Artificial intelligence governance in India: a critical legal framework for accountability, transparency, and regulatory compliance

Abstract

In India, Artificial Intelligence (AI) is now making a significant presence in the fields of governance, finance, healthcare, policing and judicial administration. However, India does not have a specific AI legislation, but it does have existing legislations like the Information Technology Act, 2000, the Digital Personal Data Protection Act, 2023 and non-binding policy guidance. The paper examines the emerging governance framework of AI in India, discusses the constitutional principles, legislative initiatives and soft-law frameworks such as the European Union's AI Act, the OECD's Principles of AI and the sectoral approach in the United States. There are significant regulatory gaps identified in the study based on four analytical criteria: accountability, transparency, enforceability, and remedies to algorithmic harms, such as the lack of a dedicated AI authority, mandatory algorithmic impact assessments, a robust explainability framework and effective remedies to harms. The paper demonstrates through examples of face recognition and automated decision-making that it is not sufficient to have voluntary governance based on principles that lead to the protection of fundamental rights for high-risk AI systems. It proposes a five-layer governance framework, comprising constitutional protection, statutory measures, soft law, sectoral oversight and a special accountability mechanism, such as the creation of a statutory AI authority and compulsory audits. Concluding the paper, it is suggested that the pro-innovation stance adopted by the Indian government fosters technological progress, but it is necessary to have more robust and legally enforceable protection mechanisms to safeguard the constitutional values of equality, privacy and due process in the era of Artificial Intelligence.