Administrative sovereignty in the age of algorithmic governance: challenges and prospects for policy autonomy of the BRICS states

Full Text

Introduction Artificial intelligence (AI) is reshaping the functioning of governments. From detecting fraud and predicting crime hotspots to managing welfare distribution, algorithms are increasingly embedded in public administration. While digital technologies bring speed and efficiency, they also create governance concerns, especially in countries dependent on foreign-made systems [1]. For the BRICS states, where infrastructure, regulatory frameworks, and technical expertise remain uneven, the adoption of algorithmic tools often depends on global technology corporations or foreign governments. These systems may not be aligned with local needs or values and can weaken the ability of national administrations to adapt policies. The key question emerges: Does technological adoption risk compromising administrative sovereignty in exchange for modernization? The aim of the study is to determine the correlation between digital modernization and the political sovereignty of Russia and India. India and Russia provide instructive comparisons. India emphasizes citizen participation and inclusion, whereas Russia pursues centralized control. Despite their differences, both countries face the challenge of leveraging AI to enhance governance while protecting institutional autonomy and democratic accountability. Theoretical Framework Several theoretical lenses inform this analysis: 1. Bureaucratic Autonomy: rooted in Weber’s theory of rule-based governance, bureaucratic autonomy stresses insulation from external pressures. In the age of AI, autonomy is increasingly tied to a state’s capacity to design, regulate, and monitor algorithms [2]. 2. Technological Determinism: this perspective argues that technological advances often drive societal change. By adopting ready-made algorithms, governments may unconsciously accept external biases embedded in their design, thereby shaping policy outcomes [3]. 3. Digital Colonialism: this concept refers to the dominance of developing nations by technologically powerful actors. Wholesale import of AI tools can result in local administrations losing decision-making independence [4]. 4. Algorithmic Sovereignty: a proposed response, where states assert control over algorithmic systems by building domestic capacity, establishing regulatory oversight, and ensuring AI is aligned with public values [5]. Methodology This study employs Qualitative Comparative Analysis (QCA) to investigate India and Russia’s approaches to AI in governance. These countries were selected because: · Both have national AI strategies and significant public-sector investments in digital programs. · They represent contrasting governance styles - India’s decentralized democratic model and Russia’s centralized, security-driven approach. Sources of data include: · Government reports and policy documents1. · Publications from international think tanks and policy research institutes [4; 7]. · Peer-reviewed academic literature and expert commentaries [2; 3; 5; 6]. Case Studies A. India India integrates AI into its Digital India mission, which seeks to use technology to empower citizens and improve public service delivery. · Aadhaar: launched in 2009, this biometric ID covers over one billion individuals and has reduced fraud in welfare delivery. However, critics raise concerns about privacy, consent, and the exclusion of citizens who fail biometric authentication2 [6]. · Predictive Policing in Uttar Pradesh: algorithms are used to forecast crime-prone areas. Yet, evidence suggests such tools replicate societal biases, disproportionately affecting marginalized groups3 [1]. · PMGDISHA: a digital literacy program targeting rural India, where AI is applied to personalize learning and assess outcomes4. · My Gov: A citizen engagement platform that incorporates AI for real-time sentiment analysis, enabling participatory governance at scale5. · B. Russia · Russia treats AI as both a developmental priority and a matter of national security. · Moscow Smart City Project: Characterized by mass surveillance, facial recognition, and traffic monitoring, the initiative boosts urban management but raises questions about privacy and civil liberties [4; 6]. · Federal Tax Service: AI-driven fraud detection has improved compliance but lacks transparency, leaving taxpayers uncertain about how decisions are reached [4]. · Digital Economy Programme: The national AI roadmap targets global leadership by 2030, with emphasis on Russian innovation and open-source tools. The National AI Strategy, approved by Presidential Decree No. 490 (2019), formalizes this goal. Results Findings reveal several common governance challenges: · Foreign Dependence: India in particular relies heavily on imported technologies, raising issues of contextual appropriateness and data sovereignty [1, 3, 4]. · Opacity of Algorithms: Lack of explainability creates gaps in accountability and public trust [2; 7]. · Bias in AI Systems: Datasets reflecting historical inequalities risk reinforcing discrimination in policy implementation [1; 3]. · Accountability Gaps: Ambiguity persists over responsibility when algorithms influence or determine outcomes [5; 6]. · Privacy Concerns: Extensive data collection without robust legal protections creates risks of misuse and surveillance [4]. · Regulatory Lag: Both states struggle to adapt legal frameworks to the pace of technological change, particularly regarding ethics and citizen redress [2; 5]. Discussion AI adoption presents a complex paradox for governance: while it significantly improves efficiency, speed, and administrative responsiveness, it simultaneously creates risks of weakening accountability structures. In India, participatory initiatives such as My Gov are intended to broaden inclusion and encourage citizen feedback. However, the reliance on foreign-developed systems often generates a gap between the country’s policy goals and the technological models being deployed [1]. For example, credit-assessment algorithms designed for global markets may overlook local socio-economic factors, leading to the unintended exclusion of rural or marginalized groups who might otherwise qualify for state benefits. In contrast, Russia prioritizes domestic technological development as a means of minimizing foreign dependence. While this approach strengthens control over critical infrastructure, it introduces a different set of challenges. Heavy reliance on surveillance-driven projects and opaque algorithmic decision-making erodes public confidence in institutions and limits civic participation. Both experiences demonstrate how algorithmic governance risks shifting authority from democratic deliberation to technocratic rule, where data models outweigh human judgment. Such a transition produces not only technical fragility but also a democratic deficit, as citizens lose both visibility into and influence over how decisions are made [7]. Recommendations To safeguard algorithmic sovereignty, the following measures are essential: 21. Strengthen Domestic AI Ecosystems: Expand research, invest in universities, and encourage local startups; promote open-source solutions [5]. 22. Inclusive Design Processes: Involve administrators, technologists, civil society, and citizens in shaping AI systems [7]. 23. Legal and Ethical Frameworks: Introduce legislation mandating transparency, explainability, and fairness in all public-sector AI [2]. 24. Capacity Building: Train bureaucrats to understand and question AI decisions; establish AI ethics units within government [5]. 25. Citizen Awareness: Expand digital literacy so citizens know their rights in interactions with automated systems6. 26. Regional Cooperation: Build Global South alliances to create standards and avoid dependence on unregulated external technologies [1]. Conclusion AI today functions not merely as a technological instrument but as a political and administrative force that reshapes how decisions are taken, whose voices are prioritized, and what values are embedded in governance systems. For the BRICS states, this transformation is particularly significant, as they must simultaneously pursue modernization and preserve their autonomy. The paradox is clear: AI promises efficiency, precision, and innovation, yet it can also erode transparency, accountability, and sovereignty when adopted without adequate safeguards. The comparative experiences of India and Russia illustrate this tension. India has attempted to harness AI for inclusion and citizen participation, yet its dependence on foreign-designed technologies has sometimes resulted in outcomes misaligned with local contexts. Russia, by contrast, seeks to strengthen domestic innovation and strategic control but often does so through opaque, surveillance-heavy approaches that limit citizen agency. Both trajectories underscore the risks of overreliance - whether on external providers or on centralized state apparatuses. Going forward, the path toward algorithmic sovereignty must involve a deliberate balance between technological adoption and democratic oversight. This requires transparent institutions, accountable regulatory frameworks, citizen engagement, and regional collaboration. Only by embedding ethical and human-centered principles into AI governance can developing nations modernize without sacrificing their policy independence and administrative sovereignty.

About the authors

Bhargav R. Moundekar

MIT Art, Design & Technology University

Author for correspondence.
Email: himanshu0701rm@gmail.com
ORCID iD: 0009-0005-8728-0483

Associate Professor, MIT School of Indian Civil Services

Rajbaug Educational Campus, Loni Kalbhor, Pune, 412201, India

Gauri S. Sharma

MIT Art, Design & Technology University

Email: gaurisharma.ml.m@gmail.com
ORCID iD: 0009-0007-9917-2986

Student, MIT School of Indian Civil Services

Rajbaug Educational Campus, Loni Kalbhor, Pune, 412201, India

References

Supplementary files