Political and legal aspects of BRICS cooperation in the field of artificial intelligence: Towards the development of an alternative regulatory approach

Full Text

Introduction

Contemporary discourse on global artificial intelligence (AI) governance has become trapped in a binary opposition between two main approaches. On the one hand, the European Union’s risk-based approach (Kunin & Uporova, 2019, p. 59), as embodied in its AI Act (Belov, 2024), advances a statist model of regulation through stringent normative frameworks. On the other hand, a liberal model associated with the United States prevails, emphasizing industry standards and corporate self-regulation (Nusratullin, 2007, p. 54). The literature often cites the ‘California effect’ and the ‘Brussels effect’ as illustrations of how policies adopted in the EU and California’s Silicon Valley exert influence beyond their respective jurisdictions (Bocken, Coffay & Dalhammar, 2025). However, growing tensions over technological sovereignty and the emergence of data as a new resource for geopolitical leverage have revealed the limitations of both models for many non-Western countries.

In this context, the BRICS countries are not merely passive observers, but active participants in shaping the global technological order. Their cooperation on artificial intelligence represents a unique politico-legal phenomenon: it is neither an attempt to create another supranational regulatory body, nor a pursuit of a competitive, fragmented ecosystem. This study proceeds from the hypothesis that BRICS is endeavoring to construct an alternative regulatory model grounded in the principle of an ‘overlapping consensus,’ where common objectives are achieved not through unification, but through the harmonization of disparate national approaches.

The aim of this research is to identify and analyze the political determinants and emerging legal contours of this nascent model of cooperation. To achieve this aim, the following tasks are set:

– to conduct a comparative analysis of the national strategies and regulatory approaches to AI of the BRICS member states,

– to identify common political motives and potential zones of conflict among the member states in the field of AI,

– to analyze the evolution of cooperative formats within BRICS, from initial discussions to concrete initiatives,

– to model the potential trajectories for the development of the legal framework for cooperation.

The scientific novelty of this research lies in applying the concept of ‘overlapping consensus’ to the analysis of regulatory processes within BRICS, thereby allowing us to overcome the methodological pitfall of treating the alliance as a monolithic entity.

Research Methodology

To address the research tasks, this study employs a set of complementary methods that combine tools from political science and law. Discourse analysis of BRICS documents was conducted based on several parameters, including the frequency and context of key concepts (‘sovereignty,’ ‘inclusiveness,’ ‘fairness’) in the final declarations of BRICS summits from 2020 to 2025, in order to trace the evolution of formulations from general principles to concrete commitments. The case-study method was applied to analyze Russian-Chinese cooperation in the field of AI. A comparative legal analysis was also carried out on the national strategies for AI development and regulation in Brazil, Russia, India, the People’s Republic of China (PRC), and the Republic of South Africa (RSA).

In this study, ‘overlapping consensus’ is understood as an agreement on general principles that is achieved because each participant imbues them with content compatible with their own interests (Rawls, 1987).

‘Digital sovereignty’ refers to the capacity of a state to maintain control over critical digital assets and technologies.

‘Soft law’ denotes normative instruments (declarations, memoranda, codes of ethics) that are not legally binding (Trubek & Trubek, 2005).

The comprehensive approach proposed by the authors in this study provides a thorough assessment of the specific features of the emerging regulatory model, which operates primarily within the domain of ‘soft law’ (Kudryashov, 2013)—that is, through codes of ethics, general principles, and joint declarations, rather than through rigid mandatory norms.

Literature Review

In the face of current geopolitical challenges, research on the role of BRICS in transforming the international architecture becomes particularly relevant. This study is situated at the intersection of several actively developing academic fields: the study of BRICS as an institution of international cooperation, research into global AI governance, and a comparative analysis of digital policies.

A significant body of literature is devoted to analyzing BRICS as a phenomenon in international relations (Wang & Long, 2024; Denisova, 2022; Glebov & Agonnoude, 2023; Cooper, 2022). These works address the stages of formation and the factors behind the unification of member states. As D.E. Denisova notes, BRICS is fundamentally an alliance of equal and independent states that not only actively participate in processes of global integration but also jointly confront contemporary threats, while building mutually beneficial relations with both traditional partners and Western countries (Denisova, 2022, p. 40).

As D.E. Denisova points out, the BRICS is essentially a group of equal and independent states that actively participate in global integration processes, while jointly confronting contemporary threats and building mutually beneficial relationships with both traditional partners and Western countries.

The second body of literature addresses the emergence of a global regulatory regime for AI. A dominant narrative, brilliantly articulated in A. Bradford’s “The Brussels Effect,” describes the confrontation between two models: the EU’s precautionary principle, based on stringent risk regulation, and the liberal, market-oriented model of the United States (Bradford, 2020). The role of China in this system is often portrayed as a ‘techno-authoritarian’ alternative, as explored in detail in a number of articles by Chinese authors (Zeng, 2020; Wu, Huang & Gong, 2020). This  tripartite perspective (EU—USA—China) effectively excludes other players from the ranks of norm-setters, relegating them to a follower role. However, in recent years, a critical trend has gained momentum, represented in the work of H. Roberts and colleagues (Roberts et al., 2022), which points to the need to take into account the voices and approaches of Global South countries on this issue. In parallel, research on highly specialized legal aspects of AI regulation is actively developing, particularly with regard to issues of liability for the actions of autonomous systems (Duffy, 2023). In the Russian literature, there are also works on global AI governance (Nalivkina, 2025; Belov, 2024; Vasilyev, 2023; Vykhodets, 2022). Thus, in the literature on global AI governance, there is a clear gap related to the analysis of BRICS as a collective actor capable of offering a normative alternative.

The third body of research includes works analyzing the policies of individual BRICS member states in the field of AI. The Russian approach, focused on ‘digital sovereignty’ and import substitution, has been extensively studied in detail in a number of publications (Karev & Chertykovtcev, 2025; Bolgov, 2024; Zhurkov, 2023). The Chinese model of statist control and support for national ‘champions’ has been analyzed in the research of Chinese scholars (Cheng & Zeng, 2023; Wu, Huang & Gong, 2020). India’s “AI for All” approach is examined in the works of A. Shaji George (2024) and K. Roy and colleagues (Roy et al., 2024), while Brazilian and South African ethical initiatives are analyzed in the studies of G. de Souza^[1] and E. Ormond (2023), respectively. These studies provide valuable empirical data; however, they are isolated and country-specific in nature. There is virtually no comparative work that systematically juxtaposes these approaches, identifying points of convergence and systemic contradictions. The question of how these disparate national strategies can be reconciled within a single multilateral framework has not yet been raised in the literature.

Finally, there is a body of theoretical literature that provides tools for analyzing complex governance regimes. John Rawls’ concept of ‘overlapping consensus’ (Rawls, 1987), when applied to international relations, enables the analysis of cooperation among actors with different value systems. The theory of ‘soft law,’ developed in the works of D. Trubek (Trubek & Trubek, 2005) and G. Shaffer (Shaffer & Pollack, 2010), provides an appropriate framework for understanding regulatory processes in conditions of institutional weakness and heterogeneity of interests. However, applying this theoretical apparatus specifically to BRICS cooperation in the field of AI is novel.

Thus, the literature review has identified the following gaps:

– the lack of studies specifically focused on cooperation in the field of AI as a key element of the BRICS technological agenda,

– the predominance of bipolar or tripolar paradigms in research on global AI governance, which ignore the potential of BRICS as a normative actor,

– the lack of comparative work synthesizing analyses of the national strategies of member states to identify the foundations for multilateral cooperation,

– the absence of application of the theoretical framework of ‘overlapping consensus’ and ‘soft law’ to this specific case.

The contribution of this study is to fill these gaps by conducting a comprehensive analysis of the politico-legal aspects of BRICS cooperation in the field of AI. The study proposes a theoretical framework that enables the analysis of BRICS not as a monolith, but as a platform for creating a hybrid regulatory regime, based on the harmonization of disparate national approaches to achieve common strategic objectives on the global stage.

Theoretical Foundations of the Study of Artificial Intelligence as an Object of International Cooperation

In contemporary academic discourse, AI has ceased to be a purely technical term describing a set of algorithms and computing power. It has transformed into a complex socio-political construct defining a new paradigm of global development. The nature of AI in the 21st century is twofold: on the one hand, it is a tool for process optimization and innovation generation (“weak AI,” or narrow AI, focused on solving highly specialized tasks); on the other hand, it is a potential agent of fundamental change in the structure of society, the economy, and international relations (“strong AI,” or artificial general intelligence—a hypothetical system surpassing human intellect) (Bory, Natale & Katzenbach, 2024). This duality predetermines its political significance. While AI development was initially concentrated within the corporate environment of technology giants, today it has become a priority of national strategies. AI has evolved into a “dual-use technology,” blurring the lines between civilian and military applications (Vaynman & Volpe, 2023). Its development is directly linked to concepts such as “digital sovereignty,” “technological autarky,” and “information security,” making it a key element of national power and competitive advantage on the global stage.

The classification of AI technologies can be carried out according to various criteria: by level of autonomy (ranging from decision-support systems to fully autonomous systems), by type of task (computer vision, natural language processing, predictive analytics, etc.), by degree of societal impact (Kukshev, 2020), and others. From a politico-economic perspective, the most relevant classification is one based on access to critical resources:

1. Data as raw material: technologies requiring massive volumes of data (Big Data) for model training. This gives rise to policies such as “data nationalism” or “digital nationalism” (Gorian, 2022) and the regulation of cross-border data flows (data localization).
2. Computing power as infrastructure: high-performance computing (HPC) and quantum computers as the foundation for complex models. This fuels the race for supercomputing supremacy and investment in relevant infrastructure.
3. Talent and knowledge as capital: the competition for highly qualified specialists leads to adjustments in states’ migration and educational policies.
4. Algorithms as instruments of power: the capacity of algorithms to influence public opinion, allocate resources, and make automated decisions concentrates considerable power in the hands of their developers and regulators.

It follows that each of these levels becomes an arena for political competition and cooperation, shaping the development trajectories of countries and interstate blocs such as BRICS.

At present, the international community is at an early stage of forming an international legal regime for AI regulation. This process involves intense competition between various regulatory approaches: from stringent, precautionary regulation oriented towards human rights (the EU model, enshrined in the AI Act),^[2] to more liberal, innovation-oriented models (such as those proposed by the US and China). In our view, the key principles around which the global discourse on AI is coalescing are:

– transparency and explainability (the requirement that algorithms should be understandable to humans),

– fairness and non-discrimination (preventing algorithmic bias and discrimination),

– accountability and responsibility (determining the subject of responsibility for the actions of autonomous systems),

– privacy and security (protection of personal data and resilience of systems to cyberattacks).

These principles cover the main problematic aspects of AI application in today’s world. Determining the subject of responsibility for the actions of autonomous systems is a primary task across all spheres of activity, including the financial sector.

For instance, while discussing the prospects of economic digitalization at the plenary session of the 10th Eastern Economic Forum, Russian President V.V. Putin touched upon the potential use of AI technologies in determining one of the most important economic indicators—the key interest rate, noting the promise of integrating AI into the decision-making process of even such a critically important structure as the Central Bank. At the same time, the President emphasized that the final choice must remain the prerogative of a human being, who must bear responsibility for the decisions made.^[3]

As E.F. Cooper rightly points out, “BRICS is above all a clear testament to the cardinal, albeit uneven, changes in the global system in the 21st century” (Cooper, 2022, p. 44). The process of establishing international regulatory mechanisms demonstrates a multi-level structure, encompassing global, regional, and bilateral dimensions of interaction (Vasilyev, 2023, p. 75). The absence of a universal international convention on AI is compensated by “proliferating” initiatives at the level of ‘soft law’ (Podolsky & Olshansky, 2025). This includes the recommendations of the Organisation for Economic Co-operation and Development (OECD), the United Nations Educational, Scientific and Cultural Organization (UNESCO), the G20, as well as industry standards from IEEE^[4] and others. This creates a complex, fragmented regulatory mosaic, in which BRICS countries are forced to maneuver, striving, on the one hand, not to find themselves on the periphery of emerging standards, and on the other hand, to defend their own model of digital development and sovereignty.

A key event of 2025 was the discussion of global AI regulation at the G20 summit in Johannesburg (South Africa). Indian Prime Minister Narendra Modi called for the formation of a “Global AI Compact,”^[5] based on principles of transparency, human control, and safety-by-design.^[6] This initiative, supported by a number of Global South countries, reflects a demand for the institutionalization of ‘soft law’ at the global level. In furtherance of this agenda, the India AI Impact Summit 2026 was held in New Delhi in February 2026, demonstrating the consolidation of efforts by a number of countries to develop universal principles for AI governance.^[7]

On the practical plane, the Institute of Electrical and Electronics Engineers (IEEE) remains a key player. In January 2026, the IEEE Standards Association launched the IEEE CertifAIEd certification program, offering two types of certifications: for professionals (professional certification in ethical AI) and for products (confirmation that products meet standards of transparency, accountability, and non-discrimination).^[8] This initiative creates a mechanism for verifying compliance with the principles of ‘soft law,’ which is particularly significant for BRICS countries seeking to develop their own standards without directly borrowing Western models. Standardization is thus becoming a flexible regulatory tool that allows for a balance between innovation and security.

The theoretical framework of this study assumes an examination of BRICS countries’ policies in the field of AI through the prism of their struggle for technological sovereignty, their adaptation to a fragmented global regulatory landscape, and efforts to establish alternative centers of influence in the sphere of norm-setting for future technologies.

Current State of BRICS Cooperation in the Field of AI

Phase of Formation (2009–2015): From Economic Cooperation to the Technological Agenda

Historically, technological cooperation within BRICS has emerged as a derivative of economic and financial initiatives. As A.V. Shelepov notes, the creation of BRICS was associated from the very beginning with the task of enhancing the role of emerging market economies in the international monetary and financial system, which made financial stabilization one of the key priorities of the group (Shelepov, 2025, p. 159). As early as the 2012 summit in Delhi, a Declaration was adopted that included a separate section on scientific and technical cooperation, which for the first time mentioned the need for joint research in the field of high technologies. In paragraph 43 of the Delhi Declaration, the BRICS countries expressed their support for “the process of cooperation both in priority areas (food, pharmaceuticals, healthcare, energy) and in the field of fundamental research in new interdisciplinary areas (nanotechnologies, biotechnologies, advanced materials, etc.).”^[9]

A turning point was the summit in Ufa (2015), which approved the BRICS Economic Partnership Strategy, which included a separate section on innovation and the digital economy. In 2015, a Memorandum of Understanding and Cooperation in the field of science, technology and innovation was signed.^[10] By that time, a conceptual shift had occurred within BRICS from purely economic cooperation towards an awareness of technological sovereignty as a key element of a multipolar world architecture (Potaptseva & Akberdina, 2023).

Institution-Building Phase (2016–2023): Creation of Specialized Structures

This stage is characterized by the establishment of specialized institutions for cooperation and a clear shift from a purely economic to a technological agenda. The following were established:

– the Working Group on Trade and Economy (demonstrating the beginning of the systematization of cooperation and its elevation to an official intergovernmental level),

– the BRICS Digital Forum (indicating the separation of the digital/technological agenda into a distinct, significant area of interaction beyond purely economic issues),

– the BRICS Young Scientists Forum (evidencing a forward-looking approach and an understanding of the need to build shared human capital and expert capacity).

Of particular significance was the creation of the BRICS Working Group on Security in the Use of Information and Communication Technologies (ICT), which became the first specialized structure addressing AI-related issues directly. The establishment of the Working Group on ICT Security signified the elevation of cooperation among BRICS member states to a level of critical, sensitive topics related to sovereignty and security, representing the highest form of trust and institutionalization.

The development of the conceptual framework of “BRICS digital sovereignty” (Karev & Chertykovtcev, 2025) has been accompanied by the establishment of specialized institutions, yet their mandates have remained limited.

Contemporary Phase (2021–2025):  AI as a Cooperation Priority

At the BRICS summit in New Delhi (2021), a Declaration was adopted that contained a separate section on cooperation in the field of artificial intelligence. For the first time, the document (para. 36) proclaimed the need to develop ethical principles for AI, approved an initiative to establish a network of centers of excellence in the field of AI, and emphasized the importance of joint research.^[11]

An important outcome of the 2022 BRICS XIV Summit was the adoption of the Beijing Declaration, which pays special attention to the problems and ethical challenges posed by the development of artificial intelligence. Paragraph 57 initiates the creation of a unified mechanism for interaction among member states to overcome existing concerns, including the exchange of best practices and joint research. It is rightly noted that this will make it possible to develop a consolidated approach to regulating AI technologies, aimed at their ethical and responsible application.^[12]

At the summit in Johannesburg (2023), the BRICS Roadmap for ICT Cooperation (para. 24) was approved, which includes the creation of a working group on the harmonization of standards.^[13]

By 2023, BRICS cooperation in the field of AI had acquired a systemic character, with clear institutional mechanisms and a roadmap for implementation. Thus, BRICS has the potential to form an alternative platform for technological cooperation. A key element of its positioning is the promotion of a regulatory model based on the principles of fairness, understood as equality in global AI governance and non‑discriminatory access to technologies, and inclusiveness, which implies the participation of a wide range of Global South states and non-state actors in norm-setting. This conceptual framework stands in direct opposition to the approach whereby standards are developed by a narrow group of technologically advanced countries.

The final stage of the XVII BRICS Summit culminated in an important achievement—the adoption of the Rio de Janeiro Declaration, as well as the agreement on a principled leaders’ statement on global AI regulation.^[14] The document states that the effective development of the digital economy and the introduction of advanced technologies, including artificial intelligence, are impossible without the creation of a system of comprehensive and equitable data governance, which is critically important for developing countries. It also highlights “the leading role of the UN in facilitating constructive dialogue to develop a common understanding of security in the use of ICTs and to discuss the development of a universal legal framework in this area, as well as in the further elaboration and implementation of agreed universally recognized norms, rules and principles of responsible state behavior in the use of ICTs.”^[15]

Despite the achievements, such as the establishment of institutional infrastructure, the launch of joint research projects, and the agreement on basic principles of cooperation, persistent problems remain. These include, in particular, the different levels of technological development among member states, the competing interests of national AI ecosystems, and divergent regulatory approaches. It can be concluded that the main achievement of BRICS has been the creation of a common platform for dialogue, allowing positions to be coordinated on key issues of technological development. However, a gap persists between declaratory documents and real mechanisms of cooperation.

Despite the progress that has been made, including the establishment of institutional infrastructure, the initiation of joint research projects, and the agreement on basic principles of cooperation, there are still several persistent challenges. These include, among other things, the varying levels of technological advancement among member states, conflicting interests of national AI ecosystems, and differing regulatory approaches. It can be said that the main accomplishment of BRICS is the establishment of a common forum for dialogue, which allows for the coordination of positions on key issues related to technological development. Nevertheless, there is still a gap between the declared intentions and the actual mechanisms of cooperation.

Discourse Analysis and Cooperation Practice: Identifying the ‘Overlapping Consensus’

Evolution of BRICS Conceptual Apparatus

The analysis of BRICS summit declarations for 2020–2025 revealed a consistent evolution of the conceptual apparatus. To ensure reliability and verifiability of the study, a frequency analysis of key concepts was conducted on a corpus of official documents, including the final declarations of the summits in Moscow (2020), New Delhi (2021), Beijing (2022), Johannesburg (2023), Kazan (2024), and Rio de Janeiro (2025). The texts were processed manually with selective inter-coder reliability checks. The following key lexemes and their derivatives were identified: ‘sovereignty,’ ‘inclusiveness,’ and ‘fairness.’

The direct use of these lexemes and their derivatives was in both the general and immediate contexts of discussions on artificial intelligence, digital economy, and technological cooperation (Table 1).

Thus, in 2020, BRICS rhetoric demonstrated a clear gap between general politico-legal principles (where sovereignty was supreme) and the specific technological agenda. While inclusiveness and fairness were part of the lexicon, they had not yet been rethought and operationalized in relation to the challenges of the digital age, such as cybersecurity, the digital divide, or AI regulation. This year can be characterized as a stage when technological cooperation was discussed more in practical than in value‑oriented terms.

Table 1. Evolution of Key Concepts in BRICS Documents, 2020–2025

Year	Key concept	Mentions in Targeted Context (AI, Digital Economy, Technology)	Mentions in General Context (Entire Document)	Key Semantic Accents
2020 (Moscow)	Sovereignty	0	10 (para. 3, 5, 6, 23, 26, 29, 33, 34, 52, 67)	Fundamental principle: emphasis on sovereign equality (para. 5), territorial integrity (paras. 5, 23), non-interference (para. 5). Support for sovereignty of Syria (para. 23), Libya (para. 34), Iraq (para. 26)
	Inclusiveness	1 (para. 66—promoting inclusive economic growth)	11 (paras. 5, 6, 26, 27, 29, 34, 47, 53, 66, 76, 90)	Quality of systems and processes: inclusive international system (para. 5), UN reform for greater inclusiveness (para. 6), inclusive negotiation process (paras. 27, 34), inclusive economic growth (paras. 47, 66)
	Fairness	0	5 (paras. 5, 12, 23, 24, 83)	Ethical imperative: fair international system (para. 5), fair access to vaccines (para. 12), fair conflict resolution (para. 24), fair distribution of benefits from genetic resources (para. 83)
2021 (New Delhi)	Sovereignty	1 (para. 27—ensuring state security and protecting their interests, ICT context)	4 (paras. 2, 14, 22, 27)	Cornerstone: sovereign equality and territorial integrity as the foundation of the international system (para. 2), non-interference (para. 22), protection of sovereignty in cyberspace (para. 27)
	Inclusiveness	2 (para. 36—sustainable and inclusive recovery; para. 48—formation of a more sustainable and inclusive tourism sector)	9 (paras. 2, 14, 18, 23, 36, 42, 43, 48, 51)	Cross-cutting principle: inclusive international system (para. 2), inclusive intra-Afghan settlement (para. 23), inclusive labor markets (para. 43), inclusive economic growth (para. 51)
	Fairness	1 (para. 36—ensuring fair and comprehensive access to digital resources)	3 (paras. 14, 36, 49)	Focus on equality: fair access to digital resources (para. 36), fair access to global public goods (para. 14), fair and equal treatment of human rights (para. 49)
2022 (Beijing)	Sovereignty	0	5 (paras. 5, 6, 21, 23, 53)	Cornerstone and protection from interference: respect for sovereignty and territorial integrity (paras. 5, 21), special emphasis on Afghanistan’s sovereignty (para. 23). Also mentioned in the context of nationally determined contributions on climate (para. 53)
	Inclusiveness	2 (para. 38—inclusive recovery, inclusive development; para. 57 — marginalized and vulnerable population groups)	14 (paras. 2, 4, 5, 6, 10, 11, 23, 37, 38, 44, 63, 64, 67, 72)	Cross-cutting principle for systems and recovery: inclusiveness as the spirit of BRICS (para. 2), quality of global governance (paras. 5, 6), trade system (para. 11), political structure of Afghanistan (para. 23), key characteristic of post-pandemic recovery (para. 38)
	Fairness	2 (para. 6—fair access to global public goods; para. 57 — fair and equal treatment of all human rights)	6 (paras. 5, 6, 9, 14, 51, 57)	Principle of access and equality: fair access to global goods (para. 6) and vaccines (para. 14), fair treatment of human rights (paras. 9, 57), foundation of international relations (spirit of fairness, para. 5)
2023 (Johannesburg)	Sovereignty	0	6 (paras. 2, 3, 5, 16, 17, 21)	Fundamental principle: sovereign equality (para. 5), support for Libya’s sovereignty (para. 16), Yemen and Syria (para. 17)
	Inclusiveness	3 (para. 33—digital economy; para. 36—“for inclusive and sustainable industrialization”; para. 31—inclusive economic recovery)	14 (paras. 1, 2, 3, 8, 17, 31, 33, 36, 38, 39, 44, 78, 86, 90)	Quality of systems and processes: inclusive multilateralism (paras. 2, 3), inclusive growth (para. 2), inclusive trading system (para. 8), inclusive negotiations (para. 17), inclusive payment systems (para. 44)
	Fairness	0	14 (paras. 2, 3, 6, 8, 9, 29, 38, 41, 53, 57, 63, 70, 74, 76)	Quality of systems and processes: fair international order (para. 2), fair trading system (para. 8), fair burden-sharing (para. 29), fair transition to a low-carbon economy (para. 57), fair standards (para. 63)
2024 (Kazan)	Sovereignty	1 (para. 54—sovereignty in the ICT sphere)	9 (paras. 3, 6, 30, 31, 34, 47, 54, 56, 92)	Fundamental principle: sovereign equality (para. 3), respect for sovereignty (paras. 6, 31, 34), sovereignty in cyberspace (para. 54), territorial integrity (paras. 31, 34)
	Inclusiveness	4 (para. 71—data governance; para. 77—digital economy; para. 80—energy transition; para. 103—tax system)	18 (paras. 3, 6, 9, 11, 12, 13, 14, 64, 65, 71, 77, 80, 87, 103, 108, 118, 122, 130)	Quality of systems and processes: inclusive growth (para. 3), inclusive multilateralism (para. 6), inclusive trading system (para. 9), inclusive recovery (para. 64), inclusive payment systems (para. 65), inclusive entrepreneurship (para. 130)
	Fairness	2 (para. 71—data governance; para. 80—energy transition)	21 (paras. 1, 3, 6, 9, 11, 15, 17, 21, 59. 60, 62, 71, 73, 80, 81, 82, 84, 87, 103, 108, 117)	Quality of systems and processes: fair world order (paras. 1, 3), fair trading system (para. 9), fair burden-sharing (para. 60), fair energy transition (paras. 80, 81), fair tax system (para. 103)
2025 (Rio de Janeiro)	Sovereignty	2 (para. 16—respect for state sovereignty in the context of global AI governance; para. 62—the technical scope of space telecommunication systems must not exceed the limits of state sovereignty)	13 (paras. 2, 5, 16, 27, 28, 29, 31, 59, 62, 66, 92, 101, 124)	Fundamental and indivisible principle: pervades all spheres—from politics (sovereign equality of member states, para. 2) to the digital space (para. 59) and the economy (national sovereignty over data, para. 66). Strictly linked to territorial integrity (paras. 29, 31, 101) and the right of the state to control activities on its territory (para. 62)
	Inclusiveness	3 (para. 16—global AI governance must consider the needs of all countries, ensuring inclusive international cooperation; para. 40—commitment to creating an open, secure, and inclusive ICT environment; para. 58—the importance of forming an inclusive and secure environment for the development of the digital economy)	32 (paras. 1, 2, 4, 5, 8, 10, 13, 15, 16, 29, 39, 40, 43, 45, 54, 55, 58, 60, 64, 69, 76, 82, 84, 89, 90, 91, 106, 112, 114, 117, 118, 124)	Cross-cutting principle for the quality of all systems and processes: describes the nature of the desired international order, economic growth, financial systems, and social policy. Key contexts: inclusive governance (paras. 1, 5), inclusive growth (para. 2), inclusive trading system  (para. 13), inclusive labor markets (para. 114). Traditional spheres (politics) and new technological spheres (AI, digital economy) are placed on an equal footing in the need to follow this principle
	Fairness	1 (para. 121—the need to ensure fair opportunities for women’s development in a sustainable and digital economy)	37 (paras. 2, 5, 8, 10, 13, 15, 16, 26, 27, 45, 62, 65, 69, 71, 76, 77, 81, 83, 84, 89, 90, 91, 92, 93, 98, 100, 101, 103, 104, 106, 107, 111, 114, 119, 120,   121, 122)	Fundamental principle for criticizing the current world order and formulating a vision for the future: emphasis on restoring fairness: in representation  (paras. 5, 10), trade (para. 13), access to resources and goods (healthcare—para. 15, finance—para. 83, genetic resources— para. 93), conflict resolution (para. 27).  Demand for fair globalization and an international tax system (para. 76)

Source: compiled by B.V.F. Agonnoude, V.A. Glebov and A.A. Maslov based on content analysis of BRICS summit declarations: Moscow Declaration of the XII BRICS Summit. November 17, 2020 // President of Russia. (In Russian). URL: http://kremlin.ru/supplement/5581 (accessed: 23.09.2025); New Delhi Declaration of the XIII BRICS Summit // BRICS Information Portal. (In Russian). URL: https://infobrics.org/files/country/russia/documents/2021/2021_09_09_India_New-Dehli__XIII_Sammit_Declaration_ru.pdf (accessed: 23.09.2025); Beijing Declaration of the XIV BRICS Summit. June 23, 2022 // President of Russia. (In Russian). URL: http://kremlin.ru/supplement/5819 (accessed: 23.09.2025); Second Johannesburg Declaration of the BRICS Countries. August 24, 2023 // President of Russia. (In Russian). URL: http://kremlin.ru/events/president/news/72103 (accessed: 23.09.2025); XVI BRICS Summit, Kazan Declaration. October 23, 2024 // President of Russia. (In Russian). URL: http://static.kremlin.ru/media/events/files/ru/MUCfWDg0QRs3xfMUiCAmF3LEh02OL3Hk.pdf (accessed: 23.09.2025); Rio de Janeiro Declaration: Strengthening Cooperation of the Global South for More Inclusive and Sustainable Governance, Rio de Janeiro, Brazil, July 6, 2025 // President of Russia. (In Russian). URL: http://static.kremlin.ru/media/events/files/ru/gvTArkWauqwuryk9xzLt3HuuI7EBmqrC.pdf (accessed: 23.09.2025).

In 2021, under India’s chairmanship, there was a clear trend towards a closer linkage of key political principles (sovereignty, inclusiveness, fairness) with the current agenda of digital transformation and technological development. This indicates a growing depth of integration of these concepts into practical areas of cooperation.

The analysis of the 2022 document reveals a consistent shift in BRICS rhetoric from an emphasis on defending sovereignty (a defensive/reactive stance) to promoting inclusiveness and fairness as principles for building an alternative world order and ensuring access to the benefits of globalization (a proactive/constructive stance).

In 2023, BRICS demonstrates a shift from classical political rhetoric on sovereignty to a more pragmatic and multi‑level agenda, where inclusiveness and fairness become central elements of global governance, economic cooperation, and sustainable development.

In 2024, the BRICS countries deepened and specified the group’s key principles. Sovereignty was adapted to the digital era, and inclusiveness became a cross‑cutting principle in economic, technological, and climate agendas. Fairness, in turn, was transformed into a demand for fair rules in new areas such as data, energy, and taxation. This reflects BRICS’s evolution from a declaration of principles towards the creation of an alternative architecture of global governance, based on sovereignty, inclusiveness, and fairness.

In 2025, BRICS consolidation is observed around a project to create a parallel, decentralized system of governance, where members’ sovereignty is protected, access to growth and resources is inclusive, and rules are based on principles of fairness rather than historical dominance. BRICS is moving from criticizing the architecture of global governance to building its own alternative system, as confirmed by the group’s first‑ever adoption of the “Leaders’ Statement on Global Artificial Intelligence Governance.”

To visually demonstrate the identified trends, the data from Table 1 were visualized as line graphs (Figures 1–3).

Figure 1. Dynamics of Mentions of the Concept ‘Sovereignty’ in BRICS Documents, 2020–2025
Source: compiled by B.V.F. Agonnoude, V.A. Glebov and A.A. Maslov.

Figure 2. Dynamics of Mentions of the Concept ‘Inclusiveness’ in BRICS Documents, 2020–2025
Source: compiled by B.V.F. Agonnoude, V.A. Glebov and A.A. Maslov.

Figure 3. Dynamics of Mentions of the Concept ‘Fairness’ in BRICS Documents, 2020–2025
Source: compiled by B.V.F. Agonnoude, V.A. Glebov and A.A. Maslov.

Figure 1 reflects the dynamics of mentions of the concept ‘sovereignty’ in targeted contexts: single appearances in 2021 and 2024 (one mention each) are followed by an increase to two mentions in 2025. The absence of mentions in 2020, 2022, and 2023 indicates that sovereignty long remained a predominantly political concept. Its inclusion in the technological agenda in 2025, with a peak of two mentions, confirms the thesis of the ‘digitalization’ of the sovereignty discourse and its adaptation to the issues of artificial intelligence regulation.

Figure 2 demonstrates a steady increase in the number of mentions of the concept ‘inclusiveness’ in targeted contexts: from one in 2020 to two in 2021–2022, then a gradual rise to three mentions in 2023 and four in 2024, which is the absolute maximum for the period under study. In 2025, the figure declines to three mentions but remains at a high level. Such dynamics indicate the transformation of inclusiveness from a general principle into a key element of the BRICS technological agenda, especially during the period of active discussion of digital transformation.

Figure 3 shows the non-linear evolution of the concept ‘fairness’: after appearing in 2021 with one mention and increasing to two mentions in 2022, the concept completely disappears from the targeted context in 2023. It returns in 2024, reaching a maximum of two mentions, after which in 2025 the figure declines to one mention. This “pulsating” dynamic indicates the instability of the inclusion of ethical language in the BRICS technological discourse; however, the very fact of the concept’s presence over the last two years suggests the emergence of a demand for fair AI regulation.

Case Study: Russian-Chinese Cooperation in the Field of AI

The establishment of the Russian-Chinese working group on experience exchange, regulation, and the application of AI^[16] in September 2025 marks a transition from declarative bilateral dialogue to a practical approach of joint design. Unlike previous formats, the new structure, according to its mandate, is focused on addressing three key tasks that reveal the logic of the ‘overlapping consensus’:

1) harmonization of standards for the compatibility of national AI ecosystems without their unification,

2) coordination of R&D projects in strategically sensitive areas, such as large language models,

3) creation of joint ‘regulatory sandboxes’ for testing mechanisms for the cross-border use of artificial intelligence.

Thus, the working group acts not merely as a new institution, but as a fully-fledged instrument for operationalizing technological sovereignty through selective cooperation.

The public statement by Russian Deputy Prime Minister D. Grigorenko about a “solid foundation for large-scale projects”^[17] should be considered as a narrative constructing political reality. The emphasis on “own large language models” and “technological sovereignty”^[18] serves several functions:

–  foreign policy: positioning the Russian-Chinese alliance as an independent pole in global AI competition,

– domestic policy: legitimizing the course towards import substitution in the IT sector,

–   pragmatic: identifying an area of synergy—combining Russian competencies in fundamental research with Chinese capabilities in scaling up.

This narrative, however, requires verification through analysis of specific projects. The initial steps, as noted by A.D. Nalivkina, are concentrated in less politicized areas, such as medical AI, where the risks of conflicts of interest are lower and the potential for quick returns is higher (Nalivkina, 2025, p. 155). The position of D.S. Krylov seems justified in this context: “Unlike traditional forms of diplomatic and military-political interaction, technological cooperation allows a state to implement a ‘soft power’ strategy in the digital dimension, combining flexibility, pragmatism, and diversification of international partnerships” (Krylov, 2025, p. 59).

Analysis of National Regulatory Landscapes of BRICS Founding Member States

A comparative analysis of the national approaches of BRICS founding member states to AI regulation (Table 2) revealed significant regulatory heterogeneity among these approaches, which initially created challenges for developing a unified policy for the group in this area. The expansion of its membership through the inclusion of new members further complicates this picture, requiring consideration of new regulatory models and strategic priorities when formulating a common BRICS approach to digital regulation.

Table 2. Comparative Analysis of Regulatory Approaches of BRICS Founding Member States in the Field of AI

Country	Dominant Regulatory Model	Key Policy Accents	Attitude to ‘Hard Law’
China	Statist-mercantilist	Security, control, technology export	Stringent regulation with emphasis on content control
Russia	Sovereign digitalization	Import substitution, critical information infrastructure security	Experimental legal regimes (‘regulatory sandboxes’)
India	“AI for Development” (AI for All)	Startup ecosystem, inclusiveness	Liberal regulation, stimulation of innovation
Brazil	Liberal-humanistic	Human rights, ethics, inequality reduction	Balance between innovation and rights protection
South Africa	Developmental	Economic growth, public welfare	Gradual formation of regulatory framework

Source: compiled by B.V.F. Agonnoude, V.A. Glebo, and A.A. Maslov based on analysis of regulatory approaches of BRICS founding states: Artificial Intelligence Development Plan // Foundation for Law & International Relations. July 8, 2017. URL: https://flia.org/notice-state-council-issuing-new-generation-artificial-intelligence-development-plan/ (accessed: 23.09.2025); National Strategy for the Development of Artificial Intelligence for the Period up to 2030. Approved by Decree of the President of the Russian Federation No. 490 of October 10, 2019 // President of Russia. (In Russian). URL: http://static.kremlin.ru/media/events/files/ru/AH4x6HgKWANwVtMOfPDhcbRpvd1HCCsv.pdf (accessed: 23.09.2025); India’s National Strategy for Artificial Intelligence // DigWatch. June 2018. URL: https://dig.watch/resource/indias-national-strategy-for-artificial-intelligence (accessed: 23.09.2025); Brazilian Strategy for Artificial Intelligence // OECD. April 15, 2025. URL: https://www.oecd.org/en/publications/access-to-public-research-data-toolkit_a12e8998-en/brazilian-strategy-for-artificial-intelligence_936c5793-en.html (accessed: 23.09.2025); South Africa National Artificial Intelligence Policy Framework // Fairbridges. August 2024. URL: https://fwblaw.co.za/wp-content/uploads/2024/10/South-Africa-National-AI-Policy-Framework-1.pdf (accessed: 23.09.2025).

China

The period 2015–2018 was marked by fundamental changes in China’s AI policy as a comprehensive state strategy was formed, supported by a series of key documents at all levels of government (Vykhodets, 2022, p. 144). In July 2017, China published its national AI strategy, the “New Generation Artificial Intelligence Development Plan” (Artificial Intelligence Development Plan, AIDP)^[19], which outlined the country’s geopolitical, fiscal, legal, and ethical goals regarding AI technologies (Roberts et al., 2022). China’s national strategy became the conceptual foundation for building a “statist-mercantilist model” of regulation.

The period from 2015 to 2018 witnessed significant changes in China’s artificial intelligence (AI) policy, as a comprehensive state strategy was developed and supported by a range of key documents across all levels of the government.

The institutional mechanisms for implementing the strategy demonstrate a systemic approach: AIDP has established a three-tier system of goals—from fundamental research to technology commercialization, with coordination between state scientific institutions and private technology companies becoming a key instrument. Regulation, as in the case of the Interim Measures on Generative AI (Migliorini, 2024), combines strict content security requirements with the creation of favorable conditions for national ‘champions’ (Baidu, Alibaba).

Technological sovereignty and security have become an absolute priority for the PRC, which is reflected in two parallel processes. On the one hand, AI is seen as a tool to enhance the capabilities of the security sector and manage social processes (Zeng, 2020); on the other hand, research groups from Chinese companies and universities are actively developing federated learning technologies (federated learning— a machine learning method that allows training models on decentralized data without centralizing it) (Wu, Huang & Gong, 2020, p. 303), which indicates a search for balance between innovation and data control.

China’s geopolitical positioning in the field of AI is characterized by an open aspiration for global leadership by 2030 (Cheng & Zeng, 2023). This implies not only technological dominance, but also active participation in shaping global standards and ethical norms, which corresponds to China’s overall strategy of becoming a “world power” in the field of AI.

Thus, the Chinese model to AI regulation represents a statist-mercantilist pole within BRICS, demonstrating a path of technological development through a rigid vertical of state governance combined with aggressive support for national ‘champions.’ The strategy, aimed at achieving global leadership by 2030, makes China a key driver of the group’s technological transformation. Within BRICS, China acts as an alternative center of technological power, offering a model based on the principles of technological sovereignty and strategic planning, which creates a counterbalance to Western approaches to AI regulation. The success of the Chinese model largely determines BRICS’s potential in forming a competitive multipolar architecture of global artificial intelligence governance.

Russian Federation

The Russian Federation is consistently developing a model of technological development that can be characterized as “sovereign digitalization.” Its conceptual foundation is the National Strategy for the Development of Artificial Intelligence until 2030, which enshrines as a key principle the “need to ensure technological sovereignty” of the Russian Federation in the field of AI.^[20]

The institutional and legal mechanisms for implementing this model include several levels. At the systemic level, Federal Law No. 187-FZ “On the Security of Critical Information Infrastructure” operates, setting the framework for the protection of critical information infrastructure (CII)^[21] facilities, which is an unconditional priority of state policy. At the same time, to stimulate innovation, Federal Law No. 258-FZ “On Experimental Legal Regimes in the Sphere of Digital Innovations” was adopted, creating legal “sandboxes” for testing AI-based solutions in limited conditions. As A.A. Zhurkov rightly notes, “For the period 2021–2030, the Russian government identified the issue of legal regulation of artificial intelligence activities as an important area of scientific research” (Zhurkov, 2023, p. 181), which underscores the state’s awareness of the complexity of the regulatory challenges it faces.

Sectoral priorities and the focus on import substitution are clearly outlined in strategic planning documents. Emphasis is placed on the application of AI in the military-industrial complex and in industries related to national security. This naturally leads to the dominance of state orders and the close linkage of developments with state scientific institutions and state corporations. In parallel, a policy of import substitution of software and hardware solutions is being implemented, aimed at reducing dependence on foreign technologies in the basic elements of the AI ecosystem.

International cooperation within this model is selective and pragmatic in nature, and it is seen, like interaction within BRICS as a whole, as a tool for diversifying international ties and creating alternative technological chains not controlled by Western countries.

The Russian model of “sovereign digitalization” brings a critically important component to BRICS cooperation, emphasizing critical information infrastructure security and technological independence in strategic industries. Through the development of “regulatory sandboxes”^[22] and selective international cooperation (primarily with China), Russia demonstrates practical mechanisms for implementing digital sovereignty under sanctions constraints. Within BRICS, this experience is of high value for countries seeking to maintain politico-technological sovereignty while integrating into global value chains. The Russian approach offers the group’s countries a model of balancing between innovation development and the protection of national interests, which strengthens BRICS’s overall position on the governance of future technologies.

Republic of India

India has implemented a unique model of artificial intelligence development within BRICS, which can be characterized as a pragmatic “AI for Development” (AI for All)^[23] model. Its formation is linked to the launch in 2018 of the National Strategy for Artificial Intelligence,^[24] which laid the foundation for sectoral transformation in the context of AI development.

The institutional architecture and priority sectors of the strategy are focused on addressing the country’s key development challenges. In contrast to the Chinese statist model, the Indian approach emphasizes the application of AI in sectors with maximum social impact: healthcare, agriculture, education, smart cities, and infrastructure. As noted by A. Shaji George (2024) and K. Roy and colleagues (Roy et al., 2024), the regulation remains sufficiently liberal, which purposefully stimulates the development of the startup ecosystem and attracts private investment. The “AI for All” government program is aimed at democratizing technology, ensuring its accessibility to broad sections of the population.

India’s position within BRICS and its international cooperation are determined by the country’s competitive advantages, as it positions itself as a global hub for outsourcing AI services and developing solutions for developing countries. Within BRICS, India advocates for fair and inclusive AI governance, which aligns with its “AI for All” strategy and allows New Delhi to act as a voice for developing economies. This approach contrasts with more rigid regulatory models of other members of the group, creating a foundation for complementary cooperation.

The key challenge and balance of interests for India lies in finding the optimal balance between stimulating innovation and ensuring inclusive growth. On the one hand, the country is interested in developing its own technology industry and startup ecosystem, which requires a liberal regulatory approach. On the other hand, as researchers note (Shaji George, 2024; Roy et al., 2024), the issue of including broad sections of the population in the digital economy and preventing the exacerbation of social inequality remains critically important.

The Indian model of AI regulation represents a pragmatic hybrid, combining a liberal approach to innovation with targeted state interventions in socially significant sectors. The “AI for All” strategy allows India not only to address internal development challenges, but also to occupy a unique niche within BRICS, acting as a bridge between technology giants and developing economies. The success of this model will depend on the government’s ability to maintain a fragile balance between technological progress and social inclusiveness, which is crucial for the sustainable development of the digital economy.

Brazil

Brazil is shaping a socially oriented model of AI regulation within BRICS, based on the principles of responsible data governance and respect for human rights. The key document is the Brazilian Artificial Intelligence Strategy (Estratégia Brasileira de Inteligência Artificial (EBIA), English National Strategy for Artificial Intelligence),^[25] adopted in 2021, which defines the country’s approach to AI technology development.

Data governance occupies a central place in the Brazilian approach. The strategy provides for the creation of a national data governance system aimed at ensuring responsible data use and facilitating data exchange between the public and private sectors. Special attention is paid to the use of government databases for the development of AI technologies, reflecting the desire to enhance the efficiency of public administration through digitalization while maintaining control over critical information assets.

The Brazilian model of AI differs from other BRICS countries due to its social orientation and ethical approach. In contrast to mercantilist approaches or those defending national sovereignty, Brazil’s AI policy emphasizes the protection of human rights, ensuring fairness in algorithmic systems, and preventing discrimination. As noted by G. de Souza, the Brazilian approach is based on the principle of algorithmic transparency and compliance with ethical standards.^[26]

Brazil’s positioning within BRICS is determined by its aspiration to act as a mediator between different regulatory approaches. On the one hand, the Brazilian model is close to liberal-humanistic principles; on the other hand, it is adapted to the realities of a developing economy. This allows Brazil to offer compromise solutions within BRICS that take into account both the need for technological development and the importance of social protection.

Thus, the Brazilian model of AI regulation represents a synthesis of principles of responsible data governance and social orientation, where technological development is combined with human rights protection. The emphasis on ethics, algorithmic transparency, and responsible data use distinguishes the Brazilian approach from the more technocratic models of other BRICS countries and allows the country to occupy an important niche in shaping standards for responsible AI. The success of this model will depend on Brazil’s ability to ensure the practical implementation of its declared principles while maintaining competitiveness in the field of technological innovation.

South Africa

The Republic of South Africa is shaping a model of artificial intelligence development focused on addressing national development challenges and positioning the country as a regional leader in innovation. The foundation of the approach is the draft National Artificial Intelligence Plan,^[27] which defines the country’s strategic priorities in this area.

The goals and priorities of the strategy are centered on three key directions: stimulating economic growth through the implementation of AI technologies, enhancing public welfare, and strengthening South Africa’s position as a center for AI innovation on the African continent. Special attention is paid to the use of artificial intelligence in healthcare, education, and agriculture—sectors that are critical to the country’s socio-economic development.

The socio-ethical orientation of the strategy allows the South African model to be classified within the liberal-humanistic paradigm, which the country shares with Brazil within BRICS. As noted by E. Ormond (2023), the draft national policy places a strong emphasis on the ethical aspects of AI use, the protection of human rights, and ensuring algorithmic transparency. This approach reflects South Africa’s aspiration to reduce social and economic inequality through the responsible implementation of artificial intelligence technologies.

Regional leadership and positioning within BRICS are important elements of South Africa’s strategy. The country strives to become a hub for AI development in Africa, which aligns with its ambitions to strengthen its position on the international stage. Within BRICS, South Africa represents the interests of the continent’s developing economies and advocates for the consideration of the specific needs of developing countries in shaping global AI standards.

Based on the foregoing, it can be concluded that South Africa’s AI regulatory model represents a balanced approach that combines technological development objectives with a strong socio-ethical component. The focus on addressing development challenges, reducing inequality, and promoting regional leadership distinguishes the South African strategy from the approaches of other BRICS countries and allows the country to occupy an important niche in shaping the global AI development agenda. The success of this model will be determined by South Africa’s ability to translate its declared priorities in practice and strengthen its position as a regional center of competence in the field of artificial intelligence.

 Synthesis of Approaches of BRICS Founding Member States in the Field of AI

The conducted analysis of national strategies for the development of artificial intelligence in the BRICS founding states reveals the formation of a unique polycentric regulatory landscape based on the principle of complementarity of different models. Each country contributes its own approach to the common space of cooperation, reflecting its global political ambitions, economic interests, and socio-cultural characteristics.

The set of national strategies of the examined BRICS countries forms not competing, but mutually complementary models, creating a foundation for an ‘overlapping consensus’ in AI regulation. It is precisely the diversity of approaches, rather than their unification, that becomes the key advantage of the grouping, allowing it to offer the world a polycentric alternative to the bipolar regulatory paradigm of “Brussels—Washington.” The strategic task of BRICS lies not in developing a single standard, but in creating a flexible architecture that enables the harmonization of different regulatory approaches to achieve common goals of technological sovereignty and equitable AI governance.

The analysis of BRICS national strategies in the field of AI reveals a system of interconnected challenges that, despite regulatory heterogeneity, form a solid foundation for cooperation.

1. Resistance to the imposition of regulatory standards from outside manifests in the unanimous rejection by BRICS countries of the prospect of simply adapting either the European or the American model. This reflects a common understanding that adopting others’ standards is tantamount to losing technological sovereignty and the ability to pursue independent economic development.
2. Countering technological hegemony is expressed in a collective demand for the demonopolization of global value chains in the field of AI. This challenge becomes particularly acute in the context of restricted access to advanced semiconductors and computing power, controlled by a narrow group of countries and corporations.
3. The imperative of digital sovereignty manifests in the desire of all member states to maintain control over national data and critical technologies. Each country implements this principle in accordance with its own politico-economic model—from statist control to liberal-humanistic regulation.

The identified system of challenges provides a unique foundation for the formation of an ‘overlapping consensus’ within BRICS, where the heterogeneity of regulatory approaches becomes not an obstacle, but a resource for creating a polycentric model of AI governance. The enlargement of BRICS strengthens the group’s representativeness within the Global South, while simultaneously confirming the objective tendency towards the dominance of ‘soft law’ instruments (Shaffer & Pollack, 2010) and the formation of flexible ‘coalitions of the willing.’ This model, based on the harmonization of heterogeneous approaches while preserving national sovereignty, represents a practical alternative to the unifying regulatory approaches of the West and reflects the real needs of a multipolar world order.

‘Overlapping Consensus’ in Action: Evolution of Discourse and Institutions of BRICS

The study of BRICS documents in recent years has demonstrated a gradual shift in the way AI is viewed, from being seen as one of several technologies to forming its own independent agenda. In the BRICS Leaders’ Statement on Global Artificial Intelligence Governance (2025), the key formulations became those concerning the need for “fair and inclusive” AI governance, the importance of “creating reliable AI systems” and sharing best practices, as well as countering “discriminatory and unilateral restrictions” in the field of technology^[28] (a direct reference to the policies of the United States and its allies).

Such formulations, like many other provisions, represent a classic example of an ‘overlapping consensus.’ For China and Russia, “inclusiveness” and “non-discriminatory restrictions” mean fighting against sanctions and gaining access to advanced chips and technologies. For India, “fairness” means access to technologies for the development of its AI ecosystem, while for Brazil and South Africa, “fairness” and “reliability” are primarily about ethical guarantees and the protection of human rights.

It is noteworthy that the current state of the grouping is characterized by its transformation into an exceptionally attractive platform for integration for developing states. The undoubted advantage of this structure is its unprecedented openness, inclusiveness, and equal approach to cooperation (Glebov & Agonnoude, 2023, p. 77). However, the expansion of BRICS does not simplify but complicates the development of a unified position. In today’s world, the common consensus must take into account not only the traditional diversity of approaches but also new axes of tension: between secular and religious models, hydrocarbon-based and agrarian economies, regions with high and low digital maturity. The key instruments of coordination remain memoranda of understanding and joint declarations as tools of ‘soft law’ (Trubek & Trubek, 2005), which corresponds to the identified model of ‘overlapping consensus.’

Within BRICS, the formation of issue-based coalitions is taking place in specific areas:

– countries with developed research bases (China, Russia) focus on joint R&D projects in the field of fundamental research. At the same time, R&D has evolved to the level of an international cluster where, within various forms of cooperation, transnational corporations, universities, industrial complexes, and state institutions generate innovative solutions (Astratova & Klimuk, 2022, p. 3),

– countries with strong IT ecosystems (India, China) develop cooperation in the field of software development and algorithms,

– countries with an emphasis on ethics (Brazil, South Africa) promote the development of ethical frameworks and principles of responsible AI.

Overall, the structure of cooperation within BRICS is evolving towards forming flexible coalitions based on interests, which allows overcoming the initial regulatory heterogeneity.

The expansion of BRICS does not abolish the model of ‘overlapping consensus,’ but brings it to its logical conclusion, creating an alternative regulatory model in the form of an even more complex mosaic of intersecting multilateral and bilateral agreements within the common political paradigm of multipolarity and digital sovereignty. The role of common BRICS institutions will consist not in dictating norms, but in functioning as a “facilitator” or “platform” for finding partners for specific projects and developing minimal common principles for coordination on the global stage (the UN and other platforms). This position makes BRICS simultaneously weaker in terms of internal cohesion, but stronger in terms of representativeness and attractiveness for the Global South.

In the beginning of 2026, one of the key events in the field of international regulation of artificial intelligence technologies was the official support by the Russian Federation of the Chinese initiative to create a World Organization for Cooperation in the Field of Artificial Intelligence. According to a statement made by the Assistant to the President of the Russian Federation Yu. Ushakov, on February 4, 2026, following the Russian-Chinese high-level talks, the leaders of the two states emphasized the strategic importance of innovative areas of international cooperation. Within the framework of the agreements reached, support was confirmed for the PRC’s initiative to form a specialized international organization aimed at coordinating efforts in the development and regulation of AI technologies.^[29]

This initiative elevates bilateral and multilateral cooperation to a fundamentally new level. While previously discussed formats, such as working groups and memoranda, were oriented primarily towards internal coordination within BRICS or bilateral relations, the proposal to establish a worldwide organization marks a transition to the phase of active construction of global institutions alternative to existing Western platforms.

The political significance of this decision is determined by three key factors.

1. The scale of ambition. We are talking not about a regional, but about a global institution, which directly corresponds to BRICS’s stated goal of forming a multipolar world. Establishing such an organization presupposes the development of universal norms and standards in the field of AI applicable at the international level.
2. Speed of reaction. The initiative was announced and immediately supported at the highest level, which indicates a high level of trust and the convergence of strategic interests between Russia and China in this area. The prompt coordination of positions points to preliminary elaboration of the issue and the existence of an agreed roadmap.
3. Potential for expansion. The establishment of such an organization could become a magnet for other Global South countries, including BRICS+ members, who are seeking a platform for cooperation in the field of AI not controlled by Western players. This opens up opportunities for the formation of a broad coalition of states interested in an alternative model of technology regulation.

At the time of this study’s writing, the details of the mandate and structure of the future organization had not been disclosed. However, the very fact of putting forward this initiative confirms the main thesis of the authors: BRICS (and its key members) are moving from a strategy of adaptation to existing rules to a strategy of creating their own global normative architecture.

Thus, the support by the Russian side of the Chinese initiative can be seen as an important step in forming the institutional foundation for global AI technology governance. This is the most vivid illustration of the practical implementation of the ‘overlapping consensus’ model, where a common political imperative (countering technological hegemony) materializes in a specific institutional project. This case demonstrates the evolution of BRICS approaches to international technological regulation—from coordination within the group to the formation of independent global institutions.

Instead of a Conclusion: Practical Recommendations for the Development of BRICS Cooperation in the Field of AI

The conducted research allows us to conclude that BRICS cooperation in the field of AI is not aimed at the rapid creation of a single rigid regulatory space along the lines of the European Union, but rather forms a unique hybrid model, the essence of which lies in the coordination of heterogeneous national regimes to achieve common strategic goals on the global stage. The main features of this emerging model include:

Dominance of ‘soft law’: harmonization occurs primarily through codes of ethics, general principles, and joint political declarations, rather than through directives.

Primacy of politics over law: the basis of cooperation is a common political imperative—building a multipolar technological order, rather than the pursuit of legislative unification.

Instrumental goal: approaches agreed within BRICS are not an end product, but rather a tool for advancing a common position on the platforms of the UN, UNESCO, and the International Telecommunication Union (ITU), where the real “battle for norms” takes place.

Thus, the BRICS “alternative” lies not in a ready-made legal package, but rather in an alternative process of norm-making—more flexible, multi-level, and sovereignty-oriented. Moreover, the ‘overlapping consensus’ model may attract Global South countries seeking to preserve room for maneuver in the context of the technological Cold War.

Based on the analysis, the following practical recommendations for the development of cooperation in the field of AI within BRICS can be formulated.

Development of a digital platform for harmonizing the standards in the field of AI. Given the differences in national regulatory approaches, a promising solution would be to create a joint digital platform that allows for the monitoring and comparison of national standards. Such initiatives already exist: they are being implemented within the G7 (Global Partnership on AI), and this experience can be adapted for BRICS countries. An example is the AI Act Single Information Platform launched by the European Commission in October 2025. The platform, created to support the implementation of the EU AI Act, includes tools such as an interactive Compliance Checker for determining applicable obligations, an AI Act Explorer (a searchable version of the legislation), and a direct inquiry channel to the European AI Office.^[30] Although the platform is oriented towards regulation within the EU, its architecture demonstrates how transparency and navigation can be ensured in a complex regulatory landscape, which is a direct analogue of the proposed harmonization platform for BRICS.

Establishment of a mechanism for expedited recognition of certification. Based on the Working Group on AI, it is advisable to develop an agreement on the mutual recognition of certification results for AI-based products within the framework of ‘regulatory sandboxes.’ This would reduce transaction costs for companies operating in BRICS markets and stimulate technology transfer. A successful example is the initiative of Singapore and Australia to ensure the interoperability of national regulatory approaches in the field of AI, recorded in the Memorandum of Understanding signed in December 2024. The document provides for “promoting the alignment of governance and regulatory frameworks and tools” for artificial intelligence.³¹

Launch of a joint research program on large language models. Countries with developed research bases (China, Russia, India) could initiate a program for the development of multilingual models that take into account the linguistic and cultural diversity of the Global South. As emphasized in the report of India’s G20 Sherpa Amitabh Kant (February 2026), without such an approach, AI development risks entrenching digital inequality and creating a “deeply unequal society.”³²

Promising directions for further research can be focused on several key aspects.

First, it is advisable to conduct a thorough analysis of the strategic approaches of new BRICS members to AI regulation issues.

Second, it is important to assess how the further expansion of the group to the BRICS+ format will affect the achievement of consensus in the field of artificial intelligence. Special attention should be paid to a detailed study of the role of non-state actors, including IT companies and expert communities, in the process of forming an effective regulatory model for this dynamically developing field. Such a comprehensive approach will make it possible not only to identify current trends but also to anticipate possible scenarios for the development of cooperation within the expanded group.

 

 

^{1 De Souza G. Artificial Intelligence in the Office and the Factory: Evidence From Administrative Software Registry Data // Federal Reserve Bank of Chicago Working Paper. 2026 (January). No. 2025-11 (Revised). URL:  https://doi.org/10.21033/wp-2025-11 (accessed: 12.02.2026).}

^{2 The EU Artificial Intelligence Act: How It Will Work // Actual Comments. June 10, 2022. (In Russian). URL: https://actualcomment.ru/zakon-es-ob-ii-ogranichennyy-effekt-bryusselya-2206101541.html (accessed: 25.09.2025).}

^{3 Steklova E. Putin Suggested the Possibility of Transferring the Key Rate Decision to AI // Komsomolskaya Pravda. September 5, 2025. (In Russian). URL: https://www.kp.ru/online/news/6555753/ (accessed: 23.09.2025).}

^{4 IEEE—Institute of Electrical and Electronics Engineers.}

^{5 Modi, Altman, Pichai, Hassabis: Key Takeaways from India AI Impact Summit 2026 // The Economic Times. February 20, 2026. URL: https://economictimes.indiatimes.com/ai/ai-insights/modi-altman-pichai-hassabis-key-takeaways-from-india-ai-impact-summit-2026/videoshow/128589745.cms?from=mdr (accessed: 18.05.2026).}

^{6 Modi Calls for Global AI Compact and Human-Centred Tech at G20 // Outlook India. November 23, 2025. URL: https://www.outlookindia.com/international/modi-calls-for-global-ai-compact-and-human-centred-tech-at-g20 (accessed: 24.02.2026).}

^{7 Amitabh Kant Warns AI Risks Inequality Without DPI, Urges 3 Pillars // Asiaтet News. February 17, 2026. URL: https://newsable.asianetnews.com/world/amitabh-kant-warns-ai-risks-inequality-without-dpi-urges-3-pillars-articleshow-bhy9y05 (accessed: 24.02.2026).}

^{8 Pretz K. Two New AI Ethics Certifications Available from IEEE: CertifAIEd Offers Training on How to Evaluate AI Products // IEEE Spectrum. December 10, 2025. URL: https://spectrum.ieee.org/two-new-ai-ethics-certifications (accessed: 24.02.2026).}

^{9 Delhi Declaration. March 29, 2012 // President of Russia. (In Russian). URL: http://kremlin.ru/supplement/1189 (accessed: 23.09.2025).}

^{10 On the Signing of the Memorandum of Cooperation in the Field of Science, Technology and Innovation between the Governments of Russia, Brazil, India, China and the Republic of South Africa // Government of Russia. March 14, 2015. (In Russian). URL: http://government.ru/docs/17313/ (accessed: 23.09.2025).}

^{11 New Delhi Declaration of the XIII BRICS Summit // BRICS Information Portal. (In Russian). URL: https://infobrics.org/files/country/russia/documents/2021/2021_09_09_India_New-Dehli__XIII_Sammit_Declaration_ru.pdf (accessed: 23.09.2025).}

^{12 Beijing Declaration of the XIV BRICS Summit. June 23, 2022 // President of Russia. (In Russian). URL: http://kremlin.ru/supplement/5819 (accessed: 23.09.2025).}

^{13 Second Johannesburg Declaration of the BRICS Countries. August 24, 2023 // President of Russia. (In Russian). URL: http://kremlin.ru/events/president/news/72103 (accessed: 23.09.2025).}

^{14 BRICS Declaration—AI and Information Security // Digital Russia. July 8, 2025. (In Russian). URL: https:// d-russia.ru/deklaracija-briks-ii-i-ib.html (accessed: 23.09.2025).}

^{15 Rio de Janeiro Declaration: Strengthening Cooperation of the Global South for More Inclusive and Sustainable Governance, Rio de Janeiro, Brazil, July 6, 2025 // President of Russia. (In Russian). URL: http://static.kremlin.ru/media/events/files/ru/gvTArkWauqwuryk9xzLt3HuuI7EBmqrC.pdf (accessed: 23.09.2025).}

^{16 Dmitry Grigorenko: Artificial Intelligence Is a Promising Area for Cooperation between Russia and China // Government of Russia. September 26, 2025. (In Russian). URL: http://government.ru/news/56332/ (accessed: 14.11.2025).}

^{17 Ibid.}

^{18 Grigorenko Named AI Cooperation as the Main Prospect for Russia and China // TASS. September 25, 2025. (In Russian). URL: https://tass.ru/ekonomika/25160035 (accessed: 26.09.2025).}

^{19 Artificial Intelligence Development Plan // Foundation for Law & International Relations. July 8, 2017. URL: https://flia.org/notice-state-council-issuing-new-generation-artificial-intelligence-development-plan/ (accessed: 23.09.2025).}

^{20 National Strategy for the Development of Artificial Intelligence for the Period up to 2030. Approved by Decree of the President of the Russian Federation No. 490 of October 10, 2019 // President of Russia. (In Russian). URL: http://static.kremlin.ru/media/events/files/ru/AH4x6HgKWANwVtMOfPDhcbRpvd1HCCsv.pdf (accessed: 23.09.2025).}

^{21 Federal Law “On the Security of Critical Information Infrastructure of the Russian Federation” No. 187-FZ of July 26, 2017 (latest version) // ConsultantPlus. (In Russian). URL: https://www.consultant.ru/document/cons_doc_LAW_220885/ (accessed: 23.09.2025).}

^{22 What Are “Regulatory Sandboxes” and How They Will Help Business // State Duma of the Federal Assembly of the Russian Federation. August 6, 2020. (In Russian). URL: http://duma.gov.ru/news/49285/ (accessed: 23.09.2025).}

^{23 Welcome to the AI for All Program // Digital India. Ministry of Education. Government of India. URL: https://ai-for-all.in/#/home (accessed: 23.09.2025).}

^{24 India’s National Strategy for Artificial Intelligence // DigWatch. June 2018. URL: https://dig.watch/resource/indias-national-strategy-for-artificial-intelligence (accessed: 23.09.2025).}

^{25 Brazilian Strategy for Artificial Intelligence // OECD. April 15, 2025. URL: https://www.oecd.org/en/publications/access-to-public-research-data-toolkit_a12e8998-en/brazilian-strategy-for-artificial-intelligence_936c5793-en.html (accessed: 23.09.2025).}

^{26 De Souza G. Artificial Intelligence in the Office and the Factory: Evidence From Administrative Software Registry Data // Federal Reserve Bank of Chicago Working Paper. 2026 (January). No. 2025-11 (Revised). URL:  https://doi.org/10.21033/wp-2025-11 (accessed: 12.02.2026).}

^{27 South Africa National Artificial Intelligence Policy Framework // Fairbridges. August 2024. URL: https://fwblaw.co.za/wp-content/uploads/2024/10/South-Africa-National-AI-Policy-Framework-1.pdf (accessed: 23.09.2025).}

^{28 BRICS Leaders’ Statement on Global Artificial Intelligence Governance, Rio de Janeiro, Brazil, July 6, 2025 // President of Russia. (In Russian). URL: http://static.kremlin.ru/media/events/files/ru/n1eGkN9xVCOwXlAGekbAlJOLApDGIXEq.pdf (accessed: 26.09.2025).}

^{29 Ushakov: Russia Supported China’s Idea of a Worldwide Organization in the Field of AI // TASS. February 4, 2026. (In Russian). URL: https://tass.ru/politika/26345195 (accessed: 24.02.2026).}

^{30 Commission Launches AI Act Service Desk and Single Information Platform to Support AI Act Implementation // European Commission. October 8, 2025. URL: https://digital-strategy.ec.europa.eu/en/news/commission-launches-ai-act-service-desk-and-single-information-platform-support-ai-act?pk_source=ec_newsroom&pk_medium=email&pk_campaign=Shaping%20Europe%E2%80%99s%20Digital%20Future/en (accessed: 24.02.2026).}

About the authors

Bidoley V.F. Agonnoude

RUDN University

Author for correspondence.
Email: freddyagonnoude@mail.ru
ORCID iD: 0000-0001-5807-6591
SPIN-code: 9457-7952

PhD (Political Science), Senior Lecturer, Department of Public Policy and History of State and Law, Law Institute

6 Miklukho-Maklaya St, Moscow, 117198, Russian Federation

Victor A. Glebov

RUDN University

Email: glebov17@mail.ru
ORCID iD: 0000-0002-4272-8927
SPIN-code: 5134-9334

PhD (Law), Associate Professor, Deputy Head of the Department of Public Policy and History of State and Law, Law Institute

6 Miklukho-Maklaya St, Moscow, 117198, Russian Federation

Alexey A. Maslov

Lomonosov Moscow State University

Email: maslov@maslov.msk.ru
ORCID iD: 0000-0001-7337-2874
SPIN-code: 8336-3284

PhD, Dr.Sc. (History), Professor, Director, Institute of Asian and African Studies

11 Mokhovaya St, Moscow, 125009, Russian Federation

References

Supplementary files