go through for a decision-making to take place in corporations. He brings up the example of standardized time zones for rationalization which is simply an example for an omission or a breakdown of information to facilitate its processing. ([https://archive.org/details/beniger-james-the-control-revolution Beniger, 1986, p.15-16]) This is also commonly seen from nowadays as some companies may use internal terminologies for the sake of both confidentiality and efficiency.</span></span>

==== <span style="font-size: 12pt;">Mechanism of Path Dependence</span> ====<p>Following along the line with the historical lineage, there are certain mechanisms at work of path dependence for information control and workplace surveillance. Bureaucracy is closely related with organizational routines. Many companies have already settled down with a fixed workflow and decision-making process, which makes it more costly for one to change from its conventional way of working. This means switching costs from an old system to a new one is clearly high considering its time and capital, not necessarily will guarantee increasing returns. Hence, many firms end up maintaining its old strategy, reluctant to make any modifications or changes.<br></p>

==== <span style="font-size: 12pt;">Politics and Distribution</span> ====

As businesses grew too large for managers to oversee everything directly, companies replaced human judgment with formal rule-systems, as we see explained by Beniger's 'intellectual technology' ([https://archive.org/details/beniger-james-the-control-revolution/page/65/mode/2up Beniger, 1986, pp. 17, 21]). Algorithmic management continues this: a call once made by a supervisor's judgment is now a rule built into software that runs without one, extending a logic Beniger traces as far back as to Taylorism and the so called 'assembly line'. Winner ([https://www.jstor.org/stable/20024652 1980]) asks whether such systems must concentrate power this way or were driven there, when other designs were equally possible; nothing required shutting workers out of the decision, so leaving them out was a political choice, and not necessarily a technical necessity. Barbrook and Cameron ([https://doi.org/10.1080/09505439609526455 1996]) explain the reasoning why some can defend this 'domination': they show how 1990s tech culture sold new technologies as freedom even as it tightened control, so the rhetoric could become one of closer oversight as flexibility and autonomy. Together, authority accrues to employers, who in some systems can set and revise these oversight rules unilaterally and face less resistance because the arrangement is framed as freedom; workers bear the cost, losing the informal give-and-take they had with a human supervisor for oversight that is harder to see, question, or contest, while being told it grants them more freedom, not less.

==== <span style="font-size: 12pt;">Contingency and Alternatives</span> ====

A plausible alternative happened on U.S. post WWII: in some workplaces, labor and its pacing was demanded to be set through collective bargaining and enforced jointly, by both management and union stewards monitoring compliance on the line ([https://doi.org/10.1093/icc/dtaf038 Breznitz et al., 2025, p. 211]). So, in this context, workplace surveillance could have developed as a bargained, dual-administered practice rather than a management monopoly. The 1950 [https://en.wikipedia.org/wiki/Reuther%27s_Treaty_of_Detroit Reuther's Treaty of Detroit ] cut that possibility off: unions won guaranteed wages and benefits in exchange for surrendering any general claim over how work was supervised ([https://doi.org/10.1093/icc/dtaf038 Breznitz et al., 2025, pp. 210–211]). Workers depended on wages to survive, and employers controlled the wage relationship. That asymmetry meant employers, given their position of power, were able to arrange things so that, at the very least, workers would have far less capacity to sustain a fight over control itself. Thus, the "trade" wasn't a fair exchange between equals. This reflects which side could afford to hold out on which issue. Hypothetically speaking, had that joint administration model prevailed instead, this could have greatly impacted norms internationally and good practices with worker-union agreed surveillance would have become a widely accepted standard practice.

==== <span style="font-size: 12pt;">Governance Implication</span> ====

An interesting proposition for present-day governance that wants to fill a gap in workplace surveillance is what [https://doi.org/10.1016/j.digbus.2025.100120 Kayas, Ong and Belal (2025)] are proposing: They argue that EU law, through the [https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng GDPR] and its required [https://en.wikipedia.org/wiki/Data_protection_impact_assessment data protection impact assessments (DPIAs)], already offers meaningful protection for workers in terms of lawful high-risk data collection and data processing. Nevertheless, these DPIAs are built around data and privacy, so they leave out several dimensions of monitoring itself: its purpose, how intrusive it is, how visible it is to the people being watched, and its effects on workers' autonomy, trust, and standing relative to management. These authors propose a "surveillance impact assessment" scoring a system against these characteristics directly, before any surveillance system is put in place. Although this mechanism "is constrained by its reliance on existing literature and the lack of empirical validation," it feels like a genuine alternative for two reasons: 1. it targets the actual mechanism at issue, and 2. at the same time it builds upon a mechanism of active participatory approach where employees have a direct say in how surveillance is designed ([https://doi.org/10.1016/j.digbus.2025.100120 Kayas et al., 2025, p. 11]), which, combined with the actual assessment, would put workers' own judgment at the centre of the one decision that affects them. It's important to note that this alternative is proposed as worth piloting, not as a proven fix, and it is fair to ask whether another mandatory assessment stacked on top of DPIAs would function as intended or simply read to workers as more compliance paperwork.

==== References ====
<br>Barbrook, R., & Cameron, A. (1996). The Californian Ideology. ''Science as Culture'', 6(1), 44–72. [https://doi.org/10.1080/09505439609526455 https://doi.org/10.1080/09505439609526455]<br>Beniger, J. R. (1986). ''The Control Revolution: Technological and Economic Origins of the Information Society''. Harvard University Press. [https://archive.org/details/beniger-james-the-control-revolution https://archive.org/details/beniger-james-the-control-revolution] <br>Breznitz, D., Levy, K., Lipartito, K., & Zehavi, A. (2025). An equity-focused research agenda for workplace surveillance. ''Industrial and Corporate Change'', 35(1), 206–221. [https://doi.org/10.1093/icc/dtaf038 https://doi.org/10.1093/icc/dtaf038]<br>Kayas, O. G., Ong, C. E., & Belal, H. M. (2025). From humans to algorithms: A sociotechnical framework of workplace surveillance. ''Digital Business'', 5, 100120. [https://doi.org/10.1016/j.digbus.2025.100120 https://doi.org/10.1016/j.digbus.2025.100120]<br>Ollier-Malaterre, A., Parent-Rocheleau, X., Benkrid, K., & Provost Savard, Y. (2026). Making Sense of Workplace Surveillance: How Workers and Union Representatives Experience, Imagine, and Resist Surveillance Exposure. ''Surveillance & Society'', 24(3), 236–254. [https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263 https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263]<br>Winner, L. (1980). Do Artifacts Have Politics? ''Daedalus'', 109(1), 121–136. [https://www.jstor.org/stable/20024652 https://www.jstor.org/stable/20024652]

<br>
== [[Algorithmic management and workplace surveillance - by Sebastián Muller and Kate Davey''' ''']] ==

=== 1. The current issue. ===
<br><p>Algorithmic management and workplace surveillance can be defined as the use of computer-programmed procedures to coordinate labor input in an organization ([https://joint-research-centre.ec.europa.eu/scientific-activities/employment/algorithmic-management-and-digital-monitoring-work_en 1]). Simply put, it is a way to control workers, and workers generally have no say in how the system that tracks them is created. As Winner says "different people are differently situated and possess unequal degrees of power as well as unequal levels of awareness" ([https://www.jstor.org/stable/20024652?seq=7 2]).</p>
<p>The population affected is workers; both “white collar” and “blue collar” workers are impacted. Whether it’s Amazon tracking warehouse workers’ “Time Off Task” with handheld scanners ([https://www.wired.com/story/amazon-worker-tracking-details-revealed 3]) or Meta employees having their every click tracked ([https://www.techpolicy.press/metas-worker-surveillance-tests-eu-rules-on-ai-and-labor/ 4]) to train AI, employers have the power to watch and control their employees in an unprecedented way.</p>
<p>It matters legally because workers have very few rights to fight back against these surveillance programs; even the The Electronic Communications Privacy Act of 1986 allows exceptions to surveillance of employees ([https://bja.ojp.gov/program/it/privacy-civil-liberties/authorities/statutes/1285 5]).<br></p>
=== <br>2. Historical lineage. ===<p><br>2.1 Panopticon: it is an architectural apparatus developed by Jeremy Bentham in 1791. The design of this building effectively generated the sense of constant monitoring, which in turn motivated the prisoners to behave as if they were being watched at all times. As human labor became commodified, capitalists began implementing measures to ensure the maximization of productivity, including the replication of this design ([https://doi.org/10.14452/MR-070-09-2019-02_2 6]). While unused today, the panopticon is widely considered as both a theoretical and practical inspiration of other forms of surveillance in the workplace.</p>
<p><br></p>
<p>2.2 Scientific management: this theory attempts to analyze and standardize workflows to improve labor productivity. Frederick Winslow Taylor, its main pioneer, used early forms of data analysis during the XX century to decrease walking distances, eliminating unnecessary movements and preventing deliberate slow-walking among workers. ([https://doi.org/10.14452/MR-070-09-2019-02_2 7]). There are many modern forms of data-driven decision-making processes in the workforce. For instance, Amazon warehouses track individual workers' scan rates in real time through handheld scanners, generating data feeds to track and evaluate performance ([https://www.theverge.com/2019/4/25/18516004/amazon-warehouse-fulfillment-centers-productivity-firing-terminations 8]).</p>
<p><br></p>
<p>2.3 PC in the workplace: John Blankenbaker designed the first commercially available personal computer in 1971 ([https://www.computerhistory.org/revolution/personal-computers/17/297 9]), however, it wasn’t until the mid-1980s that the use of personal computers in the workplace became an industry standard. This achievement gave rise to the digital monitoring of work, which in turn allowed employers to monitor employees in real-time, effectively replacing the human eye with a more efficient surveillance mechanism. This represents one of the key aspects of modern-day workplace monitoring.<br><br></p>
<p><br></p>

=== 3. Mechanism of path dependence. ===
'''<br>'''When analyzing the historic lineage of work surveillance, it is possible to point out two main path dependencies mechanisms in place.<p>Firstly, it is clear that many of the technological developments that arose through these events have one major goal, which is to increase wealth within a hierarchical ecosystem. Employers resort to several mechanisms to maximize profits and cut inefficiencies at the expense of workers. </p>
<p>In addition, standardization can also be singled out as a key mechanism at play. Once one corporation implements any of these developments (such as the digital monitoring of the workplace), the market incentives would force the rest to do the same in order to be profitable and remain competitive.</p>

=== 4. Politics and distribution. ===
<br>One simple answer to this question would be that capitalists implement and leverage these mechanisms to increase their wealth while the working class remains powerless and unable to withdraw itself from the system in force. <p>However, if we resort to Winner's framework, we see that the technical elements of these technologies are not intrinsically capitalist, since workplace surveillance can also be observed in alternative regimes (including chattel slavery and the Soviet regime). It seems that, rather than being exclusively tied to capitalistic values, the methods in question respond to strong hierarchical structures.</p>
<p>In addition, even under a libertarian framework, structural inequalities are still present, allowing a privileged class to enjoy individual freedoms while segregated groups keep performing low-income jobs ([https://www.metamute.org/editorial/articles/californian-ideology 10]), arguably still subject to surveillance. Thus, not only are these technologies compatible with hierarchical top-down forms of government, but also with structures that allow such hierarchies to form without state intervention.</p>
<p>That said, this form of management seems less compatible with democratic societies, which tend not only to value individual liberties but to turn them into enforceable rights (including workers' privacy, health and integrity), thus establishing boundaries for workplace surveillance. This could explain why studies of American business leaders have found them unhappy with democratic scruples ([https://www.jstor.org/stable/20024652 11]).</p>
<p><br></p>
=== 5. Contingency and alternatives. ===<p><br></p>
<p>Bush writes about the memex, a machine that could extend a person’s memory and that the memory would belong to the person who owned it. As Bush expands, “For mature thought there is no mechanical substitute" ([https://dl.acm.org/doi/pdf/10.1145/227181.227186 12]). His idea was that a machine could take over the repetitive nature of someone’s job, while the person could record their own thoughts for their own use. Bush’s memex was never created, but the idea inspired products like Microsoft’s Productivity Score that were made for employers (the buyers) not for the person whose memory it used (the worker).</p>
<p>This immediately brings to mind Meta's Model Capability Initiative, which was done to train AI agents the company hopes to sell to other employers as well as watch their employees ([https://www.reuters.com/sustainability/boards-policy-regulation/meta-start-capturing-employee-mouse-movements-keystrokes-ai-training-data-2026-04-21/ 13]). If this was Bush’s memex, it would belong to the employee using it, not Meta. </p>

=== 6. Governance implication.<br> ===
Democratic oversight of the use of algorithms as a form of surveillance is essential to ensure that workers’ rights are not overstepped in the name of efficiency. As such we propose: all workers should have a legal right to be consulted and approve a surveillance program before a company installs and uses it. A tracking system should be a group decision, not an inevitable, unilateral one. While the United States has very little protections for workers in terms of surveillance, the EU leads the way with the GDPR, specifically Article 22 ([http://data.europa.eu/eli/reg/2016/679/oj 14]), and the AI Act Article 5 ([http://data.europa.eu/eli/reg/2024/1689/oj 15]). It’s also worth noting that in December 2026, the Platform Work Directive ([https://eur-lex.europa.eu/eli/dir/2024/2831/oj/eng 16]) will come into effect which will ban automated termination.

= [[Platform Dependence in Public Communication (News)]] Ceren & Rodrigue =

Public communication is the conveyance of a message to a large group of people. This includes public announcements, advertisements in public spaces, social media broadcasts, news delivered in reporting, editorials, opinion pieces, and even political campaigns. Platform dependence is the reliance on a third party platform or digital infrastructure to reach the audience of the public communication, the public. Platform dependence in public communication therefore includes reliance on private companies like Meta (Facebook), state television channels, or telecommunications providers who provide and control access to, and develop the infrastructure of such platforms. For the purpose of this exercise, news will be utilized as the main example of public communication. It is very difficult to discuss news without highlighting the deep entanglement it shares with with information. For our purpose 'information' will have the the following characteristics; it can be learned, processed, archived, and it's worth correlates to its rarity.<sup>[[#fn1|[1]]]</sup> News is the end result of the process of taking this information from one form to another, and transmitting to a wider, typically public audience is what we will anchor our definition of 'news' to be.

Those who are affected by platform dependence in public communication are the communicators and the public alike. News outlets are mainly for-profit organizations. These organizations’ reliance on third party digital platforms grows day-by-day to maintain their popularity and engagement ultimately deciding the revenue generated from advertisers. The public on the other hand are provided catchier and more diluted information, gaining just enough knowledge to grab their attention while not being provided substantial knowledge on matters. Additionally, members of the public who do not have the resources to access such platforms and digital services remain “information poor”. <sup>6</sup> Digital rich, the virtual class of tech-professional and entrepreneurs benefit from the dependence to their platforms.

This is crucial as platform dependence may threaten democracy and free political expression. Platforms, run by private parties, can be run according to their political beliefs. Social media platforms aiming to keep users online utilize algorithms that provide their users with their confirmed preferences, not allowing the conveyance of a truly heterogeneous catalogue of news. The original ideal of the democratic electronic agora is undermined when private corporations control the flow of news and the infrastructure on which they are provided.

=== Historical Lineage ===

Modern form of this can be traced to Renaissance Europe,<sup>[[#fn2|[2]]]</sup> where a single innovation displaced established networks of collecting, processing and sharing news.The church, merchants, students, and diplomats were core participants in this ecosystem.<sup>[[#fn3|[3]]]</sup> The invention of printing in the 15th century was a seismic point in history as it offered new possibilities for the dissemination of information and therefore news. Initially, printing benefitted from the existing networks more than the networks benefitted from printing. This was the case until the networks were transformed in an unrecognisable manner (a trend we will see come to see throughout the history of the news. A clear instance of this is Johann Carolus's ''Relation aller Fürnemmen und gedenckwürdigen Historien'', a handwritten newspaper delivered weekly to the elite class of the time. The mode of production and the means of distribution were constrained by the available technology rather than strategy. The mode of production decreased the cost, time and later revolutionised the network as production could be localised and dependency on transporting was reduced. <br>

One of the first paying customers for press was the Church, selling indulgences. During this era the Church and the State were in a clear symbiotic relationship, and drawing a separation between the two was difficult. Both actors deeply understood Beniger's principle that control depends on information processing and communication working together.<sup>[[#fn4|[4]]] </sup><sup> </sup>By introducing hurdles (such as the requirement of licensing the type of content that could be printed) and benefiting from the industrial might of this innovation, the Church and the state had effectively consolidated power and control. Interestingly, this contrasts what would take place generations later in the computing revolutions, where proximity to the state purposely unmentioned in order to appeal to the masses.<sup>[[#fn5|[5]]]</sup> <sup></sup>

The next evolution of news took place when transmission was once again innovated with the radio. Radio’s reach could be decoupled from the cost, decreasing reliance on aging networks such as the now matured postal system. Control could be focused on content providers  and who would be allowed broadcast rights. This changed the trajectory of news as a broadcaster could deliver news to anyone with a receiver dialed in to the relevant frequency. News outlets now had legitimate, diverse, alternative methods of production and distribution. The television followed not after, however it mostly followed the trajectory of the radio with the caveat that it dominated due to ability to generate commercial revenue through advertisement and optionality it provided.  

=== Mechanism of Path Dependence ===

The public generally partakes in the consumption of content on social media. Hence, with its 3 billion monthly active users for example, Instagram has an installed base for news outlets to depend upon (Instagram Revenue and Usage Statistics (2026), n.d.). The audience of news outlets spends an increasing amount of their time on social media platforms, which results in a habitual entrenchment for these news outlets. In other words, the means of reaching the audience is dependent on the audience’s habits of content consumption. An indirect contractual entrenchment may also be speculated, as news outlets have obligations to their advertisers (who may request the upkeep of views/clicks on the news platforms they pay to advertise in). Cultural entrenchment should not be disregarded: the public expects to receive their news on social media platforms they readily frequent, and many news outlets meet this expectation. Hence, news outlets must partake in building a presence in social media platforms if they wish to compete.

=== Politics and Distribution  ===

Barbook argues that although the California Ideology claims that state power will be decreased and individuals will be empowered through decentralized networks enabled by digital technologies, he demonstrates that public communication is in fact not immune to undemocratic powers. These include private corporations: these corporations may gatekeep in choosing when and how to provide access to their platforms, while finetuning how users navigate their platforms. By enabling access to content that users tend to agree with, those who manage these platforms enable public opinion to shift to extremes. Being “fed” the same narrative and not being challenged to see an opposing perspective ultimately provides fewer choices to the consumer of the information, the general public. Therefore, individual agency is constrained by platform architecture and control, not liberated by it.

=== Point Of No Return ===
In examining the lineage of news and its eventual platform dependency, the diminishing returns of technological advancement for the industry become evident. Revenue has not been a particular focus of this article. The broadcast era was the era in which news outlets earned some of their largest revenues and were at their most independent. It was during this time that news outlets initiated the unbundling of one of their main revenue generators, the classified advertisement. By initiating this [https://stratechery.com/2017/the-great-unbundling/ unbundling] without any afterthought on capitalising on its timely advantages, news outlets entered their relationships with social media platforms in a weakened position. Searching for a saviour and giving the platform the credentials and reputation they needed in the early days in return. Reminiscent of the trend identified earlier. An example of a new's outlet that has navigated this journey with more diligence is [https://business.columbia.edu/insights/ai-transformative-tech/how-new-york-times-company-reinventing-itself-digital-age The New York Times].

=== Governance Implication & Proposition ===

It's clear that complete avoidance of these platforms on the part of news outlet is a Herculean request! New's outlets have flocked to these platforms due to their substantial audience and lock on audience. It's virtually impossible for a single organisation to mount a significant rebellion without self harming. A proposal is to approach the problem more as a herd and invest and cultivate more alternative open platforms for example [https://en.wikipedia.org/wiki/Bluesky Bluesky]  and [https://joinmastodon.org Mastadon]. The benefit that these platforms bring about begin with not only optionality from the current dependencies but governance via protocol.

=== Footnotes ===

# <span id="fn1"></span>Mark Greengrass, Thierry Rentet and Stéphane Gal, ‘The Hinterland of the Newsletter: Handling Information in Space and Time’ in Joad Raymond and Noah Moxham (eds), News Networks in Early Modern Europe (Brill 2016) http://www.jstor.org/stable/10.1163/j.ctt1w8h1ng.34 accessed 7 September 2026; ‘Michel Serres Prononce En Sorbonne La Conférence Inaugurale Du Programme Paris Nouveaux Mondes - La Chancellerie Des Universités de Paris’ (La Chancellerie des Universités de Paris, 1 January 2013) https://www.sorbonne.fr/michel-serres-prononce-en-sorbonne-la-conference-inaugurale-du-programme-paris-nouveaux-mondes/ accessed 7 September 2026. James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986).
# <span id="fn2"></span>Brendan Dooley, Making News in Renaissance Europe (1st edn, Cambridge University Press 2026) https://doi.org/10.1017/9781009358835.
# <span id="fn3"></span>Andrew Pettegree, The Invention of News: How the World Came to Know about Itself (Yale University Press 2014). Helmer Helmers, ‘Public Diplomacy in Early Modern Europe: Towards a New History of News’ (2016) 22 Media History 401 https://doi.org/10.1080/13688804.2016.1174570.
# <span id="fn4"></span>James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986) P24-25
# <span id="fn5"></span>Richard Barbrook and Andy Cameron, The Californian Ideology (1996)<br>
#Richard Barbrook and Andy Cameron, The Californian Ideology (1996)

= Centralization of AI Compute and Data =

== Centralization of AI Compute and Data - by Sisi Hammoudi & Juliette Chamot ==
=== Governance issue ===
<span class="s8">Today, several path-dependence phenomena are emerging across advanced computing technologies. This assignment addresses one of them: the </span><span class="s8">centralisation</span><span class="s8"> of AI compute and data. Training and running frontier AI systems now depends on a handful of hyperscale providers, namely Amazon, Microsoft, Google, and Nvidia, which dominates the GPU market. In other words, near monopolies exist throughout the supply chain. This concentration creates a governance problem affecting public and private actors alike, both legally and institutionally. On the institutional side, it prevents public administrations, and states other than the US or China, from reaching sovereign AI capacity. It also raises competition-law concerns, since dominant market positions can prevent the market from delivering the best products at the best prices to consumers. Finally, AI developers, like states, are often forced into contracts with hyperscale providers, creating a structural dependence on them.</span>

<span class="s8"></span>
=== <span class="s8">Historical lineage</span> ===
==== <span class="s8"><span class="s8">1945</span></span> ====
<span class="s8"><span class="s8">Wartime scientific instruments should have been redirected to extend human command over knowledge, helping establish the postwar model of sustained federal funding for computing research (Vannevar Bush's, "As We May Think », 1945).</span></span>

<span class="s8"><span class="s8"></span></span>
==== <span class="s8"><span class="s8">1990s</span></span> ====
<span class="s8"><span class="s8">The privatisation of the NSFNET established the idea that critical digital infrastructure, even though publicly subsidised, could legitimately be privately owned. The EU Commission followed the US's lead, recommending privatisation and deregulation of telecommunications and media as the best way to build the "infobahn" though EU members could instead have chosen nationalisation of the infrastructure (Barbrook & Cameron, "The Californian Ideology," 1996).</span></span>

<span class="s8"><span class="s8"></span></span>
==== <span class="s8"><span class="s8">2006</span></span> ====
<span class="s8"><span class="s8"></span></span><div class="s13">Hyperscale cloud computing turned massive fixed capital investment in data centresinto a rentable service, favouring firms able to invest at scale.</div>
<div class="s13">Same period: Nvidia's proprietary CUDA software, released alongside its GPUs, became the de facto standard for machine learning, locking in a decade of tooling and expertise around one firm's hardware.</div>
<div class="s13">

==== <span class="s8"><span class="s8"><span class="s8"><span class="s8">From 2022</span></span></span></span> ====
<span class="s8"><span class="s8"><span class="s8"><span class="s8"><span class="s8"><span class="s8">Semiconductor export controls and the CHIPS Act turned compute into an explicit instrument of geopolitical strategy.</span></span></span></span></span></span></div>

<span class="s8"></span>

=== <span class="s8">Mechanisms of path dependance</span><span class="s8"></span> ===
<span class="s8"><span class="s8">Several mechanisms drive this </span><span class="s8">centralisation</span><span class="s8">, but path-dependence ones dominate. Because AI compute is a capital-intensive industry, new entrants struggle to compete with </span><span class="s8">hyperscalers</span><span class="s8"> that have already sunk huge sums into infrastructure, a barrier that produces monopoly-like situations. This dynamic, known as increasing returns, generates system-scale economies that harden into a near-irreversibility of investment (David, "Clio and the Economics of QWERTY," 1985). It overlaps with a second mechanism, </span><span class="s8">standardisation</span><span class="s8">, itself a form of technical interrelatedness: researchers' investment in Nvidia's CUDA-based tooling locks the field into that ecosystem regardless of whether better alternatives exist. A further mechanism is legal and contractual entrenchment. For example, Microsoft's stake in OpenAI binds developers to a specific provider by agreement rather than technical necessity.</span></span>=== Politics and distribution ===<p class="s16" style="text-align: justify;"><span class="s8">Technology does not have a history of distributing power equally. Rather, its development has strengthened the interests and authority of powerful groups that already possessed economic and political resource</span><span class="s8">s, and it is yet to be determined whether AI compute and data will follow suit</span><span class="s8">. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Author Langdon Winner illustrates </span><span class="s8">the historical con</span><span class="s8">s</span><span class="s8">traints of technological development</span><span class="s8"> with the case of New York from </span><span class="s8">the </span><span class="s8">1920s to the 1970s. He argued that the city’s roads, bridges, and parks’ physical arrangement contained “explicit and implicit political purpose</span><span class="s8">.</span><span class="s8">”</span><span class="s8"> </span><span class="s8">These infrastructures privileged wealthy and white communities while restricting poor and Black populations’ choice of mobility. Similarly, the development of industrial machinery increased productivity, simultaneously weakening union and non-union worker</span><span class="s8">s’</span><span class="s8"> bargaining power.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">In the twentieth century, f</span><span class="s8">ollowing the Industrial Revolution, a Control Revolution emerged from the increasing need to manage </span><span class="s8">expanding production, transportation, and information flows requiring strengthened organizational bureaucratic control.</span><span class="s8"> This was heavily shaped by the political context of post-World War II warfare.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Therefore, the bearer of the initial cost of technology is the person putting in the capital to erect it</span><span class="s8">,</span><span class="s8"> but ultimately, the person bearing the long-term cost of technology, whose choices diminish</span><span class="s8">,</span><span class="s8"> </span><span class="s8">is</span><span class="s8"> not the rich and privileged, but the everyday person subject to technologies. Forced to adapt, to reshape their professional skills to adapt to an environment trying to make them obsolete.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">AI compute and data </span><span class="s8">are</span><span class="s8"> at risk of creating the same effect. I</span><span class="s8">n their critique of dotcom neoliberalism</span><span class="s8">,</span><span class="s8"> authors Barbrook and Cameron exposed a similar socio-economic tension. The Californian Ideology claims digital technology empowers individuals</span><span class="s8">;</span><span class="s8"> however, perpetual unequal access to it creates </span><span class="s8">an apartheid between </span><span class="s8">“information-rich” and “information-poor” populations.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Finally, the development of AI compute and data depends on expensive and scar</span><span class="s8">c</span><span class="s8">e computing infrastructure and </span><span class="s8">a </span><span class="s8">highly concentrated semiconductor supply chain.</span><span class="s8"> Today, this is one of the most concentrated industries</span><span class="s8">,</span><span class="s8"> with single companies having near</span><span class="s8">-</span><span class="s8">total monopolies over the production of chips necessary </span><span class="s8">for</span><span class="s8"> the functioning of AI. This monopoly model threatens to reinforce the existing economic hierarchy.</span></p>

<span class="s8"></span>=== <span class="s8">Contingency and alternatives</span> ===<p class="s16" style="text-align: justify;"><span class="s8">If a different technological trajectory was plausible, the most apt time for it to have occurred would be during the transition from the Industrial Revolution to </span><span class="s8">the </span><span class="s8">“Control Revolution”. When industrial production accelerated, our society experienced a “crisis of control”, creating opportunities to organize information and communication differently.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Instead of allowing control technologies to </span><span class="s8">be monopolized within businesses and states, our societies could have pursued a more democratic and decentralized system. Barbrook and Cameron provide such an alternative path through their idea of the “electronic agora</span><span class="s8">,</span><span class="s8">” where digital networks could encourage community participation </span><span class="s8">and </span><span class="s8">direct democracy</span><span class="s8">,</span><span class="s8"> where government could actively engage with the people to shape digital infrastructure and expand access.</span><span class="s8"> These alternatives did not actually prevail because corporat</span><span class="s8">e</span><span class="s8"> and state market participants possessed greater resources and authority to shape technological development at the end of the twentieth century. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Inherently, technological development is not predetermined</span><span class="s8">;</span><span class="s8"> it can support either authoritarian or democratic arrangements.</span><span class="s8"> Today, monopolies control chip production needed for AI compute and data, but other technologies like blockchain permit democratic participation.</span></p>

<span class="s8"></span>=== <span class="s8">Governance implication</span> ===<p class="s16" style="text-align: justify;"><span class="s8">History always threatens to repeat itself. As Barbrook discussed, digital societies are shaped by the political context i</span><span class="s8">n</span><span class="s8"> </span><span class="s8">which</span><span class="s8"> </span><span class="s8">they develop</span><span class="s8">. It is a valid concern, then, that the development of information technology, through AI computing, would follow the same trajectory</span><span class="s8">-</span><span class="s8"> that is, be constrained to benefit the wealthy who funded its development, and ultimately harm the middle and lower classes enabling its functioning. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">However, AI’s place in our society has not yet crystallized. Technology governance does not have to adhere to this system. Such a seemingly sustainable approach to AI compute and data governance has already been broached. In 2022, the Notre Dame Journal on Emerging Technologies published an article in </span><span class="s8">its</span><span class="s8"> issue on Artificial Intelligence where the authors proposed an AI compute system based on a “dual investment strategy”</span><span class="s8">. Firstly, the model would be launched by “subsidizing and negotiating” cloud computing for </span><span class="s17">human </span><span class="s8">AI researchers through “existing” cloud vendors</span><span class="s8">,</span><span class="s8"> which would expand on existing AI compute initiatives. Secondly, the compute system would be specifically invested in as a public</span><span class="s8">ly</span><span class="s8"> owned infrastructure that would </span><span class="s8">b</span><span class="s8">e built</span><span class="s8"> and operate similarly to national libraries which own “academic supercomputing facilities”. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">The pilot program would have a “base level of compute access”</span><span class="s8"> which would keep costs of processing </span><span class="s8">down and</span><span class="s8"> enable it to meet the compute need</span><span class="s8">s</span><span class="s8"> of </span><span class="s8">a majority of</span><span class="s8"> academic researchers. Absent any strengthened legislation on the use of AI compute and data, the hardware would be accompanied by an AI ethics framework to enable sustainable technological governance. In addition to an ethical staff, the human resources needed to keep an AI compute model running would include data officers, scientists, data center and software engineers, research developers, </span><span class="s8">and </span><span class="s8">system administrators. These positions require specialized training, allowing new types of professional</span><span class="s8">s</span><span class="s8">, from any background</span><span class="s8">,</span><span class="s8"> to have access to professional prosperity in the field of AI. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">By implementing these frameworks, never neglecting the human element, and the economic and intellectual impact AI has on our society, we move from a historical determinist perspective to one acknowledging historical constraints and </span><span class="s8">learning</span><span class="s8"> from them.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8"></span></p>

=== <span class="s8">Bibliography</span> ===
<span class="s8"></span><div><div><p>[1] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 124, (1980)</p></div>
<div><p>[2] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 124, (1980)</p></div>
<div><p>[3] James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Harvard University Press, p. 443, 450 (1986)</p></div>
<div><p>[4] Richard Barbrook, Andy Cameron, The Californian Ideology, ''Science as Culture'', ''6''(1), 44–72, 61 (1996)</p></div>
<div><p>[5] Anton Korinek, Jai Vipra, Concentrating Intelligence: Scaling and Market Structure in Artificial Intelligence, National Bureau of Economic Research, 10.3386/w33139, p. 11, 13-15 (2024)</p></div></div>
<div><div><p>[6] James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Harvard University Press, p. 443, 442 (1986)</p></div>
<div><p>[7] Richard Barbrook, Andy Cameron, The Californian Ideology, ''Science as Culture'', ''6''(1), 44–72 (1996)</p></div>
<div><p>[8] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 130 (1980)</p></div>
<div><p>[9] Daniel E. Ho, Jennifer King, Russell C. Wald & Christopher Wan, The Centrality of Data and Compute for AI Innovation: A Blueprint for the National Research Cloud, 3 Notre Dame J. on Emerging Tech. 71, 81(2022)</p></div>
<div><p>[10] Daniel E. Ho, Jennifer King, Russell C. Wald & Christopher Wan, The Centrality of Data and Compute for AI Innovation: A Blueprint for the National Research Cloud, 3 Notre Dame J. on Emerging Tech. 71, 82 (2022)</p></div></div>

== Centralization of AI compute and Data - by Atmika Dechamma & Pablo Latorre ==
=== Governance Issue ===
AI computing capacity and data presents the governance problem of an extreme concentration of technological infrastructure and information in a restricted number of companies that privately and globally dominate the market. This affects public and private actors - such as small and medium enterprises, potential new competitors, and Global South countries - that increasingly depend on AI systems outside their control and/or are deterred from competing with established companies because of impenetrable entry barriers - most notably initial investments and operational costs (1). The institutional and legal relevance of this is that it consolidates, reproduces and amplifies asymmetries of power in favor of the “Big Tech”, reverting the technological trend from digital decentralization to physical infrastructure dependency and curbing innovation and market efficiency.

=== Historical Lineage ===
==== Limitation of Machine Learning ====<p style="text-align: justify;">The AI revolution was built on the limitation of human capabilities to process large information and derive efficient results - where before there existed only facilitative tools to allow for human-enabled processes, AI now offered derivations and cognitive reasoning in moments and at large scales.</p>

<br>==== Revert to Physical Infrastructure Dependency in the rush to be the best ====<p style="text-align: justify;">The present AI model heavily relies on hardware in order to perform complex data processing for deep learning. The data centre models followed require larger model sizes (which require higher resource, information and capital investment) to be able to yield better results. Due to the limitation of resources and a rush to establish the best working model of an AI agent, data centres are built on a rudimentary model heavily dependent on massive amounts of electricity and fresh water (2).</p>

<br>==== Regulation and Decentralization of Information ====<p style="text-align: justify;">Regulation and influence of sovereigns and institutions has resulted in data sharing and information exchange structurally through open source and cloud computing. But, while AI models claims their selling point to be accessibility for all, their information and derivations remain in a closed server, stored for the purpose of training their AI agents to be the best (3). Big AI contenders attract higher consumer and information flows overall resulting in a better developed AI model, proving disadvantageous in any market where early contenders have an advantage and accumulation of power.</p>
=== Mechanisms of Path Dependence ===<p style="text-align: justify;">Early on in the industry there is an expectation of human and capital investment from all stakeholders involved, this high cost for entry into the market and expected capital to sustain and develop a model by both consumer and enterprise makes for a high switching cost. For the AI model to run and scale there is a presumption that the rate of growth is dependent on the size of data fed, resulting in a better model and hence, more consumer usage. A closed centralised data loop is created but the requirement for storage and processing far surpasses current infrastructures capabilities - heavily relying on early market entry advantage and sustained capital investments taking away from smaller aspirants and hoarding of power within closed circles. Despite its physical dependency, AI data processing has unique requirements beyond legacy infrastructures (4) - larger demand and production of GPUs and designed to handle immense thermal heat generated while consuming electricity beyond current capabilities. </p>

<br><p style="text-align: justify;">The AI space largely remains unregulated due to its undefined form and fluid nature, however, this has given way for unethical and morality issues regarding data privacy, exploitation of labour and environmental harms.</p>

=== Politics and Distribution ===
Concentration strengthens the values, interests, capacities, and authority of the “Big Tech” over other actors of public and private nature. In this context, Barbrook and Cameron (5) illuminate the ethos behind AI technology development while Winner (6) makes observable the power relations surrounding it: concentration is the consequence of an AI development anchored in the belief that it must be free from all institutional and normative restriction since it is inherently liberating and non-political which, in turn, provides a political/moral justification for the concentration (if technology is inherently liberating and non-political then there is no power relations to observe or question) that translates itself into power dynamics where the AI developer not only owns the AI technology but also decides how it is trained and deployed and which models the user can or cannot access.

=== Contingency and Alternatives ===
Concentration is due, first and foremost, to the fact that AI computing capacity requires massive and cutting-edge physical infrastructure - such as data centers and microprocessors - which, in turn, requires upfront investments that are unaffordable to most companies. In David’s (7) language, companies that could finance the infrastructure and position themselves first obtained self-reinforcing advantages and a “lock-in” effect in the market. In other words, concentration is the consequence of the development of AI physical infrastructure privately with no mixed or public alternatives as a public service. Therefore, there was a space for an alternative future at the end of the 1980s or the beginning of the 1990s when the privatization model of the technological infrastructure was consolidated in line with the “Californian Ideology”. In that moment governments had the chance to treat technological infrastructure as a mixed or public good rendering it universally accessible as it was the case of the Internet; this would have permitted companies of all sizes to profit from economies of scale, reducing entry barriers and fostering competition. However, this alternative path did not prevail since the Californian Ideology provided the moral/political justification to exclude governments from the development of technological infrastructure in the exact moment in which such a decision would produce structural consequences. If that path had been followed it could have contributed to collective access to economies of scale, lower entry barriers, and a more competitive AI market.

=== Governance Implication ===
Considering the above, public or mixed investment on AI physical infrastructure is a necessary governance measure for market decentralization and collective access to AI computing capacity and data. From an historical perspective, the proposed measure is not merely a corrective action for a market failure but for a contingent and ideological decision made between the late 1980s and early 1990s that produced a structural “path dependence” that now is necessary to change.
==== References ====
(1) The Quiet Re-Centralization of Tech: Why AI Is Reversing 20 Years of Platform Decentralization. (2026, January 10). Retrieved on 8 September 2026, from Medium website : [https://medium.com/@sumitbhardwaj1357/the-quiet-re-centralization-of-tech-why-ai-is-reversing-20-years-of-platform-decentralization-ce15962b6a13 https://medium.com/@sumitbhardwaj1357/the-quiet-re-centralization-of-tech-why-ai-is-reversing-20-years-of-platform-decentralization-ce15962b6a13]

(2) [https://www.mdpi.com/1996-1073/19/1/137 Ginzburg-Ganz, E., Lifshits, P., Machlev, R., Belikov, J., Krieger, Z., & Levron, Y. (2025). Technical challenges of AI data center integration into power grids—A survey. ''Energies'', ''19''(1), 137.]

(3) [https://decai-resources.s3.us-east-1.amazonaws.com/decentralized_ML_perspective-16.pdf Singh, A., Lu, C., Gupta, G., Chopra, A., Blanc, J., Klinghoffer, T., ... & Raskar, R. (2024). A perspective on decentralizing AI.]

(4) The Rise of AI Data Center Models and the Decline of the General-Purpose Data Center. (Circa 2026). Retrieved on 8 September 2026, from SubZero Engineering website : [https://www.subzeroeng.com/the-decline-of-the-general-purpose-data-center/ https://www.subzeroeng.com/the-decline-of-the-general-purpose-data-center/]

(5) [https://git.xpub.nl/onebigear/SI_09_reference_library/raw/commit/22bf61e1dcfed35d58fba23808212bd00584be47/Richard_Barbrook_Andy_Cameron_Californian_ideology.pdf Barbrook, R., & Cameron, A. (1996). The Californian Ideology. ''Science as Culture'', ''6''(1), 44-72].

(6) [https://d1wqtxts1xzle7.cloudfront.net/87708502/langdon_winner_do_artifacts_have_politics-libre.pdf?1655588533=&response-content-disposition=inline%3B+filename%3DDo_Artifacts_Have_Politics.pdf&Expires=1788890463&Signature=UBBiMqkip5MfYZbkbLNAN6aE2hyZdKkER-HFcmyxXkwIZuTnMKdOGoao8ddFr-rQ5T55A8NWUOlnZkGYmi66e7XR9h99DgTELMCrh48sXjOb8KYgA48r~SUUymt1IW~R~aOKJzitMIUHcrc6yEGKK1OuaA~mrEIVOXpUjQMYjqOd26Xmk9YJEglTkqaxeR47VCCth5FvozNQIowdfjzeu8aNHnI8V7Yur1IiXY~PGvYRcN6wBtcgpVMHZUtNaI8NjmYbp2p5TBYrPMF9MCLtvpGhPlsCfsvA8Bdm6XulGwRlksZBKWvVUQ8D~M8oLu12V39YaiETIObyJqbE7ctIcQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA Winner, L. (2017). Do artifacts have politics?. In ''Computer ethics''(pp. 177-192). Routledge].

(7) [https://sjbae.pbworks.com/f/David+1985.pdf David, P. A. (1985). Clio and the Economics of QWERTY. ''The American economic review'', ''75''(2), 332-337].

== Centralization of AI compute and Data - by Sergey Abramov and Paul Stein ==<div>
=== The current governance issue ===
<p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>
<br>
=== Historical lineage ===
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p>
<br><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p>
<br><p>Soon after Amazon launched AWS, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p>
<br><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>
<br>
=== Mechanism of path dependence ===
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers<sup>4</sup>.</p>
<br><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>
<br>
=== Politics and distribution ===
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities. The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
<br>
=== Contingency and Alternatives ===
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p>
<br>
=== Governance implication ===
<p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p>

=== References & Notes ===
1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from 'https://epoch.ai/data-insights/hyperscalers-control-most-compute' [online resource]. Accessed 7 Sept 2026.</p>2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026) https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world accessed 7 September 2026.

3. Amazon Web Services (AWS), a cloud computing platform

4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025) https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf accessed 7 September 2026.

5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.

6. Ibid. 123.

7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.

8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025) https://texasroyaltybrokers.com/ai-clusters-by-state/ accessed 7 September 2026.

9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.

10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.

11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026) https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 accessed 7 September 2026.

12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'') https://www.nsf.gov/impacts/internet accessed 7 September 2026.

13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'') https://casrai.org/dictionary/term/national-ai-research-resource-nairr accessed 7 September 2026.

14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.

15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024) https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en accessed 7 September 2026.</div>

Governance by history

2026-09-13T18:52:57Z

Sergey Abramov: /* contributions */

[[File:Old internet.jpg|300px|thumb]]

= contributions =
*can
*you 
*please
*create
*individual pages
*from the contributions below
*and link them back here?
*'''<span class="mw-page-title-main">[https://wiki.ivir.nl/index.php?title=Centralization_of_AI_compute_and_Data_-_by_Sergey_Abramov_and_Paul_Stein Centralization of AI compute and Data - by Sergey Abramov and Paul Stein]</span>'''

= links things that popped up during class =
*[https://craphound.com/ Cory Doctorow]
*[https://dougengelbart.org/content/view/209/ The mother of All demos] (first mouse, graphical interface, word processor, etc)
*Can a simple sensor be political? [https://gizmodo.com/why-cant-this-soap-dispenser-identify-dark-skin-1797931773 well, this "racist" soap dispenser made headlines a couple of years ago.]
*Podcast series: [https://www.the-santiago-boys.com/ Morozov: the Santiago Boys.]
*Anatomy on AN AI. [[File:Ai-anatomy-map.pdf|check out the pdf here]]

= Workplace surveillance =
== Information Control and Workplace Surveillance - by Jihae Park and Mariona Fortuny Jané ==

==== <span style="font-size: 12pt;">The Current Governance Issue</span> ====
<span style="font-size: 10pt;">There are numerous current governance problems arising from technological innovations, yet here we will specifically focus on the information control and workplace surveillance. Whether intended or unintended, there were certain historical incidents which resulted in social and political consequences. As noticed from the Control Revolution, the dramatic evolution in mass media and telecommunications technologies has led to the reinforcement of the economic and political control during the Industrial Revolution. ([https://archive.org/details/beniger-james-the-control-revolution Beniger, 1986]) By applying this logic to a modern example, this is clearly how many corporations are getting in control of their employees through surveillance technologies. Along with the emergence of remote and hybrid work after Covid-19, employers are now relying on “bossware” to analyze the emails, chats, and messages of their employees on platforms such as Teams or Slack. ([https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263 Ollier-Malaterre et al., 2026, p.237]) Some companies have strict restrictions on transferring data both internally and externally in protection of confidential information and consistently monitor their workers to the certain extent which could slow down their efficiency. This is problematic as it’s not only violating the privacy of their employees but also depriving employees of rights they should be naturally be able to exercise.</span>

<span style="font-size: 12pt;"></span>

==== <span style="font-size: 12pt;">Historical Lineage</span> ====
<span style="font-size: 12pt;"><span style="font-size: 10pt;">In relation to the current issue of the information control and workplace surveillance, Max Weber’s argument of bureaucracy and rationalization in the middle and late nineteenth century are good historical examples which are still frequently visible nowadays. Weber stated how the range of control could extend not only by expanding the capability to process information but also by limiting the amount of information to be processed. Here, the former is directly related to the control implemented by bureaucratization (in the mid 19th century) and computerization (in the late 20th century), whereas the latter addresses rationalization. An example from bureaucracy is the expansion of paperwork in response to the growth in information. We can still note these traces of bureaucracy from numerous procedures that we have to go through for a decision-making to take place in corporations. He brings up the example of standardized time zones for rationalization which is simply an example for an omission or a breakdown of information to facilitate its processing. ([https://archive.org/details/beniger-james-the-control-revolution Beniger, 1986, p.15-16]) This is also commonly seen from nowadays as some companies may use internal terminologies for the sake of both confidentiality and efficiency.</span></span>

==== <span style="font-size: 12pt;">Mechanism of Path Dependence</span> ====<p>Following along the line with the historical lineage, there are certain mechanisms at work of path dependence for information control and workplace surveillance. Bureaucracy is closely related with organizational routines. Many companies have already settled down with a fixed workflow and decision-making process, which makes it more costly for one to change from its conventional way of working. This means switching costs from an old system to a new one is clearly high considering its time and capital, not necessarily will guarantee increasing returns. Hence, many firms end up maintaining its old strategy, reluctant to make any modifications or changes.<br></p>

==== <span style="font-size: 12pt;">Politics and Distribution</span> ====

As businesses grew too large for managers to oversee everything directly, companies replaced human judgment with formal rule-systems, as we see explained by Beniger's 'intellectual technology' ([https://archive.org/details/beniger-james-the-control-revolution/page/65/mode/2up Beniger, 1986, pp. 17, 21]). Algorithmic management continues this: a call once made by a supervisor's judgment is now a rule built into software that runs without one, extending a logic Beniger traces as far back as to Taylorism and the so called 'assembly line'. Winner ([https://www.jstor.org/stable/20024652 1980]) asks whether such systems must concentrate power this way or were driven there, when other designs were equally possible; nothing required shutting workers out of the decision, so leaving them out was a political choice, and not necessarily a technical necessity. Barbrook and Cameron ([https://doi.org/10.1080/09505439609526455 1996]) explain the reasoning why some can defend this 'domination': they show how 1990s tech culture sold new technologies as freedom even as it tightened control, so the rhetoric could become one of closer oversight as flexibility and autonomy. Together, authority accrues to employers, who in some systems can set and revise these oversight rules unilaterally and face less resistance because the arrangement is framed as freedom; workers bear the cost, losing the informal give-and-take they had with a human supervisor for oversight that is harder to see, question, or contest, while being told it grants them more freedom, not less.

==== <span style="font-size: 12pt;">Contingency and Alternatives</span> ====

A plausible alternative happened on U.S. post WWII: in some workplaces, labor and its pacing was demanded to be set through collective bargaining and enforced jointly, by both management and union stewards monitoring compliance on the line ([https://doi.org/10.1093/icc/dtaf038 Breznitz et al., 2025, p. 211]). So, in this context, workplace surveillance could have developed as a bargained, dual-administered practice rather than a management monopoly. The 1950 [https://en.wikipedia.org/wiki/Reuther%27s_Treaty_of_Detroit Reuther's Treaty of Detroit ] cut that possibility off: unions won guaranteed wages and benefits in exchange for surrendering any general claim over how work was supervised ([https://doi.org/10.1093/icc/dtaf038 Breznitz et al., 2025, pp. 210–211]). Workers depended on wages to survive, and employers controlled the wage relationship. That asymmetry meant employers, given their position of power, were able to arrange things so that, at the very least, workers would have far less capacity to sustain a fight over control itself. Thus, the "trade" wasn't a fair exchange between equals. This reflects which side could afford to hold out on which issue. Hypothetically speaking, had that joint administration model prevailed instead, this could have greatly impacted norms internationally and good practices with worker-union agreed surveillance would have become a widely accepted standard practice.

==== <span style="font-size: 12pt;">Governance Implication</span> ====

An interesting proposition for present-day governance that wants to fill a gap in workplace surveillance is what [https://doi.org/10.1016/j.digbus.2025.100120 Kayas, Ong and Belal (2025)] are proposing: They argue that EU law, through the [https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng GDPR] and its required [https://en.wikipedia.org/wiki/Data_protection_impact_assessment data protection impact assessments (DPIAs)], already offers meaningful protection for workers in terms of lawful high-risk data collection and data processing. Nevertheless, these DPIAs are built around data and privacy, so they leave out several dimensions of monitoring itself: its purpose, how intrusive it is, how visible it is to the people being watched, and its effects on workers' autonomy, trust, and standing relative to management. These authors propose a "surveillance impact assessment" scoring a system against these characteristics directly, before any surveillance system is put in place. Although this mechanism "is constrained by its reliance on existing literature and the lack of empirical validation," it feels like a genuine alternative for two reasons: 1. it targets the actual mechanism at issue, and 2. at the same time it builds upon a mechanism of active participatory approach where employees have a direct say in how surveillance is designed ([https://doi.org/10.1016/j.digbus.2025.100120 Kayas et al., 2025, p. 11]), which, combined with the actual assessment, would put workers' own judgment at the centre of the one decision that affects them. It's important to note that this alternative is proposed as worth piloting, not as a proven fix, and it is fair to ask whether another mandatory assessment stacked on top of DPIAs would function as intended or simply read to workers as more compliance paperwork.

==== References ====
<br>Barbrook, R., & Cameron, A. (1996). The Californian Ideology. ''Science as Culture'', 6(1), 44–72. [https://doi.org/10.1080/09505439609526455 https://doi.org/10.1080/09505439609526455]<br>Beniger, J. R. (1986). ''The Control Revolution: Technological and Economic Origins of the Information Society''. Harvard University Press. [https://archive.org/details/beniger-james-the-control-revolution https://archive.org/details/beniger-james-the-control-revolution] <br>Breznitz, D., Levy, K., Lipartito, K., & Zehavi, A. (2025). An equity-focused research agenda for workplace surveillance. ''Industrial and Corporate Change'', 35(1), 206–221. [https://doi.org/10.1093/icc/dtaf038 https://doi.org/10.1093/icc/dtaf038]<br>Kayas, O. G., Ong, C. E., & Belal, H. M. (2025). From humans to algorithms: A sociotechnical framework of workplace surveillance. ''Digital Business'', 5, 100120. [https://doi.org/10.1016/j.digbus.2025.100120 https://doi.org/10.1016/j.digbus.2025.100120]<br>Ollier-Malaterre, A., Parent-Rocheleau, X., Benkrid, K., & Provost Savard, Y. (2026). Making Sense of Workplace Surveillance: How Workers and Union Representatives Experience, Imagine, and Resist Surveillance Exposure. ''Surveillance & Society'', 24(3), 236–254. [https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263 https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/19100/13263]<br>Winner, L. (1980). Do Artifacts Have Politics? ''Daedalus'', 109(1), 121–136. [https://www.jstor.org/stable/20024652 https://www.jstor.org/stable/20024652]

<br>
== [[Algorithmic management and workplace surveillance - by Sebastián Muller and Kate Davey''' ''']] ==

=== 1. The current issue. ===
<br><p>Algorithmic management and workplace surveillance can be defined as the use of computer-programmed procedures to coordinate labor input in an organization ([https://joint-research-centre.ec.europa.eu/scientific-activities/employment/algorithmic-management-and-digital-monitoring-work_en 1]). Simply put, it is a way to control workers, and workers generally have no say in how the system that tracks them is created. As Winner says "different people are differently situated and possess unequal degrees of power as well as unequal levels of awareness" ([https://www.jstor.org/stable/20024652?seq=7 2]).</p>
<p>The population affected is workers; both “white collar” and “blue collar” workers are impacted. Whether it’s Amazon tracking warehouse workers’ “Time Off Task” with handheld scanners ([https://www.wired.com/story/amazon-worker-tracking-details-revealed 3]) or Meta employees having their every click tracked ([https://www.techpolicy.press/metas-worker-surveillance-tests-eu-rules-on-ai-and-labor/ 4]) to train AI, employers have the power to watch and control their employees in an unprecedented way.</p>
<p>It matters legally because workers have very few rights to fight back against these surveillance programs; even the The Electronic Communications Privacy Act of 1986 allows exceptions to surveillance of employees ([https://bja.ojp.gov/program/it/privacy-civil-liberties/authorities/statutes/1285 5]).<br></p>
=== <br>2. Historical lineage. ===<p><br>2.1 Panopticon: it is an architectural apparatus developed by Jeremy Bentham in 1791. The design of this building effectively generated the sense of constant monitoring, which in turn motivated the prisoners to behave as if they were being watched at all times. As human labor became commodified, capitalists began implementing measures to ensure the maximization of productivity, including the replication of this design ([https://doi.org/10.14452/MR-070-09-2019-02_2 6]). While unused today, the panopticon is widely considered as both a theoretical and practical inspiration of other forms of surveillance in the workplace.</p>
<p><br></p>
<p>2.2 Scientific management: this theory attempts to analyze and standardize workflows to improve labor productivity. Frederick Winslow Taylor, its main pioneer, used early forms of data analysis during the XX century to decrease walking distances, eliminating unnecessary movements and preventing deliberate slow-walking among workers. ([https://doi.org/10.14452/MR-070-09-2019-02_2 7]). There are many modern forms of data-driven decision-making processes in the workforce. For instance, Amazon warehouses track individual workers' scan rates in real time through handheld scanners, generating data feeds to track and evaluate performance ([https://www.theverge.com/2019/4/25/18516004/amazon-warehouse-fulfillment-centers-productivity-firing-terminations 8]).</p>
<p><br></p>
<p>2.3 PC in the workplace: John Blankenbaker designed the first commercially available personal computer in 1971 ([https://www.computerhistory.org/revolution/personal-computers/17/297 9]), however, it wasn’t until the mid-1980s that the use of personal computers in the workplace became an industry standard. This achievement gave rise to the digital monitoring of work, which in turn allowed employers to monitor employees in real-time, effectively replacing the human eye with a more efficient surveillance mechanism. This represents one of the key aspects of modern-day workplace monitoring.<br><br></p>
<p><br></p>

=== 3. Mechanism of path dependence. ===
'''<br>'''When analyzing the historic lineage of work surveillance, it is possible to point out two main path dependencies mechanisms in place.<p>Firstly, it is clear that many of the technological developments that arose through these events have one major goal, which is to increase wealth within a hierarchical ecosystem. Employers resort to several mechanisms to maximize profits and cut inefficiencies at the expense of workers. </p>
<p>In addition, standardization can also be singled out as a key mechanism at play. Once one corporation implements any of these developments (such as the digital monitoring of the workplace), the market incentives would force the rest to do the same in order to be profitable and remain competitive.</p>

=== 4. Politics and distribution. ===
<br>One simple answer to this question would be that capitalists implement and leverage these mechanisms to increase their wealth while the working class remains powerless and unable to withdraw itself from the system in force. <p>However, if we resort to Winner's framework, we see that the technical elements of these technologies are not intrinsically capitalist, since workplace surveillance can also be observed in alternative regimes (including chattel slavery and the Soviet regime). It seems that, rather than being exclusively tied to capitalistic values, the methods in question respond to strong hierarchical structures.</p>
<p>In addition, even under a libertarian framework, structural inequalities are still present, allowing a privileged class to enjoy individual freedoms while segregated groups keep performing low-income jobs ([https://www.metamute.org/editorial/articles/californian-ideology 10]), arguably still subject to surveillance. Thus, not only are these technologies compatible with hierarchical top-down forms of government, but also with structures that allow such hierarchies to form without state intervention.</p>
<p>That said, this form of management seems less compatible with democratic societies, which tend not only to value individual liberties but to turn them into enforceable rights (including workers' privacy, health and integrity), thus establishing boundaries for workplace surveillance. This could explain why studies of American business leaders have found them unhappy with democratic scruples ([https://www.jstor.org/stable/20024652 11]).</p>
<p><br></p>
=== 5. Contingency and alternatives. ===<p><br></p>
<p>Bush writes about the memex, a machine that could extend a person’s memory and that the memory would belong to the person who owned it. As Bush expands, “For mature thought there is no mechanical substitute" ([https://dl.acm.org/doi/pdf/10.1145/227181.227186 12]). His idea was that a machine could take over the repetitive nature of someone’s job, while the person could record their own thoughts for their own use. Bush’s memex was never created, but the idea inspired products like Microsoft’s Productivity Score that were made for employers (the buyers) not for the person whose memory it used (the worker).</p>
<p>This immediately brings to mind Meta's Model Capability Initiative, which was done to train AI agents the company hopes to sell to other employers as well as watch their employees ([https://www.reuters.com/sustainability/boards-policy-regulation/meta-start-capturing-employee-mouse-movements-keystrokes-ai-training-data-2026-04-21/ 13]). If this was Bush’s memex, it would belong to the employee using it, not Meta. </p>

=== 6. Governance implication.<br> ===
Democratic oversight of the use of algorithms as a form of surveillance is essential to ensure that workers’ rights are not overstepped in the name of efficiency. As such we propose: all workers should have a legal right to be consulted and approve a surveillance program before a company installs and uses it. A tracking system should be a group decision, not an inevitable, unilateral one. While the United States has very little protections for workers in terms of surveillance, the EU leads the way with the GDPR, specifically Article 22 ([http://data.europa.eu/eli/reg/2016/679/oj 14]), and the AI Act Article 5 ([http://data.europa.eu/eli/reg/2024/1689/oj 15]). It’s also worth noting that in December 2026, the Platform Work Directive ([https://eur-lex.europa.eu/eli/dir/2024/2831/oj/eng 16]) will come into effect which will ban automated termination.

= [[Platform Dependence in Public Communication (News)]] Ceren & Rodrigue =

Public communication is the conveyance of a message to a large group of people. This includes public announcements, advertisements in public spaces, social media broadcasts, news delivered in reporting, editorials, opinion pieces, and even political campaigns. Platform dependence is the reliance on a third party platform or digital infrastructure to reach the audience of the public communication, the public. Platform dependence in public communication therefore includes reliance on private companies like Meta (Facebook), state television channels, or telecommunications providers who provide and control access to, and develop the infrastructure of such platforms. For the purpose of this exercise, news will be utilized as the main example of public communication. It is very difficult to discuss news without highlighting the deep entanglement it shares with with information. For our purpose 'information' will have the the following characteristics; it can be learned, processed, archived, and it's worth correlates to its rarity.<sup>[[#fn1|[1]]]</sup> News is the end result of the process of taking this information from one form to another, and transmitting to a wider, typically public audience is what we will anchor our definition of 'news' to be.

Those who are affected by platform dependence in public communication are the communicators and the public alike. News outlets are mainly for-profit organizations. These organizations’ reliance on third party digital platforms grows day-by-day to maintain their popularity and engagement ultimately deciding the revenue generated from advertisers. The public on the other hand are provided catchier and more diluted information, gaining just enough knowledge to grab their attention while not being provided substantial knowledge on matters. Additionally, members of the public who do not have the resources to access such platforms and digital services remain “information poor”. <sup>6</sup> Digital rich, the virtual class of tech-professional and entrepreneurs benefit from the dependence to their platforms.

This is crucial as platform dependence may threaten democracy and free political expression. Platforms, run by private parties, can be run according to their political beliefs. Social media platforms aiming to keep users online utilize algorithms that provide their users with their confirmed preferences, not allowing the conveyance of a truly heterogeneous catalogue of news. The original ideal of the democratic electronic agora is undermined when private corporations control the flow of news and the infrastructure on which they are provided.

=== Historical Lineage ===

Modern form of this can be traced to Renaissance Europe,<sup>[[#fn2|[2]]]</sup> where a single innovation displaced established networks of collecting, processing and sharing news.The church, merchants, students, and diplomats were core participants in this ecosystem.<sup>[[#fn3|[3]]]</sup> The invention of printing in the 15th century was a seismic point in history as it offered new possibilities for the dissemination of information and therefore news. Initially, printing benefitted from the existing networks more than the networks benefitted from printing. This was the case until the networks were transformed in an unrecognisable manner (a trend we will see come to see throughout the history of the news. A clear instance of this is Johann Carolus's ''Relation aller Fürnemmen und gedenckwürdigen Historien'', a handwritten newspaper delivered weekly to the elite class of the time. The mode of production and the means of distribution were constrained by the available technology rather than strategy. The mode of production decreased the cost, time and later revolutionised the network as production could be localised and dependency on transporting was reduced. <br>

One of the first paying customers for press was the Church, selling indulgences. During this era the Church and the State were in a clear symbiotic relationship, and drawing a separation between the two was difficult. Both actors deeply understood Beniger's principle that control depends on information processing and communication working together.<sup>[[#fn4|[4]]] </sup><sup> </sup>By introducing hurdles (such as the requirement of licensing the type of content that could be printed) and benefiting from the industrial might of this innovation, the Church and the state had effectively consolidated power and control. Interestingly, this contrasts what would take place generations later in the computing revolutions, where proximity to the state purposely unmentioned in order to appeal to the masses.<sup>[[#fn5|[5]]]</sup> <sup></sup>

The next evolution of news took place when transmission was once again innovated with the radio. Radio’s reach could be decoupled from the cost, decreasing reliance on aging networks such as the now matured postal system. Control could be focused on content providers  and who would be allowed broadcast rights. This changed the trajectory of news as a broadcaster could deliver news to anyone with a receiver dialed in to the relevant frequency. News outlets now had legitimate, diverse, alternative methods of production and distribution. The television followed not after, however it mostly followed the trajectory of the radio with the caveat that it dominated due to ability to generate commercial revenue through advertisement and optionality it provided.  

=== Mechanism of Path Dependence ===

The public generally partakes in the consumption of content on social media. Hence, with its 3 billion monthly active users for example, Instagram has an installed base for news outlets to depend upon (Instagram Revenue and Usage Statistics (2026), n.d.). The audience of news outlets spends an increasing amount of their time on social media platforms, which results in a habitual entrenchment for these news outlets. In other words, the means of reaching the audience is dependent on the audience’s habits of content consumption. An indirect contractual entrenchment may also be speculated, as news outlets have obligations to their advertisers (who may request the upkeep of views/clicks on the news platforms they pay to advertise in). Cultural entrenchment should not be disregarded: the public expects to receive their news on social media platforms they readily frequent, and many news outlets meet this expectation. Hence, news outlets must partake in building a presence in social media platforms if they wish to compete.

=== Politics and Distribution  ===

Barbook argues that although the California Ideology claims that state power will be decreased and individuals will be empowered through decentralized networks enabled by digital technologies, he demonstrates that public communication is in fact not immune to undemocratic powers. These include private corporations: these corporations may gatekeep in choosing when and how to provide access to their platforms, while finetuning how users navigate their platforms. By enabling access to content that users tend to agree with, those who manage these platforms enable public opinion to shift to extremes. Being “fed” the same narrative and not being challenged to see an opposing perspective ultimately provides fewer choices to the consumer of the information, the general public. Therefore, individual agency is constrained by platform architecture and control, not liberated by it.

=== Point Of No Return ===
In examining the lineage of news and its eventual platform dependency, the diminishing returns of technological advancement for the industry become evident. Revenue has not been a particular focus of this article. The broadcast era was the era in which news outlets earned some of their largest revenues and were at their most independent. It was during this time that news outlets initiated the unbundling of one of their main revenue generators, the classified advertisement. By initiating this [https://stratechery.com/2017/the-great-unbundling/ unbundling] without any afterthought on capitalising on its timely advantages, news outlets entered their relationships with social media platforms in a weakened position. Searching for a saviour and giving the platform the credentials and reputation they needed in the early days in return. Reminiscent of the trend identified earlier. An example of a new's outlet that has navigated this journey with more diligence is [https://business.columbia.edu/insights/ai-transformative-tech/how-new-york-times-company-reinventing-itself-digital-age The New York Times].

=== Governance Implication & Proposition ===

It's clear that complete avoidance of these platforms on the part of news outlet is a Herculean request! New's outlets have flocked to these platforms due to their substantial audience and lock on audience. It's virtually impossible for a single organisation to mount a significant rebellion without self harming. A proposal is to approach the problem more as a herd and invest and cultivate more alternative open platforms for example [https://en.wikipedia.org/wiki/Bluesky Bluesky]  and [https://joinmastodon.org Mastadon]. The benefit that these platforms bring about begin with not only optionality from the current dependencies but governance via protocol.

=== Footnotes ===

# <span id="fn1"></span>Mark Greengrass, Thierry Rentet and Stéphane Gal, ‘The Hinterland of the Newsletter: Handling Information in Space and Time’ in Joad Raymond and Noah Moxham (eds), News Networks in Early Modern Europe (Brill 2016) http://www.jstor.org/stable/10.1163/j.ctt1w8h1ng.34 accessed 7 September 2026; ‘Michel Serres Prononce En Sorbonne La Conférence Inaugurale Du Programme Paris Nouveaux Mondes - La Chancellerie Des Universités de Paris’ (La Chancellerie des Universités de Paris, 1 January 2013) https://www.sorbonne.fr/michel-serres-prononce-en-sorbonne-la-conference-inaugurale-du-programme-paris-nouveaux-mondes/ accessed 7 September 2026. James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986).
# <span id="fn2"></span>Brendan Dooley, Making News in Renaissance Europe (1st edn, Cambridge University Press 2026) https://doi.org/10.1017/9781009358835.
# <span id="fn3"></span>Andrew Pettegree, The Invention of News: How the World Came to Know about Itself (Yale University Press 2014). Helmer Helmers, ‘Public Diplomacy in Early Modern Europe: Towards a New History of News’ (2016) 22 Media History 401 https://doi.org/10.1080/13688804.2016.1174570.
# <span id="fn4"></span>James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986) P24-25
# <span id="fn5"></span>Richard Barbrook and Andy Cameron, The Californian Ideology (1996)<br>
#Richard Barbrook and Andy Cameron, The Californian Ideology (1996)

= Centralization of AI Compute and Data =

== Centralization of AI Compute and Data - by Sisi Hammoudi & Juliette Chamot ==
=== Governance issue ===
<span class="s8">Today, several path-dependence phenomena are emerging across advanced computing technologies. This assignment addresses one of them: the </span><span class="s8">centralisation</span><span class="s8"> of AI compute and data. Training and running frontier AI systems now depends on a handful of hyperscale providers, namely Amazon, Microsoft, Google, and Nvidia, which dominates the GPU market. In other words, near monopolies exist throughout the supply chain. This concentration creates a governance problem affecting public and private actors alike, both legally and institutionally. On the institutional side, it prevents public administrations, and states other than the US or China, from reaching sovereign AI capacity. It also raises competition-law concerns, since dominant market positions can prevent the market from delivering the best products at the best prices to consumers. Finally, AI developers, like states, are often forced into contracts with hyperscale providers, creating a structural dependence on them.</span>

<span class="s8"></span>
=== <span class="s8">Historical lineage</span> ===
==== <span class="s8"><span class="s8">1945</span></span> ====
<span class="s8"><span class="s8">Wartime scientific instruments should have been redirected to extend human command over knowledge, helping establish the postwar model of sustained federal funding for computing research (Vannevar Bush's, "As We May Think », 1945).</span></span>

<span class="s8"><span class="s8"></span></span>
==== <span class="s8"><span class="s8">1990s</span></span> ====
<span class="s8"><span class="s8">The privatisation of the NSFNET established the idea that critical digital infrastructure, even though publicly subsidised, could legitimately be privately owned. The EU Commission followed the US's lead, recommending privatisation and deregulation of telecommunications and media as the best way to build the "infobahn" though EU members could instead have chosen nationalisation of the infrastructure (Barbrook & Cameron, "The Californian Ideology," 1996).</span></span>

<span class="s8"><span class="s8"></span></span>
==== <span class="s8"><span class="s8">2006</span></span> ====
<span class="s8"><span class="s8"></span></span><div class="s13">Hyperscale cloud computing turned massive fixed capital investment in data centresinto a rentable service, favouring firms able to invest at scale.</div>
<div class="s13">Same period: Nvidia's proprietary CUDA software, released alongside its GPUs, became the de facto standard for machine learning, locking in a decade of tooling and expertise around one firm's hardware.</div>
<div class="s13">

==== <span class="s8"><span class="s8"><span class="s8"><span class="s8">From 2022</span></span></span></span> ====
<span class="s8"><span class="s8"><span class="s8"><span class="s8"><span class="s8"><span class="s8">Semiconductor export controls and the CHIPS Act turned compute into an explicit instrument of geopolitical strategy.</span></span></span></span></span></span></div>

<span class="s8"></span>

=== <span class="s8">Mechanisms of path dependance</span><span class="s8"></span> ===
<span class="s8"><span class="s8">Several mechanisms drive this </span><span class="s8">centralisation</span><span class="s8">, but path-dependence ones dominate. Because AI compute is a capital-intensive industry, new entrants struggle to compete with </span><span class="s8">hyperscalers</span><span class="s8"> that have already sunk huge sums into infrastructure, a barrier that produces monopoly-like situations. This dynamic, known as increasing returns, generates system-scale economies that harden into a near-irreversibility of investment (David, "Clio and the Economics of QWERTY," 1985). It overlaps with a second mechanism, </span><span class="s8">standardisation</span><span class="s8">, itself a form of technical interrelatedness: researchers' investment in Nvidia's CUDA-based tooling locks the field into that ecosystem regardless of whether better alternatives exist. A further mechanism is legal and contractual entrenchment. For example, Microsoft's stake in OpenAI binds developers to a specific provider by agreement rather than technical necessity.</span></span>=== Politics and distribution ===<p class="s16" style="text-align: justify;"><span class="s8">Technology does not have a history of distributing power equally. Rather, its development has strengthened the interests and authority of powerful groups that already possessed economic and political resource</span><span class="s8">s, and it is yet to be determined whether AI compute and data will follow suit</span><span class="s8">. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Author Langdon Winner illustrates </span><span class="s8">the historical con</span><span class="s8">s</span><span class="s8">traints of technological development</span><span class="s8"> with the case of New York from </span><span class="s8">the </span><span class="s8">1920s to the 1970s. He argued that the city’s roads, bridges, and parks’ physical arrangement contained “explicit and implicit political purpose</span><span class="s8">.</span><span class="s8">”</span><span class="s8"> </span><span class="s8">These infrastructures privileged wealthy and white communities while restricting poor and Black populations’ choice of mobility. Similarly, the development of industrial machinery increased productivity, simultaneously weakening union and non-union worker</span><span class="s8">s’</span><span class="s8"> bargaining power.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">In the twentieth century, f</span><span class="s8">ollowing the Industrial Revolution, a Control Revolution emerged from the increasing need to manage </span><span class="s8">expanding production, transportation, and information flows requiring strengthened organizational bureaucratic control.</span><span class="s8"> This was heavily shaped by the political context of post-World War II warfare.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Therefore, the bearer of the initial cost of technology is the person putting in the capital to erect it</span><span class="s8">,</span><span class="s8"> but ultimately, the person bearing the long-term cost of technology, whose choices diminish</span><span class="s8">,</span><span class="s8"> </span><span class="s8">is</span><span class="s8"> not the rich and privileged, but the everyday person subject to technologies. Forced to adapt, to reshape their professional skills to adapt to an environment trying to make them obsolete.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">AI compute and data </span><span class="s8">are</span><span class="s8"> at risk of creating the same effect. I</span><span class="s8">n their critique of dotcom neoliberalism</span><span class="s8">,</span><span class="s8"> authors Barbrook and Cameron exposed a similar socio-economic tension. The Californian Ideology claims digital technology empowers individuals</span><span class="s8">;</span><span class="s8"> however, perpetual unequal access to it creates </span><span class="s8">an apartheid between </span><span class="s8">“information-rich” and “information-poor” populations.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Finally, the development of AI compute and data depends on expensive and scar</span><span class="s8">c</span><span class="s8">e computing infrastructure and </span><span class="s8">a </span><span class="s8">highly concentrated semiconductor supply chain.</span><span class="s8"> Today, this is one of the most concentrated industries</span><span class="s8">,</span><span class="s8"> with single companies having near</span><span class="s8">-</span><span class="s8">total monopolies over the production of chips necessary </span><span class="s8">for</span><span class="s8"> the functioning of AI. This monopoly model threatens to reinforce the existing economic hierarchy.</span></p>

<span class="s8"></span>=== <span class="s8">Contingency and alternatives</span> ===<p class="s16" style="text-align: justify;"><span class="s8">If a different technological trajectory was plausible, the most apt time for it to have occurred would be during the transition from the Industrial Revolution to </span><span class="s8">the </span><span class="s8">“Control Revolution”. When industrial production accelerated, our society experienced a “crisis of control”, creating opportunities to organize information and communication differently.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Instead of allowing control technologies to </span><span class="s8">be monopolized within businesses and states, our societies could have pursued a more democratic and decentralized system. Barbrook and Cameron provide such an alternative path through their idea of the “electronic agora</span><span class="s8">,</span><span class="s8">” where digital networks could encourage community participation </span><span class="s8">and </span><span class="s8">direct democracy</span><span class="s8">,</span><span class="s8"> where government could actively engage with the people to shape digital infrastructure and expand access.</span><span class="s8"> These alternatives did not actually prevail because corporat</span><span class="s8">e</span><span class="s8"> and state market participants possessed greater resources and authority to shape technological development at the end of the twentieth century. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">Inherently, technological development is not predetermined</span><span class="s8">;</span><span class="s8"> it can support either authoritarian or democratic arrangements.</span><span class="s8"> Today, monopolies control chip production needed for AI compute and data, but other technologies like blockchain permit democratic participation.</span></p>

<span class="s8"></span>=== <span class="s8">Governance implication</span> ===<p class="s16" style="text-align: justify;"><span class="s8">History always threatens to repeat itself. As Barbrook discussed, digital societies are shaped by the political context i</span><span class="s8">n</span><span class="s8"> </span><span class="s8">which</span><span class="s8"> </span><span class="s8">they develop</span><span class="s8">. It is a valid concern, then, that the development of information technology, through AI computing, would follow the same trajectory</span><span class="s8">-</span><span class="s8"> that is, be constrained to benefit the wealthy who funded its development, and ultimately harm the middle and lower classes enabling its functioning. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">However, AI’s place in our society has not yet crystallized. Technology governance does not have to adhere to this system. Such a seemingly sustainable approach to AI compute and data governance has already been broached. In 2022, the Notre Dame Journal on Emerging Technologies published an article in </span><span class="s8">its</span><span class="s8"> issue on Artificial Intelligence where the authors proposed an AI compute system based on a “dual investment strategy”</span><span class="s8">. Firstly, the model would be launched by “subsidizing and negotiating” cloud computing for </span><span class="s17">human </span><span class="s8">AI researchers through “existing” cloud vendors</span><span class="s8">,</span><span class="s8"> which would expand on existing AI compute initiatives. Secondly, the compute system would be specifically invested in as a public</span><span class="s8">ly</span><span class="s8"> owned infrastructure that would </span><span class="s8">b</span><span class="s8">e built</span><span class="s8"> and operate similarly to national libraries which own “academic supercomputing facilities”. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">The pilot program would have a “base level of compute access”</span><span class="s8"> which would keep costs of processing </span><span class="s8">down and</span><span class="s8"> enable it to meet the compute need</span><span class="s8">s</span><span class="s8"> of </span><span class="s8">a majority of</span><span class="s8"> academic researchers. Absent any strengthened legislation on the use of AI compute and data, the hardware would be accompanied by an AI ethics framework to enable sustainable technological governance. In addition to an ethical staff, the human resources needed to keep an AI compute model running would include data officers, scientists, data center and software engineers, research developers, </span><span class="s8">and </span><span class="s8">system administrators. These positions require specialized training, allowing new types of professional</span><span class="s8">s</span><span class="s8">, from any background</span><span class="s8">,</span><span class="s8"> to have access to professional prosperity in the field of AI. </span></p>
<p class="s16" style="text-align: justify;"><span class="s8">By implementing these frameworks, never neglecting the human element, and the economic and intellectual impact AI has on our society, we move from a historical determinist perspective to one acknowledging historical constraints and </span><span class="s8">learning</span><span class="s8"> from them.</span></p>
<p class="s16" style="text-align: justify;"><span class="s8"></span></p>

=== <span class="s8">Bibliography</span> ===
<span class="s8"></span><div><div><p>[1] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 124, (1980)</p></div>
<div><p>[2] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 124, (1980)</p></div>
<div><p>[3] James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Harvard University Press, p. 443, 450 (1986)</p></div>
<div><p>[4] Richard Barbrook, Andy Cameron, The Californian Ideology, ''Science as Culture'', ''6''(1), 44–72, 61 (1996)</p></div>
<div><p>[5] Anton Korinek, Jai Vipra, Concentrating Intelligence: Scaling and Market Structure in Artificial Intelligence, National Bureau of Economic Research, 10.3386/w33139, p. 11, 13-15 (2024)</p></div></div>
<div><div><p>[6] James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Harvard University Press, p. 443, 442 (1986)</p></div>
<div><p>[7] Richard Barbrook, Andy Cameron, The Californian Ideology, ''Science as Culture'', ''6''(1), 44–72 (1996)</p></div>
<div><p>[8] Langdon Winner, Do Artifacts Have Politics, Daedalus, Winter, 1980, Vol. 109, No. 1, Modern Technology: Problem or Opportunity? pp. 121-136, 130 (1980)</p></div>
<div><p>[9] Daniel E. Ho, Jennifer King, Russell C. Wald & Christopher Wan, The Centrality of Data and Compute for AI Innovation: A Blueprint for the National Research Cloud, 3 Notre Dame J. on Emerging Tech. 71, 81(2022)</p></div>
<div><p>[10] Daniel E. Ho, Jennifer King, Russell C. Wald & Christopher Wan, The Centrality of Data and Compute for AI Innovation: A Blueprint for the National Research Cloud, 3 Notre Dame J. on Emerging Tech. 71, 82 (2022)</p></div></div>

== Centralization of AI compute and Data - by Atmika Dechamma & Pablo Latorre ==
=== Governance Issue ===
AI computing capacity and data presents the governance problem of an extreme concentration of technological infrastructure and information in a restricted number of companies that privately and globally dominate the market. This affects public and private actors - such as small and medium enterprises, potential new competitors, and Global South countries - that increasingly depend on AI systems outside their control and/or are deterred from competing with established companies because of impenetrable entry barriers - most notably initial investments and operational costs (1). The institutional and legal relevance of this is that it consolidates, reproduces and amplifies asymmetries of power in favor of the “Big Tech”, reverting the technological trend from digital decentralization to physical infrastructure dependency and curbing innovation and market efficiency.

=== Historical Lineage ===
==== Limitation of Machine Learning ====<p style="text-align: justify;">The AI revolution was built on the limitation of human capabilities to process large information and derive efficient results - where before there existed only facilitative tools to allow for human-enabled processes, AI now offered derivations and cognitive reasoning in moments and at large scales.</p>

<br>==== Revert to Physical Infrastructure Dependency in the rush to be the best ====<p style="text-align: justify;">The present AI model heavily relies on hardware in order to perform complex data processing for deep learning. The data centre models followed require larger model sizes (which require higher resource, information and capital investment) to be able to yield better results. Due to the limitation of resources and a rush to establish the best working model of an AI agent, data centres are built on a rudimentary model heavily dependent on massive amounts of electricity and fresh water (2).</p>

<br>==== Regulation and Decentralization of Information ====<p style="text-align: justify;">Regulation and influence of sovereigns and institutions has resulted in data sharing and information exchange structurally through open source and cloud computing. But, while AI models claims their selling point to be accessibility for all, their information and derivations remain in a closed server, stored for the purpose of training their AI agents to be the best (3). Big AI contenders attract higher consumer and information flows overall resulting in a better developed AI model, proving disadvantageous in any market where early contenders have an advantage and accumulation of power.</p>
=== Mechanisms of Path Dependence ===<p style="text-align: justify;">Early on in the industry there is an expectation of human and capital investment from all stakeholders involved, this high cost for entry into the market and expected capital to sustain and develop a model by both consumer and enterprise makes for a high switching cost. For the AI model to run and scale there is a presumption that the rate of growth is dependent on the size of data fed, resulting in a better model and hence, more consumer usage. A closed centralised data loop is created but the requirement for storage and processing far surpasses current infrastructures capabilities - heavily relying on early market entry advantage and sustained capital investments taking away from smaller aspirants and hoarding of power within closed circles. Despite its physical dependency, AI data processing has unique requirements beyond legacy infrastructures (4) - larger demand and production of GPUs and designed to handle immense thermal heat generated while consuming electricity beyond current capabilities. </p>

<br><p style="text-align: justify;">The AI space largely remains unregulated due to its undefined form and fluid nature, however, this has given way for unethical and morality issues regarding data privacy, exploitation of labour and environmental harms.</p>

=== Politics and Distribution ===
Concentration strengthens the values, interests, capacities, and authority of the “Big Tech” over other actors of public and private nature. In this context, Barbrook and Cameron (5) illuminate the ethos behind AI technology development while Winner (6) makes observable the power relations surrounding it: concentration is the consequence of an AI development anchored in the belief that it must be free from all institutional and normative restriction since it is inherently liberating and non-political which, in turn, provides a political/moral justification for the concentration (if technology is inherently liberating and non-political then there is no power relations to observe or question) that translates itself into power dynamics where the AI developer not only owns the AI technology but also decides how it is trained and deployed and which models the user can or cannot access.

=== Contingency and Alternatives ===
Concentration is due, first and foremost, to the fact that AI computing capacity requires massive and cutting-edge physical infrastructure - such as data centers and microprocessors - which, in turn, requires upfront investments that are unaffordable to most companies. In David’s (7) language, companies that could finance the infrastructure and position themselves first obtained self-reinforcing advantages and a “lock-in” effect in the market. In other words, concentration is the consequence of the development of AI physical infrastructure privately with no mixed or public alternatives as a public service. Therefore, there was a space for an alternative future at the end of the 1980s or the beginning of the 1990s when the privatization model of the technological infrastructure was consolidated in line with the “Californian Ideology”. In that moment governments had the chance to treat technological infrastructure as a mixed or public good rendering it universally accessible as it was the case of the Internet; this would have permitted companies of all sizes to profit from economies of scale, reducing entry barriers and fostering competition. However, this alternative path did not prevail since the Californian Ideology provided the moral/political justification to exclude governments from the development of technological infrastructure in the exact moment in which such a decision would produce structural consequences. If that path had been followed it could have contributed to collective access to economies of scale, lower entry barriers, and a more competitive AI market.

=== Governance Implication ===
Considering the above, public or mixed investment on AI physical infrastructure is a necessary governance measure for market decentralization and collective access to AI computing capacity and data. From an historical perspective, the proposed measure is not merely a corrective action for a market failure but for a contingent and ideological decision made between the late 1980s and early 1990s that produced a structural “path dependence” that now is necessary to change.
==== References ====
(1) The Quiet Re-Centralization of Tech: Why AI Is Reversing 20 Years of Platform Decentralization. (2026, January 10). Retrieved on 8 September 2026, from Medium website : [https://medium.com/@sumitbhardwaj1357/the-quiet-re-centralization-of-tech-why-ai-is-reversing-20-years-of-platform-decentralization-ce15962b6a13 https://medium.com/@sumitbhardwaj1357/the-quiet-re-centralization-of-tech-why-ai-is-reversing-20-years-of-platform-decentralization-ce15962b6a13]

(2) [https://www.mdpi.com/1996-1073/19/1/137 Ginzburg-Ganz, E., Lifshits, P., Machlev, R., Belikov, J., Krieger, Z., & Levron, Y. (2025). Technical challenges of AI data center integration into power grids—A survey. ''Energies'', ''19''(1), 137.]

(3) [https://decai-resources.s3.us-east-1.amazonaws.com/decentralized_ML_perspective-16.pdf Singh, A., Lu, C., Gupta, G., Chopra, A., Blanc, J., Klinghoffer, T., ... & Raskar, R. (2024). A perspective on decentralizing AI.]

(4) The Rise of AI Data Center Models and the Decline of the General-Purpose Data Center. (Circa 2026). Retrieved on 8 September 2026, from SubZero Engineering website : [https://www.subzeroeng.com/the-decline-of-the-general-purpose-data-center/ https://www.subzeroeng.com/the-decline-of-the-general-purpose-data-center/]

(5) [https://git.xpub.nl/onebigear/SI_09_reference_library/raw/commit/22bf61e1dcfed35d58fba23808212bd00584be47/Richard_Barbrook_Andy_Cameron_Californian_ideology.pdf Barbrook, R., & Cameron, A. (1996). The Californian Ideology. ''Science as Culture'', ''6''(1), 44-72].

(6) [https://d1wqtxts1xzle7.cloudfront.net/87708502/langdon_winner_do_artifacts_have_politics-libre.pdf?1655588533=&response-content-disposition=inline%3B+filename%3DDo_Artifacts_Have_Politics.pdf&Expires=1788890463&Signature=UBBiMqkip5MfYZbkbLNAN6aE2hyZdKkER-HFcmyxXkwIZuTnMKdOGoao8ddFr-rQ5T55A8NWUOlnZkGYmi66e7XR9h99DgTELMCrh48sXjOb8KYgA48r~SUUymt1IW~R~aOKJzitMIUHcrc6yEGKK1OuaA~mrEIVOXpUjQMYjqOd26Xmk9YJEglTkqaxeR47VCCth5FvozNQIowdfjzeu8aNHnI8V7Yur1IiXY~PGvYRcN6wBtcgpVMHZUtNaI8NjmYbp2p5TBYrPMF9MCLtvpGhPlsCfsvA8Bdm6XulGwRlksZBKWvVUQ8D~M8oLu12V39YaiETIObyJqbE7ctIcQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA Winner, L. (2017). Do artifacts have politics?. In ''Computer ethics''(pp. 177-192). Routledge].

(7) [https://sjbae.pbworks.com/f/David+1985.pdf David, P. A. (1985). Clio and the Economics of QWERTY. ''The American economic review'', ''75''(2), 332-337].

== Centralization of AI compute and Data - by Sergey Abramov and Paul Stein ==<div>
=== The current governance issue ===
<p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>
<br>
=== Historical lineage ===
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p>
<br><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p>
<br><p>Soon after Amazon launched AWS, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p>
<br><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>
<br>
=== Mechanism of path dependence ===
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers<sup>4</sup>.</p>
<br><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>
<br>
=== Politics and distribution ===
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities. The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
<br>
=== Contingency and Alternatives ===
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p>
<br>
=== Governance implication ===
<p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p>

=== References & Notes ===
1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from 'https://epoch.ai/data-insights/hyperscalers-control-most-compute' [online resource]. Accessed 7 Sept 2026.</p>2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026) https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world accessed 7 September 2026.

3. Amazon Web Services (AWS), a cloud computing platform

4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025) https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf accessed 7 September 2026.

5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.

6. Ibid. 123.

7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.

8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025) https://texasroyaltybrokers.com/ai-clusters-by-state/ accessed 7 September 2026.

9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.

10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.

11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026) https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 accessed 7 September 2026.

12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'') https://www.nsf.gov/impacts/internet accessed 7 September 2026.

13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'') https://casrai.org/dictionary/term/national-ai-research-resource-nairr accessed 7 September 2026.

14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.

15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024) https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en accessed 7 September 2026.</div>

Centralization of AI compute and Data - by Sergey Abramov and Paul Stein

2026-09-13T18:49:57Z

Sergey Abramov: Created page with "<div class="mw-heading mw-heading3"> ====== <span style="font-size: 12pt;" >The current governance issue</span> ====== </div><p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastr..."

<div class="mw-heading mw-heading3">
====== <span style="font-size: 12pt;" >The current governance issue</span> ======
</div><p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>
====== <span style="color: inherit; font-family: inherit; font-size: 12pt; font-style: inherit; font-variant-ligatures: inherit; font-variant-caps: inherit; font-weight: inherit;" >Historical lineage</span> ======
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p><p>Soon after Amazon launched AWS, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>
====== <span style="color: inherit; font-family: inherit; font-size: 12pt; font-style: inherit; font-variant-ligatures: inherit; font-variant-caps: inherit; font-weight: inherit;" >Mechanism of path dependence</span> ======
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers<sup>4</sup>.</p><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>
====== <span style="color: inherit; font-family: inherit; font-size: 12pt; font-style: inherit; font-variant-ligatures: inherit; font-variant-caps: inherit; font-weight: inherit;" >Politics and distribution</span> ======
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities. The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
====== <span style="color: inherit; font-family: inherit; font-size: 12pt; font-style: inherit; font-variant-ligatures: inherit; font-variant-caps: inherit; font-weight: inherit;" >Contingency and Alternatives</span> ======
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p><div class="mw-heading mw-heading3">
====== <span style="font-size: 12pt;" >Governance implication</span> ======
</div><p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p><div class="mw-heading mw-heading3">
====== <span style="font-size: 12pt;" >References & Notes</span> ======
</div>

1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from '</p>[https://epoch.ai/data-insights/hyperscalers-control-most-compute' https://epoch.ai/data-insights/hyperscalers-control-most-compute']<p>[online resource]. Accessed 7 Sept 2026.</p>

2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026)[https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world]

accessed 7 September 2026.<p>3. Amazon Web Services (AWS), a cloud computing platform</p><p>4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025)</p>[https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf]<p>accessed 7 September 2026.</p><p>5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.</p><p>6. Ibid. 123.</p><p>7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.</p><p>8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025)</p>[https://texasroyaltybrokers.com/ai-clusters-by-state/ https://texasroyaltybrokers.com/ai-clusters-by-state/]<p>accessed 7 September 2026.</p><p>9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.</p><p>10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.</p><p>11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026)</p>[https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026]<p>accessed 7 September 2026.</p><p>12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'')</p>[https://www.nsf.gov/impacts/internet https://www.nsf.gov/impacts/internet]<p>accessed 7 September 2026.</p><p>13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'')</p>[https://casrai.org/dictionary/term/national-ai-research-resource-nairr https://casrai.org/dictionary/term/national-ai-research-resource-nairr]<p>accessed 7 September 2026.</p><p>14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.</p><p>15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.</p>

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024)[https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en]

accessed 7 September 2026.

Assignment 1

2026-09-12T16:40:26Z

Sergey Abramov:

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

====== ''Who is speaking? What is the speakers position, in terms of political, economic, technological power?'' ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== ''How that position shapes what they speak about? How it shapes their perspective?'' ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== ''What is missing from the perspective? What is a blind spot? What is intentionally left out?'' ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

====== ''Who is the intended audience of the text? Why is the text speaking to that particular audience?'' ======
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
====== ''What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?'' ======
<p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
====== ''What is the wider context of the text? What are the conversations/events/in which this text is situated?'' ======
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p><p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p><p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p><p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p><p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
====== ''How does this text shifts/alters the context? What is the reaction to the text?'' ======
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p><p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p><p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p><p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
====== ''What is your reaction/analysis?'' ======
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p><p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p><p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p><p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p><p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p><p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
====== ''References'' ======
<p><br></p>

Assignment 1

2026-09-12T16:33:14Z

Sergey Abramov:

'''The article [https://about.fb.com/news/2026/08/the-future-is-for-everyone/ The Future is for Everyone] review by Sergey Abramov (assignment 1) '''

<br><p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

<br><p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

<br><p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

Assignment 1

2026-09-12T16:31:35Z

Sergey Abramov:

'''The article “[[https://about.fb.com/news/2026/08/the-future-is-for-everyone/|The Future is for Everyone]]” review by Sergey Abramov (assignment 1) '''

<br><p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p><p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p><p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p><p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p><p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

<br><p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p><p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p><p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

<br><p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p><p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p><p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p><p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p><p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p><p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p><p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p><p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p><p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p><p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p><p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p><p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p><p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p><p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p><p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

Assignment 1

2026-09-12T16:29:07Z

Sergey Abramov:

====== The article “The Future is for Everyone” review by Sergey Abramov (assignment 1) ======
====== <span id="HTMLENTITY:1789312265152" style="font-family: <span class=" title=""" contenteditable="false" draggable="true" mwt-noneditable="" mwt-wikimagic="" mwt-placeholder="" mwt-htmlentity="" span="">times new roman", times;" class="mwt-preserveHtml">''Who is speaking? What is the speakers position, in terms of political, economic, technological power?''</span> ======<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>
<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>
====== <span id="HTMLENTITY:1789324340622" style="font-family: <span class=" title=""" contenteditable="false" draggable="true" mwt-noneditable="" mwt-wikimagic="" mwt-placeholder="" mwt-htmlentity="" span="">times new roman", times;" class="mwt-preserveHtml">''How that position shapes what they speak about? How it shapes their perspective?''</span> ======<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

====== <span id="HTMLENTITY:1789277026851" style="font-family: <span class=" title=""" contenteditable="false" draggable="true" mwt-noneditable="" mwt-wikimagic="" mwt-placeholder="" mwt-htmlentity="" span="">times new roman", times;" class="mwt-preserveHtml">''What is missing from the perspective? What is a blind spot? What is intentionally left out?''</span> ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

Assignment 1

2026-09-12T16:28:23Z

Sergey Abramov:

====== The article “The Future is for Everyone” review by Sergey Abramov (assignment 1) ======
====== <span style="font-family: "times new roman", times;">''Who is speaking? What is the speakers position, in terms of political, economic, technological power?''</span> ======<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p><p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p><p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p><p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p><p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>

====== <span style="font-family: "times new roman", times;">''How that position shapes what they speak about? How it shapes their perspective?''</span> ======<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p><p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p><p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>
====== <span style="font-family: "times new roman", times;" >''What is missing from the perspective? What is a blind spot? What is intentionally left out?''</span> ======
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p><p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p><p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p><p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p><p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p><p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p><p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p><p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p><p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p><p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p><p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p><p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p><p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p><p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p><p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>

Assignment 1

2026-09-12T16:25:52Z

Sergey Abramov:

<h6 class="mwt-heading" >The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</h6>

====== <span style="font-family: "times new roman", times;" >''Who is speaking? What is the speakers position, in terms of political, economic, technological power?''</span> ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p><p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p><p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p><p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p><p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>
====== <span style="font-family: "times new roman", times;" >''How that position shapes what they speak about? How it shapes their perspective?''</span> ======
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p><p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p><p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>

Assignment 1

2026-09-12T16:23:33Z

Sergey Abramov: Undid revision 123805 by Sergey Abramov (talk)

<p>The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</p>

<br>

Assignment 1

2026-09-12T16:22:57Z

Sergey Abramov:

====== <span style="font-size: 12pt;" >The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</span> ======
====== <span style="font-family: "times new roman", times;" >''Who is speaking? What is the speakers position, in terms of political, economic, technological power?''</span> ======
<p>The author of the article “The Future is for Everyone” is Mark Zuckerberg, the founder, chairman and CEO of Meta Platforms, Inc., an American multinational technology company which owns and operates several prominent social media platforms and communication services, including Facebook, Instagram, WhatsApp, Messenger, and Threads. </p>

<p>Meta has been described as a part of Big Tech, which refers to a few largest tech companies in the United States along with Microsoft, Alphabet (Google), Amazon, Apple, Nvidia and Tesla. Meta has historically posted huge profits and free cash flow, but it has been intensively investing in AI infrastructure projects for recent years. In the last three years (2023-2025), the technology giant has spent roughly $140bn in an attempt to get ahead of the AI boom. Furthermore, Meta plans to spend $125 billion to $145 billion on capital expenditures (capex) in 2026, mostly for AI infrastructure, including The Meta’s gigantic ‘Hyperion’ data-center project in Louisiana, which is the size of about 1,700 football fields. </p>
===== <br> =====
<p>The above illustrates that Meta has a tremendous economic and technological power, although recent spendings on AI infrastructure have significantly decreased free cash flow of the company, resulting in just $784 million in 1 quarter of 2026 which is a staggering 91% drop from the $8.55 billion recorded in the same period last year. </p>
===== <br> =====
<p>Mark Zuckerberg’s fortune is $196.6B according to Forbes, ranked on the 6th place in The World’s Real-Time Billionaires List. The Meta’s 2024 proxy statement reveals that CEO Mark Zuckerberg owns 99.7% of the outstanding Class B shares. While this represents only 13% of the economic ownership, it grants him 61% of the voting power. Most importantly, it also makes him the sole deciding vote on each proposal presented to shareholders. Given that, he is responsible for setting the overall direction and product strategy for the company, so he has enormous economic and technological power. </p>
===== <br> =====
<p>When it comes to his political weight, he is currently trying to build a strong and productive relationship with the US government and particularly Donald Trump’s administration. Meta previously confirmed that it had donated $1 million to Trump's inaugural fund in 2025.</p>
===== <br> =====
====== <span style="font-family: "times new roman", times;" >''How that position shapes what they speak about? How it shapes their perspective?''</span> ======
===== <br> =====
<p>Taking into account Meta’s tremendous spending on AI infrastructure and other AI projects ($140bn in 2023-2025 and up to $145 bn in 2026), it is clear that Meta is making a huge bet on AI. So despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust from investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. Apart from that Meta is going through big trials which might potentially result in additional massive expenses for the company.  </p>
===== <br> =====
<p>That pushes Meta to this article which constitutes a company’s manifestation on the bright future of humanity in the upcoming era of superintelligence and the ways and principles of achieving common prosperity and overall better life by means of products and ideas suggested by Meta’s CEO. </p>
===== <br> =====
<p>Obviously, in an attempt to regain investors' trust and win the favour of the US government, the author tries to contrast Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence.  </p>
===== <br> =====
====== <span style="font-family: "times new roman", times;" >''What is missing from the perspective? What is a blind spot? What is intentionally left out?''</span> ======
===== <br> =====
<p>The author notes that it is important to develop a philosophy for how we can best use superintelligence to ensure it improves all of our lives, work, communities, freedom and safety. He also comes up with a few specific ways Meta is building personal superintelligence to improve all of our lives.</p>
===== <br> =====
<p>At the same time, although the author acknowledges that new technologies also bring new risks, the way Meta attempts to address existing concerns doesn’t embrace completely all the negative aspects of possible superintelligence and existing problems caused by the AI development and AI infrastructure build-in.  </p>
===== <br> =====
<p>More specifically, the author mentions important concerns Meta is focused on addressing — “from concerns about job displacement and ensuring local communities benefit from data center builds, to safety concerns around AI misuse related to cybersecurity and biorisk, to avoiding government tyranny and surveillance, ensuring the US and democratic countries lead, and ultimately making sure humanity maintains control over superintelligence so it serves rather than endangers us”.</p>
===== <br> =====
<p>While speculating on the concerns about job displacement, the author makes too optimistic forecasts about jobs growth in the future thanks to superintelligence and personal superintelligent agents developed by Meta, but doesn’t mention massive AI-related lay-offs made over recent years. </p>
===== <br> =====
<p>Zuckerberg notes that there are several reasons to be optimistic that there will be an abundance of jobs in the future. No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it. If people can use AI to invent incredibly valuable new things, then it will make more sense to allocate it towards that rather than automating existing jobs. The more superintelligence serves as a tool of invention, the more likely that individual capability outpaces automation and the future is better for people.</p>
===== <br> =====
<p>Meanwhile, these optimistic forecasts are not aligned with the Meta’s lay-offs made during recent years. Meta had already eliminated around 25% of its workforce between late 2022 and early 2023.  Employees subsequently became increasingly concerned that the company’s AI strategy would accelerate job losses. Meta also laid off roughly 600 employees within its artificial intelligence unit in 2025. </p>
===== <br> =====
<p>Later this year, Meta explored cutting the size of many teams by as much as 60% and handing the work to AI agents, under a plan code-named Project OT. Meta cut about 8,000 jobs, roughly 10% of its staff in May 2026. It moved 7,000 people into AI roles in the same week.</p>
===== <br> =====
<p>The plan envisioned AI taking over much of the routine work performed by thousands of employees, with smaller teams of highly skilled workers supervising AI-powered operations.</p>
===== <br> =====
<p>One of the main concerns related to AI and mentioned by the author is sustainable AI infrastructure development which according to the author means that communities must benefit significantly from each project. This includes high-paying local jobs, investment in schools and public services, ensuring energy prices don’t rise, and taking care of the environment.</p>
===== <br> =====
<p>Undoubtedly, launching a Future Is For Everyone Fund, $1 bn community-focused initiative created to financially support the U.S. towns and regions that host Meta’s massive, resource-heavy AI data centers, and an America's Workforce Academy, a $115 million nationwide program offering zero-cost, 4 to 5-week training for in-demand skilled trades, are great attempts to contribute to local communities, but there are another more crucial aspects of AI data centers build-out which slightly mentioned on the Meta’s CEO vision on the future supported by superintelligence. </p>
===== <br> =====
<p>The most crucial concern is the negative environmental impact of AI infrastructure, particularly on adjacent local communities. Karen Hao in Empire of AI notes that hyperscalers’ data centers are large tracts of land with several massive buildings densely packed with racks and racks of computers which emanate an unseemly amount of heat. To keep them from overheating, the buildings also have massive cooling systems - large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of different noises that can be heard for miles, twenty for hours a day, creating a relentless and body-warping source of noise pollution. </p>
===== <br> =====
<p>Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. </p>
===== <br> =====
<p>By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022. And the extraordinary volume of potable water often needed to cool servers. According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally by 2027, or half the water annually consumed in the UK which can lead to a severe water crisis.    </p>
===== <br> =====
<p>Zuckerberg underlines that their data centers are designed to be among the most water-efficient in the world. According to him Meta is committed to being water-positive, meaning that it’ll restore more water than it uses in the watersheds where Meta operates by 2030. Zuckerberg assures that Meta helps keep electricity prices low by building their own energy-generating infrastructure wherever they invest. This ensures that not only is Meta not consuming energy that could have gone to the local communities, but in some cases Meta even supplies a surplus of low-cost energy back to the communities. </p>
===== <br> =====
<p>These statements look reasonable in theory, but there is no data or evidence that these measures will be implemented and how sufficient they will be for sustainability.</p>
===== <br> =====
====== <span style="font-family: "times new roman", times;" >''Who is the intended audience of the text? Why is the text speaking to that particular audience?''</span> ======
===== <br> =====
<p>Taking into account several policy recommendations related to cooperation between AI companies and government and regulation of AI development, including American leadership, and measures to reduce cybersecurity, biological and chemical risks, which Zuckerberg has outlined throughout this article, presumably, the intended audience of this article is largely the US government, law- and policymakers. The second potential category of intended audience presumably could be investors and potential users of personal superintelligence agents and other AI tools developed by Meta as Zuckerberg proposes a few specific ways Meta is building personal superintelligence to improve all of our lives. </p>
===== <br> =====
<p>What is the intended goal of the text? Why has the author written this text? What does he/she want to achieve?</p><p>I suppose that this article has two main intended goals. </p><p>The first goal is to persuade law-, policymakers and the US government to support laws or rules based on recommendations outlined by Zuckerberg that give Meta an advantage compared to its competitors.</p><p>The second goal is to regain long-term investors' trust and confidence in positive outcomes and great long-term returns after the current enormous spending on AI infrastructure and persuade them to invest their money in the company. </p>
===== <br> =====
====== <span style="font-family: "times new roman", times;" >''What is the wider context of the text? What are the conversations/events/in which this text is situated?''</span> ======
===== <br> =====
<p>As i mentioned earlier, Meta has spent on AI infrastructure and other AI projects enormous amount of money ($140bn in 2023-2025 and up to $145 bn in 2026) and because of that despite great revenue Meta is experiencing significant reduction of free cash flow which results in loss of trust of investors who are not certain whether all these massive spendings will pay off in the future. Consequently, the economic position of Meta compared to its competitors from BigTech is weaker at the moment. </p>
===== <br> =====
<p>Apart from that, Meta has recently suffered major court losses in child safety and addiction lawsuits in the US before agreeing to an $18 billion multi-state settlement in August 2026 to resolve hundreds of similar pending claims. In the European Union, Meta has suffered several major losses in front of European regulators and top EU courts, particularly surrounding market dominance, copyright laws, and privacy.</p>
===== <br> =====
<p>Some researchers advocate the idea that superintelligent AI could theoretically eliminate humanity, others, including some Meta’s AI rivals, e.g. Anthropic, urge to focus on safe and responsible AI to harness it and set clear boundaries for superintelligence or even stop further AI development. </p>
===== <br> =====
<p>One of Meta’s AI rivals, Anthropic, has recently been involved in conflict with the US government. The conflict stemmed from Anthropic refusing the U.S. Department of Defense's demand to remove AI safety guardrails, drawing strict red lines against using its technology for mass domestic surveillance and fully autonomous weapons. In response, the Trump administration blacklisted the company as a "supply chain risk," a punitive measure that a federal judge permanently blocked in August 2026, ruling it as unlawful government retaliation. Nevertheless, for some time it caused serious problems for Anthropic. </p>
===== <br> =====
<p>There have also been 2 cybersecurity incidents with Anthropic and OpenAI recently where AI hacked real-world companies. </p>
===== <br> =====
====== <span style="font-family: "times new roman", times;" >''How does this text shifts/alters the context? What is the reaction to the text?''</span> ======
===== <br> =====
<p>It seems unlikely that this text might alter the context significantly. Meta is experiencing troubles now and making a huge bet on AI spending billions of US dollars on AI infrastructure without sensible returns. </p>
===== <br> =====
<p>This text could help to rebuild trust of small group of investors and win the favour of the US government, as the author contrasts Meta favourably with AI rivals, sugarcoat reality and persuade customers that their products are well-positioned for the bright future with superintelligence for individuals  where democracy and western values dominate.  </p>
===== <br> =====
<p>There are various reactions to the text, but the majority does not take it very seriously. Chinasa T Okolo argues that Meta has ‘grossly overstated the economic benefits of its data centers’, noting that employment benefits are concentrated primarily during construction and that operational data centres employ only a small fraction of permanent workers. Okolo further highlights that data centers are also associated with negative health and environmental impacts, which are likely to be even more challenging in African countries, which have fewer protections and lack the ability for governments to oversee strict compliance if environmental laws exist.</p>
===== <br> =====
<p>Abeba Birhane points out that while the first sections of the document target the global community — and regulators in places like Brazil, India, Turkey, and Nigeria — with a message of transparency, horizontality, and willingness to cooperate broadly, the latter sections are directed to U.S. regulators and power brokers who expect Meta to be a vector of U.S. power. Abeba Birhane further notes that nobody should take Meta or Zuckerberg’s letter to “benefit everyone” seriously when the corporation that continues to silence criticism, retaliates against its own workers for exposing its technology’s harms, buries evidence of harm to minors, actively launches problematic products (the perv glasses) when overworked and underfunded civil society organizations have their attention focused elsewhere, and continually violates its legal obligations somehow claims to have people’s — especially vulnerable groups’ — best interests at heart. It’s laughable”.</p>
===== <br> =====
====== ''<span style="font-family: "times new roman", times;" >What is your reaction/analysis?</span>'' ======
===== <br> =====
<p>This text is a manifesto of Meta Platforms Inc. on the potential usage of superintelligence, access to it with balance of powers in favour of individuals rather than institutions, ways of governance and control over it. It also tries to address important concerns related to superintelligence by introducing new financial support initiatives, governance and oversight structures, government policy recommendations etc. and making forecasts about the future.</p>
===== <br> =====
<p>All new financial support initiatives he has announced seem positive, but not sufficient for solving all issues with job displacement, environmental harm, people’s security, imbalance of power and support of democracy values. Given that the budget of these financial support initiatives is less than 1 percent of total capital expenditure of Meta for this year, it raises a question whether Meta is really focused on addressing job displacement and economy transformation concerns or just feign concern. The AI governance initiatives look vague at the moment and seemingly efficient.  </p>
===== <br> =====
<p>Taking into account that on the 8th of April, 2026 Meta introduced Muse Spark, a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration as a first step toward a personal superintelligence, and the way how Zuckerberg describes the features and advantages of the personal superintelligence tools, this article can be considered as a part of PR and advertisement campaign promoting personal superintelligent agents and models which are being developed by Meta in an attempt to draw attention to it and regain investors’ trust and confidence in long-term positive return on huge AI infrastructure spendings.      </p>
===== <br> =====
<p>Zuckerberg commits to build a fully private mode where even Meta cannot see or grant access to users information and focus heavily on alignment — defined as helping each person achieve their goals, not Meta’s. He further suggests implementing a governance structure that gives Meta’s independent board of directors the power to approve the safety criteria for releasing models and reviewing whether each model release adheres to the criteria. These commitments can be viewed as trustworthy elements of governance of superintelligence tools. </p>
===== <br> =====
<p>I agree with Zuckerberg’s argument that open source is a positive and important force for empowering people and preventing centralization that is detrimental for both safety and the economy. Building on this argument, Zuckerberg emphasises that Meta continues to be strongly supportive of open source, including open source AI models, which can be interpreted both as means of promoting Meta’s superintelligence agents and models based on open-source technology and an effort to persuade the broader community, policymakers, and the US government that this vision merits financial support and a permissive regulatory approach. </p>
===== <br> =====
<p>A critical reading of this article suggests that Zuckerberg’s policy recommendations such as encouraging close cooperation between frontier labs and the government and ensuring that the government has the technical resources to harden critical systems and make effective use of advanced models can be interpreted as an intention to establish closer relationship with the US government. Furthermore, Zuckerberg claims that it is also important for national security that the American AI industry continues to lead and therefore the ability to build infrastructure - whether that’s energy, data centers, or silicon - should be accelerated. This can be considered as an effort to lobby Meta’s interests related to AI infrastructure  and win favour of the US government because Meta currently faces severe pushback over their Hyperion facility air pollution permit, ongoing legal battles to hide its projected energy data from the public, and fierce opposition from consumer groups demanding strict approvals to ensure local ratepayers are not stuck paying for the massive grid upgrades needed to power the campus.</p>
===== <br> =====
====== References ======

Assignment 1

2026-09-12T16:09:09Z

Sergey Abramov: Undid revision 123803 by Sergey Abramov (talk)

<p>The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</p>

<br>

Assignment 1

2026-09-12T16:07:56Z

Sergey Abramov:

<p>The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</p>
<p>Who is speaking? What is the speakers position, in terms of political, economic, technological power?</p>

Assignment 1

2026-09-12T16:05:47Z

Sergey Abramov: Created page with "<p>The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</p> <br>"

<p>The article “The Future is for Everyone” review by Sergey Abramov (assignment 1)</p>

<br>

Governance by history

2026-09-08T18:08:34Z

Sergey Abramov: /* Mechanism of path dependence */

= [[Workplace surveillance]] =
== [[Platform Dependency]] ==
<p class="p1"><span class="s1">8 September 2026 - Platform Dependence in Public Communication - Ceren and Rodrique (work in progress)</span></p><p class="p1"><span class="s1">I.                PLATFORM DEPENDENCE IN PUBLIC COMMUNICATION – WHAT IS IT , WHO IS AFFECTED BY IT AND WHY IT MATTERS LEGALLY AND INSTITUTIONALLY – C</span></p><p class="p1"><span class="s1">A.           Public communication is the conveyance of a message to a large group of people. This includes public announcements, advertisements in public spaces, social media broadcasts, news delivered in reporting, editorials, opinion pieces, and even political campaigns. Platform dependence is the reliance on a third party platform or digital infrastructure to reach the audience of the public communication, the public. Platform dependence in public communication therefore includes reliance on private companies like Meta (Facebook), state television channels, or telecommunications providers who provide and control access to, and develop the infrastructure of such platforms. For the purpose of this exercise, news will be utilized as the main example of public communication.</span></p><p class="p1"><span class="s1">B.          Those who are affected by platform dependence in public communication are the communicators and the public alike. News outlets are mainly for-profit organizations. These organizations’ reliance on third party digital platforms grows day-by-day to maintain their popularity and engagement (such as on social media platforms) ultimately deciding the revenue generated from advertisers. The public on the other hand are provided catchier and more diluted information, gaining just enough knowledge to grab their attention while not being provided substantial knowledge on matters. Additionally,i members of the public who do not have the resources to access such platforms and digital services remain “information poor”. Digital rich, the virtual class of tech-professional and entrepreneurs benefit from the dependence to their platforms.</span></p><p class="p1"><span class="s1">C.            This is crucial as platform dependence may threaten democracy and free political expression. Platforms, run by private parties, can be run according to their political beliefs. Social media platforms aiming to keep users online utilize algorithms that provide their users with their confirmed preferences, not allowing the conveyance of a truly heteregenious catalogue of news. The original ideal of the democratic electronic agora is undermined when private corporations control the flow of news and the infrastructure on which they are provided.</span></p><p class="p1"><span class="s1">II.              HISTORICAL LINEAGE – HISTORICAL MOMENTS/DECISIONS/INSTITUTIONAL ARRANGEMENTS - R</span></p><p class="p1"><span class="s1">It is very difficult to discuss news without highlighting the deep entanglement it shares with with information. For our purpose 'information' will have the the following characteristics; it can be learned, processed, archived, and it's worth correlates to its rarity.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-1 [1]

<span class="s2">]</span><p class="p1"><span class="s1">News is the end result of the process of taking this information from one form to another, and transmitting to a wider, typically public audience is what we will anchor our definition of 'news' to be. There is and has always been an obvious tension between the rarity that gives information its worth and the act of transmission that makes it less rare. Modern form of this can be traced to Renaissance Europe,</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-2 [2]

<span class="s2">]</span><p class="p1"><span class="s1">where a single innovation not only leaned on but also displaced established pattern of trusted networks collecting, processing and sharing news by way of the church, merchants, students, diplomats and so on.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-3 [3]

<span class="s2">]</span><p class="p1"><span class="s1">This innovation is the invention of printing in the 15th century. It had consequences beyond those imagined by its inventor, as it offered new possibilities for the dissemination of news. It must be reiterated that printing benefitted from the existing networks and not the other way around. This is evident by Johann Carolus'  ''Relation aller Fürnemmen und gedenckwürdigen Historien'' handwritten weekly newspaper for an elite audience fruitfully running prior to the advent of printing. However, the printing press coupled with existing network and infrastructure of the postal system brought industrialisation and scale unknown to the trade and the world at the time. One of the first paying customer for press was the Church, selling indulgences. Beniger's stands: control depends on information processing and communication working together, and the this innovation at that current point time supplied both at once.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-4 [4]

<span class="s2">]</span><p class="p1"><span class="s1">Nonetheless, a strong parallel between the press and later computing revolutions it's stated relationship with state.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-5 [5]

<span class="s2">]</span><p class="p1"><span class="s1">With the former seeming to lean more towards the state and the other in the opposite direction.</span></p><p class="p1"><span class="s1">The next pivotal historical point which disturbs the way in which news is transmitted is by way of the radio. The radio is huge leap frog as it permitted the decoupling of reach from cost, reliance on aging network such as the postal system. Where the method of control under print was through the content printed, the radio moved closer to physics on licensing broadcast rights. This changed the trajectory of new's as it birth the state owned monopoly broadcaster, responsible for broadcasting news to anyone with a receiver on the other end within appropriate frequency. News outlet now had more than one method for distributing news and for the first time had a way of engaging in dialogue on the news of the day as it makes it way via print and sound-wave. Diversity of dependency is an important aspect of this stage in history.</span></p><p class="p3"><span class="s1"></span><br></p><p class="p1"><span class="s1">III.             MECHANISM OF PATH DEPENDENCE - C</span></p><p class="p1"><span class="s1">A.            The public is generally partaking in consuming content on social media. Hence, with its 3 billion monthly active users for example, Instagram has an installed base for news outlets to depend upon (Instagram Revenue and Usage Statistics (2026), n.d.). The audiences of news outlets spend an increasing amount of their time on social media platforms, which gives results in a habitual entrenchment for these news outlets: the means of reaching the audience is dependent on their habits of content consumption. An indirect contractual entrenchment may also be discussed, as news outlets have obligations to their advertisers (who may request the upkeep of views/clicks on the news platforms they pay to advertise in). Cultural entrenchment should not be disregarded: the public expects to receive their news on social media platforms they readily frequent, and many news outlets meet this expectation. Hence, news outlets must partake in building a presence in social media platforms if they wish to compete.</span></p><p class="p1"><span class="s1">IV.            POLITICS AND DISTRIBUTION – WHOSE INTERESTS ARE STRENGTHENED AND WHO HAS FEWER CHOICES - C</span></p><p class="p1"><span class="s1">A.            Barbook argues that although the California Ideology claims that state power will be decreased and individuals will be empowered through decentralized networks enabled by digital technologies, he demonstrates that public communication is in fact not immune to undemocratic powers. These include private corporations: these corporations may gatekeep in choosing when and how to provide access to their platforms, while finetuning how users navigate their platforms.</span></p><p class="p1"><span class="s1">By enabling access to content that users tend to agree with, those who manage these platforms enable public opinion to shift to extremes. Being “fed” the same narrative and not being challenged to see an opposing perspective ultimately provides fewer choices to the consumer of the information, the general public. Therefore, individual agency is constrained by platform architecture and control, not liberated by it.</span></p><p class="p1"><span class="s1">V.              AT WHAT POINT IN HISTORY WAS A DIFFERENT TRAJECTORY POSSIBLE? - R</span></p><p class="p1"><span class="s1">A.            News not subscribing to / presenting news on social media as platform (?)</span></p><p class="p1"><span class="s1">VI.            GOVERNANCE IMPLICATION – ONE PROPOSITION FOR CURRENT DAY GOVERNANCE - R</span></p><p class="p1"><span class="s1">OPEN PLATFORMS (ALTERNATIVE FACEBOOK A)</span></p><p class="p1"><span class="s1"><section class="footnotes"></span></p><p class="p4"><span class="s1"></span><br></p>
#<span class="s3">Mark Greengrass, Thierry Rentet and Stéphane Gal, ‘The Hinterland of the Newsletter: Handling Information in Space and Time’ in Joad Raymond and Noah Moxham (eds), News Networks in Early Modern Europe (Brill 2016)</span><span class="s1"><br>[http://www.jstor.org/stable/10.1163/j.ctt1w8h1ng.34 http://www.jstor.org/stable/10.1163/j.ctt1w8h1ng.34]</span><span class="s3">accessed 7 September 2026; ‘Michel Serres Prononce En Sorbonne La Conférence Inaugurale Du Programme Paris Nouveaux Mondes - La Chancellerie Des Universités de Paris’ (La Chancellerie des Universités de Paris, 1 January 2013)</span><span class="s1"><br>[https://www.sorbonne.fr/michel-serres-prononce-en-sorbonne-la-conference-inaugurale-du-programme-paris-nouveaux-mondes/ https://www.sorbonne.fr/michel-serres-prononce-en-sorbonne-la-conference-inaugurale-du-programme-paris-nouveaux-mondes/]</span><span class="s3">accessed 7 September 2026. James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986).</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-1 ↩︎]</span>
#<span class="s3">Brendan Dooley, Making News in Renaissance Europe (1st edn, Cambridge University Press 2026)</span><span class="s1"><br>[https://doi.org/10.1017/9781009358835 https://doi.org/10.1017/9781009358835]</span><span class="s3">.</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-2 ↩︎]</span>
#<span class="s3">Andrew Pettegree, The Invention of News: How the World Came to Know about Itself (Yale University Press 2014). Helmer Helmers, ‘Public Diplomacy in Early Modern Europe: Towards a New History of News’ (2016) 22 Media History 401</span><span class="s1"><br>[https://doi.org/10.1080/13688804.2016.1174570 https://doi.org/10.1080/13688804.2016.1174570]</span><span class="s3">.</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-3 ↩︎]</span>
#<span class="s3">James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986) P24-25</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-4 ↩︎]</span>
#<span class="s3">Richard Barbrook and Andy Cameron, The Californian Ideology (1996)</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-5 ↩︎]</span>
<p class="p1"><span class="s1"></section></span></p>

== [[First One]] ==
= Centralization of AI Compute and Data =

== Centralization of AI compute and Data - by Atmika Dechamma & Pablo Latorre ==
=== Governance Issue ===
AI computing capacity and data presents the governance problem of an extreme concentration of technological infrastructure and information in a restricted number of companies that privately and globally dominate the market. This affects public and private actors - such as small and medium enterprises, potential new competitors, and Global South countries - that increasingly depend on AI systems outside their control and/or are deterred from competing with established companies because of impenetrable entry barriers - most notably initial investments and operational costs. The institutional and legal relevance of this is that it consolidates, reproduces and amplifies asymmetries of power in favor of the “Big Tech”, reverting the technological trend from digital decentralization to physical infrastructure dependency and curbing innovation and market efficiency.

=== Historical Lineage ===
==== Limitation of Machine Learning ====
<p style="text-align: justify;">The AI revolution was built on the limitation of human capabilities to process large information and derive efficient results - where before there existed only facilitative tools to allow for human-enabled processes, AI now offered derivations and cognitive reasoning in moments and at large scales.</p>

<br>
==== Revert to Physical Infrastructure Dependency in the Rush to Be the Vest ====
<p style="text-align: justify;">The present AI model heavily relies on hardware in order to perform complex data processing for deep learning. The data centre models followed require larger model sizes (which require higher resource, information and capital investment) to be able to yield better results. Due to the limitation of resources and a rush to establish the best working model of an AI agent, data centres are built on a rudimentary model heavily dependent on massive amounts of electricity and fresh water.</p>

<br>
==== Regulation and Decentralization of Information ====
<p style="text-align: justify;">Regulation and influence of sovereigns and institutions has resulted in data sharing and information exchange structurally through open source and cloud computing. But, while AI models claims their selling point to be accessibility for all, their information and derivations remain in a closed server, stored for the purpose of training their AI agents to be the best. Big AI contenders attract higher consumer and information flows overall resulting in a better developed AI model, proving disadvantageous in any market where early contenders have an advantage and accumulation of power.</p>

=== Mechanisms of Path Dependence ===
<p style="text-align: justify;">Early on in the industry there is an expectation of human and capital investment from all stakeholders involved, this high cost for entry into the market and expected capital to sustain and develop a model by both consumer and enterprise makes for a high switching cost. For the AI model to run and scale there is a presumption that the rate of growth is dependent on the size of data fed, resulting in a better model and hence, more consumer usage. A closed centralised data loop is created but the requirement for storage and processing far surpasses current infrastructures capabilities - heavily relying on early market entry advantage and sustained capital investments taking away from smaller aspirants and hoarding of power within closed circles. Despite its physical dependency, AI data processing has unique requirements beyond legacy infrastructures - larger demand and production of GPUs and designed to handle immense thermal heat generated while consuming electricity beyond current capabilities. </p>

<br><p style="text-align: justify;">The AI space largely remains unregulated due to its undefined form and fluid nature, however, this has given way for unethical and morality issues regarding data privacy, exploitation of labour and environmental harms.</p>

=== Politics and Distribution ===
Concentration strengthens the values, interests, capacities, and authority of the “Big Tech” over other actors of public and private nature. In this context, Barbrook and Cameron (1) illuminate the ethos behind AI technology development while Winner (2) makes observable the power relations surrounding it: concentration is the consequence of an AI development anchored in the belief that it must be free from all institutional and normative restriction since it is inherently liberating and non-political which, in turn, provides a political/moral justification for the concentration (if technology is inherently liberating and non-political then there is no power relations to observe or question) that translates itself into power dynamics where the AI developer not only owns the AI technology but also decides how it is trained and deployed and which models the user can or cannot access.

=== Contingency and Alternatives ===
Concentration is due, first and foremost, to the fact that AI computing capacity requires massive and cutting-edge physical infrastructure - such as data centers and microprocessors - which, in turn, requires upfront investments that are unaffordable to most companies. In David’s (3) language, companies that could finance the infrastructure and position themselves first obtained self-reinforcing advantages and a “lock-in” effect in the market. In other words, concentration is the consequence of the development of AI physical infrastructure privately with no mixed or public alternatives as a public service. Therefore, there was a space for an alternative future at the end of the 1980s or the beginning of the 1990s when the privatization model of the technological infrastructure was consolidated in line with the “Californian Ideology”. In that moment governments had the chance to treat technological infrastructure as a mixed or public good rendering it universally accessible as it was the case of the Internet; this would have permitted companies of all sizes to profit from economies of scale, reducing entry barriers and fostering competition. However, this alternative path did not prevail since the Californian Ideology provided the moral/political justification to exclude governments from the development of technological infrastructure in the exact moment in which such a decision would produce structural consequences. If that path had been followed it could have contributed to collective access to economies of scale, lower entry barriers, and a more competitive AI market.

=== Governance Implication ===
Considering the above, public or mixed investment on AI physical infrastructure is a necessary governance measure for market decentralization and collective access to AI computing capacity and data. From an historical perspective, the proposed measure is not merely a corrective action for a market failure but for a contingent and ideological decision made between the late 1980s and early 1990s that produced a structural “path dependence” that now is necessary to change.
==== References ====
(1) [https://git.xpub.nl/onebigear/SI_09_reference_library/raw/commit/22bf61e1dcfed35d58fba23808212bd00584be47/Richard_Barbrook_Andy_Cameron_Californian_ideology.pdf Barbrook, R., & Cameron, A. (1996). The Californian Ideology. <em>Science as Culture</em>, <em>6</em>(1), 44-72].

(2) [https://d1wqtxts1xzle7.cloudfront.net/87708502/langdon_winner_do_artifacts_have_politics-libre.pdf?1655588533=&response-content-disposition=inline%3B+filename%3DDo_Artifacts_Have_Politics.pdf&Expires=1788890463&Signature=UBBiMqkip5MfYZbkbLNAN6aE2hyZdKkER-HFcmyxXkwIZuTnMKdOGoao8ddFr-rQ5T55A8NWUOlnZkGYmi66e7XR9h99DgTELMCrh48sXjOb8KYgA48r~SUUymt1IW~R~aOKJzitMIUHcrc6yEGKK1OuaA~mrEIVOXpUjQMYjqOd26Xmk9YJEglTkqaxeR47VCCth5FvozNQIowdfjzeu8aNHnI8V7Yur1IiXY~PGvYRcN6wBtcgpVMHZUtNaI8NjmYbp2p5TBYrPMF9MCLtvpGhPlsCfsvA8Bdm6XulGwRlksZBKWvVUQ8D~M8oLu12V39YaiETIObyJqbE7ctIcQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA Winner, L. (2017). Do artifacts have politics?. In <em>Computer ethics</em>(pp. 177-192). Routledge].

(3) [https://sjbae.pbworks.com/f/David+1985.pdf David, P. A. (1985). Clio and the Economics of QWERTY. <em>The American economic review</em>, <em>75</em>(2), 332-337].

== Centralization of AI compute and Data - by Sergey Abramov and Paul Stein ==
<div>
=== The current governance issue ===
<p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>
<br>
=== Historical lineage ===
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p>
<br><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p>
<br><p>Soon after Amazon launched AWS, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p>
<br><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>
<br>
=== Mechanism of path dependence ===
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers<sup>4</sup>.</p>

<br><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>

<br>

=== Politics and distribution ===
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities.</p>
<br><p>The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
<br>
=== Contingency and Alternatives ===
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. </p>
<br><p>The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p>
<br>
=== Governance implication ===
<p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup></p>
<br><p>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. </p>
<br><p>Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p>

=== References & Notes ===
1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from 'https://epoch.ai/data-insights/hyperscalers-control-most-compute' [online resource]. Accessed 7 Sept 2026.</p>2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026) https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world accessed 7 September 2026.

3. Amazon Web Services (AWS), a cloud computing platform

4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025) https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf accessed 7 September 2026.

5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.

6. Ibid. 123.

7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.

8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025) https://texasroyaltybrokers.com/ai-clusters-by-state/ accessed 7 September 2026.

9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.

10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.

11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026) https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 accessed 7 September 2026.

12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'') https://www.nsf.gov/impacts/internet accessed 7 September 2026.

13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'') https://casrai.org/dictionary/term/national-ai-research-resource-nairr accessed 7 September 2026.

14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.

15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024) https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en accessed 7 September 2026.</div>

Governance by history

2026-09-08T18:07:26Z

Sergey Abramov: /* Mechanism of path dependence */

= [[Workplace surveillance]] =
== [[Platform Dependency]] ==
<p class="p1"><span class="s1">8 September 2026 - Platform Dependence in Public Communication - Ceren and Rodrique (work in progress)</span></p><p class="p1"><span class="s1">I.                PLATFORM DEPENDENCE IN PUBLIC COMMUNICATION – WHAT IS IT , WHO IS AFFECTED BY IT AND WHY IT MATTERS LEGALLY AND INSTITUTIONALLY – C</span></p><p class="p1"><span class="s1">A.           Public communication is the conveyance of a message to a large group of people. This includes public announcements, advertisements in public spaces, social media broadcasts, news delivered in reporting, editorials, opinion pieces, and even political campaigns. Platform dependence is the reliance on a third party platform or digital infrastructure to reach the audience of the public communication, the public. Platform dependence in public communication therefore includes reliance on private companies like Meta (Facebook), state television channels, or telecommunications providers who provide and control access to, and develop the infrastructure of such platforms. For the purpose of this exercise, news will be utilized as the main example of public communication.</span></p><p class="p1"><span class="s1">B.          Those who are affected by platform dependence in public communication are the communicators and the public alike. News outlets are mainly for-profit organizations. These organizations’ reliance on third party digital platforms grows day-by-day to maintain their popularity and engagement (such as on social media platforms) ultimately deciding the revenue generated from advertisers. The public on the other hand are provided catchier and more diluted information, gaining just enough knowledge to grab their attention while not being provided substantial knowledge on matters. Additionally,i members of the public who do not have the resources to access such platforms and digital services remain “information poor”. Digital rich, the virtual class of tech-professional and entrepreneurs benefit from the dependence to their platforms.</span></p><p class="p1"><span class="s1">C.            This is crucial as platform dependence may threaten democracy and free political expression. Platforms, run by private parties, can be run according to their political beliefs. Social media platforms aiming to keep users online utilize algorithms that provide their users with their confirmed preferences, not allowing the conveyance of a truly heteregenious catalogue of news. The original ideal of the democratic electronic agora is undermined when private corporations control the flow of news and the infrastructure on which they are provided.</span></p><p class="p1"><span class="s1">II.              HISTORICAL LINEAGE – HISTORICAL MOMENTS/DECISIONS/INSTITUTIONAL ARRANGEMENTS - R</span></p><p class="p1"><span class="s1">It is very difficult to discuss news without highlighting the deep entanglement it shares with with information. For our purpose 'information' will have the the following characteristics; it can be learned, processed, archived, and it's worth correlates to its rarity.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-1 [1]

<span class="s2">]</span><p class="p1"><span class="s1">News is the end result of the process of taking this information from one form to another, and transmitting to a wider, typically public audience is what we will anchor our definition of 'news' to be. There is and has always been an obvious tension between the rarity that gives information its worth and the act of transmission that makes it less rare. Modern form of this can be traced to Renaissance Europe,</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-2 [2]

<span class="s2">]</span><p class="p1"><span class="s1">where a single innovation not only leaned on but also displaced established pattern of trusted networks collecting, processing and sharing news by way of the church, merchants, students, diplomats and so on.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-3 [3]

<span class="s2">]</span><p class="p1"><span class="s1">This innovation is the invention of printing in the 15th century. It had consequences beyond those imagined by its inventor, as it offered new possibilities for the dissemination of news. It must be reiterated that printing benefitted from the existing networks and not the other way around. This is evident by Johann Carolus'  ''Relation aller Fürnemmen und gedenckwürdigen Historien'' handwritten weekly newspaper for an elite audience fruitfully running prior to the advent of printing. However, the printing press coupled with existing network and infrastructure of the postal system brought industrialisation and scale unknown to the trade and the world at the time. One of the first paying customer for press was the Church, selling indulgences. Beniger's stands: control depends on information processing and communication working together, and the this innovation at that current point time supplied both at once.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-4 [4]

<span class="s2">]</span><p class="p1"><span class="s1">Nonetheless, a strong parallel between the press and later computing revolutions it's stated relationship with state.</span></p><p class="p2"><span class="s1"></span></p>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fn-5 [5]

<span class="s2">]</span><p class="p1"><span class="s1">With the former seeming to lean more towards the state and the other in the opposite direction.</span></p><p class="p1"><span class="s1">The next pivotal historical point which disturbs the way in which news is transmitted is by way of the radio. The radio is huge leap frog as it permitted the decoupling of reach from cost, reliance on aging network such as the postal system. Where the method of control under print was through the content printed, the radio moved closer to physics on licensing broadcast rights. This changed the trajectory of new's as it birth the state owned monopoly broadcaster, responsible for broadcasting news to anyone with a receiver on the other end within appropriate frequency. News outlet now had more than one method for distributing news and for the first time had a way of engaging in dialogue on the news of the day as it makes it way via print and sound-wave. Diversity of dependency is an important aspect of this stage in history.</span></p><p class="p3"><span class="s1"></span><br></p><p class="p1"><span class="s1">III.             MECHANISM OF PATH DEPENDENCE - C</span></p><p class="p1"><span class="s1">A.            The public is generally partaking in consuming content on social media. Hence, with its 3 billion monthly active users for example, Instagram has an installed base for news outlets to depend upon (Instagram Revenue and Usage Statistics (2026), n.d.). The audiences of news outlets spend an increasing amount of their time on social media platforms, which gives results in a habitual entrenchment for these news outlets: the means of reaching the audience is dependent on their habits of content consumption. An indirect contractual entrenchment may also be discussed, as news outlets have obligations to their advertisers (who may request the upkeep of views/clicks on the news platforms they pay to advertise in). Cultural entrenchment should not be disregarded: the public expects to receive their news on social media platforms they readily frequent, and many news outlets meet this expectation. Hence, news outlets must partake in building a presence in social media platforms if they wish to compete.</span></p><p class="p1"><span class="s1">IV.            POLITICS AND DISTRIBUTION – WHOSE INTERESTS ARE STRENGTHENED AND WHO HAS FEWER CHOICES - C</span></p><p class="p1"><span class="s1">A.            Barbook argues that although the California Ideology claims that state power will be decreased and individuals will be empowered through decentralized networks enabled by digital technologies, he demonstrates that public communication is in fact not immune to undemocratic powers. These include private corporations: these corporations may gatekeep in choosing when and how to provide access to their platforms, while finetuning how users navigate their platforms.</span></p><p class="p1"><span class="s1">By enabling access to content that users tend to agree with, those who manage these platforms enable public opinion to shift to extremes. Being “fed” the same narrative and not being challenged to see an opposing perspective ultimately provides fewer choices to the consumer of the information, the general public. Therefore, individual agency is constrained by platform architecture and control, not liberated by it.</span></p><p class="p1"><span class="s1">V.              AT WHAT POINT IN HISTORY WAS A DIFFERENT TRAJECTORY POSSIBLE? - R</span></p><p class="p1"><span class="s1">A.            News not subscribing to / presenting news on social media as platform (?)</span></p><p class="p1"><span class="s1">VI.            GOVERNANCE IMPLICATION – ONE PROPOSITION FOR CURRENT DAY GOVERNANCE - R</span></p><p class="p1"><span class="s1">OPEN PLATFORMS (ALTERNATIVE FACEBOOK A)</span></p><p class="p1"><span class="s1"><section class="footnotes"></span></p><p class="p4"><span class="s1"></span><br></p>
#<span class="s3">Mark Greengrass, Thierry Rentet and Stéphane Gal, ‘The Hinterland of the Newsletter: Handling Information in Space and Time’ in Joad Raymond and Noah Moxham (eds), News Networks in Early Modern Europe (Brill 2016)</span><span class="s1"><br>[http://www.jstor.org/stable/10.1163/j.ctt1w8h1ng.34 http://www.jstor.org/stable/10.1163/j.ctt1w8h1ng.34]</span><span class="s3">accessed 7 September 2026; ‘Michel Serres Prononce En Sorbonne La Conférence Inaugurale Du Programme Paris Nouveaux Mondes - La Chancellerie Des Universités de Paris’ (La Chancellerie des Universités de Paris, 1 January 2013)</span><span class="s1"><br>[https://www.sorbonne.fr/michel-serres-prononce-en-sorbonne-la-conference-inaugurale-du-programme-paris-nouveaux-mondes/ https://www.sorbonne.fr/michel-serres-prononce-en-sorbonne-la-conference-inaugurale-du-programme-paris-nouveaux-mondes/]</span><span class="s3">accessed 7 September 2026. James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986).</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-1 ↩︎]</span>
#<span class="s3">Brendan Dooley, Making News in Renaissance Europe (1st edn, Cambridge University Press 2026)</span><span class="s1"><br>[https://doi.org/10.1017/9781009358835 https://doi.org/10.1017/9781009358835]</span><span class="s3">.</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-2 ↩︎]</span>
#<span class="s3">Andrew Pettegree, The Invention of News: How the World Came to Know about Itself (Yale University Press 2014). Helmer Helmers, ‘Public Diplomacy in Early Modern Europe: Towards a New History of News’ (2016) 22 Media History 401</span><span class="s1"><br>[https://doi.org/10.1080/13688804.2016.1174570 https://doi.org/10.1080/13688804.2016.1174570]</span><span class="s3">.</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-3 ↩︎]</span>
#<span class="s3">James Ralph Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Harvard Univ Press 1986) P24-25</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-4 ↩︎]</span>
#<span class="s3">Richard Barbrook and Andy Cameron, The Californian Ideology (1996)</span><span class="s1"><br>[https://wiki.ivir.nl/index.php?title=Platform_dependence_in_public_communication&action=tinymceedit#fnref-5 ↩︎]</span>
<p class="p1"><span class="s1"></section></span></p>

== [[First One]] ==
= Centralization of AI Compute and Data =

== Centralization of AI compute and Data - by Atmika Dechamma & Pablo Latorre ==
=== Governance Issue ===
AI computing capacity and data presents the governance problem of an extreme concentration of technological infrastructure and information in a restricted number of companies that privately and globally dominate the market. This affects public and private actors - such as small and medium enterprises, potential new competitors, and Global South countries - that increasingly depend on AI systems outside their control and/or are deterred from competing with established companies because of impenetrable entry barriers - most notably initial investments and operational costs. The institutional and legal relevance of this is that it consolidates, reproduces and amplifies asymmetries of power in favor of the “Big Tech”, reverting the technological trend from digital decentralization to physical infrastructure dependency and curbing innovation and market efficiency.

=== Historical Lineage ===
==== Limitation of Machine Learning ====
<p style="text-align: justify;">The AI revolution was built on the limitation of human capabilities to process large information and derive efficient results - where before there existed only facilitative tools to allow for human-enabled processes, AI now offered derivations and cognitive reasoning in moments and at large scales.</p>

<br>
==== Revert to Physical Infrastructure Dependency in the Rush to Be the Vest ====
<p style="text-align: justify;">The present AI model heavily relies on hardware in order to perform complex data processing for deep learning. The data centre models followed require larger model sizes (which require higher resource, information and capital investment) to be able to yield better results. Due to the limitation of resources and a rush to establish the best working model of an AI agent, data centres are built on a rudimentary model heavily dependent on massive amounts of electricity and fresh water.</p>

<br>
==== Regulation and Decentralization of Information ====
<p style="text-align: justify;">Regulation and influence of sovereigns and institutions has resulted in data sharing and information exchange structurally through open source and cloud computing. But, while AI models claims their selling point to be accessibility for all, their information and derivations remain in a closed server, stored for the purpose of training their AI agents to be the best. Big AI contenders attract higher consumer and information flows overall resulting in a better developed AI model, proving disadvantageous in any market where early contenders have an advantage and accumulation of power.</p>

=== Mechanisms of Path Dependence ===
<p style="text-align: justify;">Early on in the industry there is an expectation of human and capital investment from all stakeholders involved, this high cost for entry into the market and expected capital to sustain and develop a model by both consumer and enterprise makes for a high switching cost. For the AI model to run and scale there is a presumption that the rate of growth is dependent on the size of data fed, resulting in a better model and hence, more consumer usage. A closed centralised data loop is created but the requirement for storage and processing far surpasses current infrastructures capabilities - heavily relying on early market entry advantage and sustained capital investments taking away from smaller aspirants and hoarding of power within closed circles. Despite its physical dependency, AI data processing has unique requirements beyond legacy infrastructures - larger demand and production of GPUs and designed to handle immense thermal heat generated while consuming electricity beyond current capabilities. </p>

<br><p style="text-align: justify;">The AI space largely remains unregulated due to its undefined form and fluid nature, however, this has given way for unethical and morality issues regarding data privacy, exploitation of labour and environmental harms.</p>

=== Politics and Distribution ===
Concentration strengthens the values, interests, capacities, and authority of the “Big Tech” over other actors of public and private nature. In this context, Barbrook and Cameron (1) illuminate the ethos behind AI technology development while Winner (2) makes observable the power relations surrounding it: concentration is the consequence of an AI development anchored in the belief that it must be free from all institutional and normative restriction since it is inherently liberating and non-political which, in turn, provides a political/moral justification for the concentration (if technology is inherently liberating and non-political then there is no power relations to observe or question) that translates itself into power dynamics where the AI developer not only owns the AI technology but also decides how it is trained and deployed and which models the user can or cannot access.

=== Contingency and Alternatives ===
Concentration is due, first and foremost, to the fact that AI computing capacity requires massive and cutting-edge physical infrastructure - such as data centers and microprocessors - which, in turn, requires upfront investments that are unaffordable to most companies. In David’s (3) language, companies that could finance the infrastructure and position themselves first obtained self-reinforcing advantages and a “lock-in” effect in the market. In other words, concentration is the consequence of the development of AI physical infrastructure privately with no mixed or public alternatives as a public service. Therefore, there was a space for an alternative future at the end of the 1980s or the beginning of the 1990s when the privatization model of the technological infrastructure was consolidated in line with the “Californian Ideology”. In that moment governments had the chance to treat technological infrastructure as a mixed or public good rendering it universally accessible as it was the case of the Internet; this would have permitted companies of all sizes to profit from economies of scale, reducing entry barriers and fostering competition. However, this alternative path did not prevail since the Californian Ideology provided the moral/political justification to exclude governments from the development of technological infrastructure in the exact moment in which such a decision would produce structural consequences. If that path had been followed it could have contributed to collective access to economies of scale, lower entry barriers, and a more competitive AI market.

=== Governance Implication ===
Considering the above, public or mixed investment on AI physical infrastructure is a necessary governance measure for market decentralization and collective access to AI computing capacity and data. From an historical perspective, the proposed measure is not merely a corrective action for a market failure but for a contingent and ideological decision made between the late 1980s and early 1990s that produced a structural “path dependence” that now is necessary to change.
==== References ====
(1) [https://git.xpub.nl/onebigear/SI_09_reference_library/raw/commit/22bf61e1dcfed35d58fba23808212bd00584be47/Richard_Barbrook_Andy_Cameron_Californian_ideology.pdf Barbrook, R., & Cameron, A. (1996). The Californian Ideology. <em>Science as Culture</em>, <em>6</em>(1), 44-72].

(2) [https://d1wqtxts1xzle7.cloudfront.net/87708502/langdon_winner_do_artifacts_have_politics-libre.pdf?1655588533=&response-content-disposition=inline%3B+filename%3DDo_Artifacts_Have_Politics.pdf&Expires=1788890463&Signature=UBBiMqkip5MfYZbkbLNAN6aE2hyZdKkER-HFcmyxXkwIZuTnMKdOGoao8ddFr-rQ5T55A8NWUOlnZkGYmi66e7XR9h99DgTELMCrh48sXjOb8KYgA48r~SUUymt1IW~R~aOKJzitMIUHcrc6yEGKK1OuaA~mrEIVOXpUjQMYjqOd26Xmk9YJEglTkqaxeR47VCCth5FvozNQIowdfjzeu8aNHnI8V7Yur1IiXY~PGvYRcN6wBtcgpVMHZUtNaI8NjmYbp2p5TBYrPMF9MCLtvpGhPlsCfsvA8Bdm6XulGwRlksZBKWvVUQ8D~M8oLu12V39YaiETIObyJqbE7ctIcQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA Winner, L. (2017). Do artifacts have politics?. In <em>Computer ethics</em>(pp. 177-192). Routledge].

(3) [https://sjbae.pbworks.com/f/David+1985.pdf David, P. A. (1985). Clio and the Economics of QWERTY. <em>The American economic review</em>, <em>75</em>(2), 332-337].

== Centralization of AI compute and Data - by Sergey Abramov and Paul Stein ==
<div>
=== The current governance issue ===
<p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>
<br>
=== Historical lineage ===
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p>
<br><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p>
<br><p>Soon after Amazon launched AWS, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p>
<br><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>
<br>
=== Mechanism of path dependence ===
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers.</p>

<br><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>

<br>

=== Politics and distribution ===
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities.</p>
<br><p>The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
<br>
=== Contingency and Alternatives ===
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. </p>
<br><p>The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p>
<br>
=== Governance implication ===
<p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup></p>
<br><p>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. </p>
<br><p>Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p>

=== References & Notes ===
1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from 'https://epoch.ai/data-insights/hyperscalers-control-most-compute' [online resource]. Accessed 7 Sept 2026.</p>2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026) https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world accessed 7 September 2026.

3. Amazon Web Services (AWS), a cloud computing platform

4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025) https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf accessed 7 September 2026.

5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.

6. Ibid. 123.

7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.

8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025) https://texasroyaltybrokers.com/ai-clusters-by-state/ accessed 7 September 2026.

9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.

10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.

11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026) https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 accessed 7 September 2026.

12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'') https://www.nsf.gov/impacts/internet accessed 7 September 2026.

13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'') https://casrai.org/dictionary/term/national-ai-research-resource-nairr accessed 7 September 2026.

14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.

15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024) https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en accessed 7 September 2026.</div>

Centralization of Compute and Data

2026-09-08T17:49:51Z

Sergey Abramov: /* Mechanism of path dependence */

<div>

== The current governance issue ==
<p>Advanced AI currently depends on a narrow group of tech giants<sup>1</sup> and semiconductor oligopolies controlling microchips and software, hyperscale cloud infrastructure, large datasets, frontier models, and distribution platforms. This presents a governance problem where this narrow group exercises control over foundational digital infrastructure. This directly affects consumers, businesses, and academia out of AI. Legally and institutionally, this matters because it creates a profound democratic deficit: this neutralizes antitrust frameworks designed for traditional markets, often bypassing conventional copyright and data privacy protections through unchecked data harvesting thereby complicating enforcement.This ultimately might affect national sovereignty as the countries that host the buildout will shape the future of AI: who controls it, what values it contains, how it is used<sup>2</sup>.</p>

<br>

== Historical lineage ==
<p>In 1987, Taiwan Semiconductor Manufacturing Company (TSMC) was founded, establishing the world’s first semiconductor fab. Instead of designing and making its own chips, TSMC only manufactured chips designed by other companies. Today, nearly 100% of the world's most advanced AI processors are manufactured by TSMC, leaving the entire global AI compute supply chain centralized under one company.</p>

<br><p>In 2006, Nvidia launched CUDA (Compute Unified Device Architecture), a free software layer that allowed developers to use Nvidia's graphics chips (GPUs) which later became perfect for training deep neural networks. Today, this creates a near-monopoly on AI compute hardware.</p>

<br><p>Soon after Amazon launched AWS<sup>3</sup>, Microsoft and Google launched their own cloud platforms marking the structural pivot to the cloud service provider model, shifting computing from a distributed architecture (where companies bought and managed their own physical servers) to a centralized infrastructure. As training modern AI models requires tens of thousands of GPUs, today’s frontier AI companies are entirely dependent on the extremely expensive physical infrastructure of a handful of cloud hyper-scalers (Microsoft, Google, AWS, and Meta).</p>

<br><p>In 2009, Fei-Fei Li launched ImageNet, a massive database of over 14 million images, which proved that AI model performance scales directly with data volume. This created an immediate institutional advantage for consumer internet platforms (like Google and Meta), because they were already architected to collect, centralize, and store masses of user data through search engines and social networks, they possessed the unique data stack required to train high-performing AI.</p>

<br>

== Mechanism of path dependence ==
<p>The centralization of AI compute and data is a classic case of technological lock-in. When structural choices were made decades ago, they triggered powerful, self-reinforcing loops. One such loop is the switching costs and software standardisation. As developers have spent more than a decade working with CUDA, moving to a rival hardware provider (or a decentralized compute network) requires an enterprise to completely rewrite its foundational software stack. The cost of code migration, developer retraining, and potential hardware downtime force companies to stay tied to dominant hardware monopolies such as Nvidia’s CUDA. Commitments, licensing conditions, and exclusivity provisions add contractual costs. Fewer than 1% of cloud customers switch provider annually, identifying technical differentiation and data-transfer charges as important barriers<sup>4</sup>.</p>

<br><p>Network effects also matter for computing power. More customers generate more revenue, usage information, and opportunities to improve infrastructure and models. Better services attract additional users and developers, whose applications make the platform more valuable. In cloud–AI partnerships, providers may additionally obtain information about model performance and future infrastructure demand. These feedback loops strengthen firms already operating across cloud, models, software, and distribution.</p>

<br>

== Politics and distribution ==
<p>According to Langdon Winner's definition, modern AI computing infrastructure may qualify as an "inherently political technology" because its large scale and significant financial requirements not only lend themselves to, but necessitate,<sup>5</sup> centralized, hierarchical control. The physical siting of this infrastructure shows evidence of Winner’s second category of politicization of technologies, the ‘political arrangement’:<sup>6</sup> Of the five largest U.S. AI firms, four are headquartered in California,<sup>7</sup> while just one is (partially) domiciled in Texas. Texas, however, hosts twice the GPU clusters.<sup>8</sup> This placement geographically insulates beneficiaries of the data infrastructure boom from its negative externalities.</p>
<br><p>The specific choice of location may itself be a deliberate preference on account of Texas’ unique approach to water usage rights.<sup>9</sup> Juxtaposition of this unequal and even exploitative balance of interests with the pontificating rhetoric of the urban, information rich, beneficiaries is evocative of the internal contradictions of the Californian Ideology - the virtual artisans existence and privileged position is still based on the exploitation of a vulnerable out-group, the "information-rich" consuming the output, the "information-poor" hosting the extraction.<sup>10</sup></p>
<br>
== Contingency and Alternatives ==
<p>Profit motive and economies of scale now concentrate investment among a few hyperscalers,<sup>11</sup> but this reflects policy choice, not government incapacity: Compute infrastructure may be inherently political, but the technology is agnostic as to where the organizing authority originates, and states have historically funded infrastructure of comparable or greater scale, while facing less pressure to show short-term returns. </p>
<br><p>The clearest turning point was 1995, when the National Science Foundation retired NSFNET, the publicly funded backbone that had connected U.S. research and supercomputing centers since the mid-1980s,<sup>12</sup> handing internet infrastructure to commercial providers with no conditions attached. This was a deliberate policy decision made amid the deregulatory push of the Clinton era. The privatization that followed ushered in the private industrial dominance that persists today in the form of hyperscalers. Had a publicly governed backbone instead been scaled up, today's compute infrastructure might rest on a different foundation entirely.</p>
<br>
== Governance implication ==
<p>Despite private hyperscaler dominance, the U.S. does have a nominal public alternative: the National AI Research Resource (NAIRR), established following the National AI Initiative Act of 2020 and piloted from January 2024, which explicitly aims to democratize access to the computational resources essential to AI research.<sup>13</sup></p>
<br><p>In Europe, a comparable program is the EuroHPCJU,<sup>14</sup> an initiative intended to pool resources to build supercomputing capacity. The undertaking’s legislative basis has since been extended to include networked ‘AI factories', as well as the so-called “Gigafactories”.<sup>15</sup> The framing in language of sovereignty indicates intent to at least compete with private infrastructure initiatives. </p>
<br><p>Though government-led measures cannot reverse the centralization of compute, they can democratize access - provided they are intensified. The European Commission has “set a goal of investing more than €1 billion per year in AI research and innovation”.<sup>16</sup> While laudable in its own right, it pales in comparison to the investments of industry leaders, which are approaching, if not exceeding, trillions of Dollars.</p>

== References & Notes ==
1.Amazon, Google, Meta, Microsoft, and Oracle collectively hold an estimated 71% of the world’s cumulative AI compute as of Q4 2025, measured in H100-equivalents of computing power. <p>Luke Emberson, Josh You and Venkat Somala (2026), "Five hyperscalers now own over two-thirds of global AI compute". Published online at epoch.ai. Retrieved from 'https://epoch.ai/data-insights/hyperscalers-control-most-compute' [online resource]. Accessed 7 Sept 2026.</p>

2. Alasdair Phillips-Robins, Teddy Tawil and Sam Winter-Levy, ‘The Compute Coalition: How to Build the Future of AI in the Free World’ (Carnegie Endowment for International Peace, 8 June 2026) https://carnegieendowment.org/research/2026/06/the-compute-coalition-how-to-build-the-future-of-ai-in-the-free-world accessed 7 September 2026.

3. Amazon Web Services (AWS), a cloud computing platform

4. Competition and Markets Authority, Cloud Infrastructure Services: Final Decision Report (31 July 2025) https://assets.publishing.service.gov.uk/media/688b8891fdde2b8f73469544/final_decision_report.pdf accessed 7 September 2026.

5. Langdon Winner, 'Do Artifacts Have Politics?' (1980) 109(1) ''Daedalus'' 121, 128.

6. Ibid. 123.

7. OpenAI, Anthropic, Google DeepMind, Meta AI. xAI is an exception to the rule, domiciled between Palo Alto, CA and San Antonio, TX.

8. Texas’ 17 to California’s 8, cf. Texas Royalty Brokers, 'AI Clusters by State' (''Texas Royalty Brokers'', 2025) https://texasroyaltybrokers.com/ai-clusters-by-state/ accessed 7 September 2026.

9. The ‘rule of capture’, established in Houston & TC Railway Co v East (Tex 1904), permits landowners to extract groundwater beneath their land without liability for depleting adjacent wells or aquifers, a doctrine now exploited by hyperscale AI data centres to draw large cooling-water volumes with minimal regulatory constraint.

10. Richard Barbrook and Andy Cameron, 'The Californian Ideology' (1996) 6(1) ''Science as Culture'' 44.

11. Goldman Sachs, 'Global AI Investment Is Forecast to Exceed $1 Trillion in 2026' (''Goldman Sachs Insights'', 7 August 2026) https://www.goldmansachs.com/insights/articles/global-investment-is-forecast-to-exceed-1-trillion-in-2026 accessed 7 September 2026.

12.  National Science Foundation, 'Birth of the Commercial Internet' (''NSF Impacts'') https://www.nsf.gov/impacts/internet accessed 7 September 2026.

13. CASRAI, 'NAIRR: National AI Research Resource, Explained' (''CASRAI'') https://casrai.org/dictionary/term/national-ai-research-resource-nairr accessed 7 September 2026.

14. Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 [2021] OJ L256/1.

15. Council Regulation (EU) 2026/150 of 16 January 2026 amending Regulation (EU) 2021/1173 on establishing the European High Performance Computing Joint Undertaking [2026] OJ L 2026/150.

16. European Commission, 'Realising Europe's Digital Decade' (''European Commission'', 11 October 2024) https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/story-von-der-leyen-commission/realising-europes-digital-decade_en accessed 7 September 2026.