India releases national AI governance guidelines
India's Ministry of Electronics and Information Technology (MeitY) has released new AI Governance Guidelines aimed at balancing innovation with safety and trust. The framework emphasizes sectoral deployment and ethical principles, with coordination led by the AI Governance Group and AI Safety Institute. The guidelines focus on enabling “safe and trusted AI innovation” through technical and procedural safeguards.
- The framework is built on seven core principles, or "sutras," including "Innovation over Restraint," "People First," and "Accountability," which are designed to be technology-agnostic and applicable across all sectors. - Rather than creating new standalone AI legislation, the guidelines advocate for a "lightweight" and adaptive regulatory approach that leverages and amends existing legal frameworks like the Information Technology Act, 2000, and the Digital Personal Data Protection Act, 2023. - A key institutional proposal is the AI Safety Institute (AISI), which will function on a hub-and-spoke model, collaborating with academic institutions and private partners to advance indigenous research on AI safety, develop benchmarks, and test models. - The guidelines propose a phased, long-term action plan that includes establishing common standards for content authentication and data integrity, piloting regulatory sandboxes, and eventually expanding global engagement to contribute to international standards-setting. - This approach contrasts with the European Union's legally binding AI Act, which classifies systems by risk, and the United States' focus on voluntary commitments and innovation-first policies. - The AI Governance Group (AIGG), to be chaired by the Principal Scientific Adviser to the Government of India, will serve as the primary decision-making body for policy coordination and oversight across sectors. - The framework calls for the development of an India-specific AI risk assessment and classification methodology and a national, federated AI incident reporting mechanism to track and analyze harms. - The drafting process was a multi-stakeholder effort that began with a sub-committee in 2023, involved a public consultation that received over 2,500 submissions, and concluded with a drafting committee formed in July 2025 comprising representatives from government, academia, and industry.
