Algorithmic Bias and Regulatory Oversight: Examining the Role of AI Governance in India's Public Sector

The rapid expansion of artificial intelligence (AI) across India’s public sector from welfare delivery and policing to taxation and judicial administration has intensified concerns about opacity, discrimination, and due-process deficits. This paper evaluates how India’s evolving AI governance framework confronts algorithmic bias and identifies persistent regulatory gaps. Situating India within global debates on algorithmic fairness, it draws on socio-legal scholarship showing that data-driven systems can reproduce structural inequalities even without explicit discriminatory intent. India’s transition from early policy statements to recent instruments such as the Digital Personal Data Protection Act 2023, the IndiaAI Mission (2024), and the India AI Governance Guidelines (2025), along with sectoral initiatives like the RBI’s FREE-AI framework, reflects growing acknowledgment of fairness and accountability. However, these norms remain fragmented and largely non-binding, with crucial omissions such as the absence of safeguards against fully automated decision-making and limited mandates for explainability or impact assessments. Through case studies on digital welfare systems, biometric and predictive policing, and AI-enabled judicial administration, the paper demonstrates how such systems risk deepening exclusion and indirect discrimination. It argues that constitutional guarantees of equality and privacy provide a normative basis for redress, and proposes a rights-based AI law, mandatory impact assessments, stronger procurement transparency, and empowered oversight institutions.