MeitY Issues India AI Governance Guidelines 2025

The Ministry of Electronics and Information Technology (MeitY) has released the India AI Governance Guidelines 2025, introducing a comprehensive framework to ensure the safe, trusted, and inclusive development of Artificial Intelligence (AI) in India. Developed by a MeitY-appointed committee in July 2025, the Guidelines aim to balance innovation and competitiveness with ethical safeguards and public accountability, aligning with the Government's vision of AI for All and Viksit Bharat 2047.

Key Highlights

1. Seven Guiding Sutras for Responsible AI

The governance framework is founded on seven core principles designed to be technology-neutral and cross-sectoral:

Trust is the Foundation, People First, Innovation over Restraint, Fairness & Equity, Accountability, Understandable by Design, and Safety, Resilience & Sustainability. These sutras collectively promote innovation while embedding human oversight, transparency, and ethical safeguards across the AI lifecycle.

2. Institutional and Governance Structure

A three-tier institutional model has been proposed:

• AI Governance Group (AIGG): apex inter-ministerial body to coordinate AI policy and oversee governance.

• Technology & Policy Expert Committee (TPEC): provides technical, legal, and policy expertise to the AIGG.

• AI Safety Institute (AISI): serves as India's research and validation body for safety testing, risk evaluation, and global collaboration on AI standards.

Together, these institutions will operationalise a whole-of-government approach to AI governance.

3. Six Pillars of AI Governance

The Guidelines set out recommendations under six key pillars:

• Infrastructure: expand access to compute and datasets through initiatives like IndiaAI Mission and AIKosh

• Capacity Building: invest in AI skilling, literacy, and awareness for citizens and public officials.

• Policy & Regulation: adopt flexible, principle-based, and sector-specific frameworks instead of a standalone AI law.

• Risk Mitigation: develop India-specific frameworks for AI risk assessment, deepfake detection, and incident reporting.

• Accountability: enforce graded liability, grievance redressal, and transparency obligations across the AI value chain.

• Institutions: ensure coordinated oversight among regulators, standard-setting bodies, and ministries.

4. Legal and Regulatory Context

The Guidelines clarify that AI systems remain governed under existing Indian laws, including the Information Technology Act (2000), Digital Personal Data Protection Act (2023), Copyright Act (1957), and other civil and criminal statutes. The Committee concludes that a separate AI-specific legislation is not required at this stage, as most risks can be addressed through current laws and timely enforcement. However, it recommends that targeted legislative amendments may be considered to:

• Clarify the classification and liability of different AI actors (developers, deployers, users) under the IT Act;

• Review data protection provisions under the DPDP Act to ensure compatibility with AI training and processing;

• Address sector-specific risks through domain regulators such as RBI, SEBI, TRAI, and others.

5. Action Plan and Implementation Roadmap

The Guidelines outline short-, medium-, and long-term actions, including:

• Establishing the AIGG and TPEC as permanent governance institutions.

• Creating a national AI incident-reporting system and risk classification framework.

• Publishing voluntary industry standards and transparency requirements.

• Integrating AI solutions with Digital Public Infrastructure (DPI) to promote scalable, inclusive deployment.

• Conducting periodic legal reviews to align regulatory frameworks with emerging AI capabilities.

Implications for Industry

• No blanket AI law has been introduced; instead, existing sectoral regulators (RBI, SEBI, TRAI, etc.) will oversee compliance for AI applications in their domains.

• Organisations developing or deploying AI in India must ensure compliance-by-design, data protection, and risk mitigation aligned with Indian law.

• Voluntary codes of practice, algorithmic audits, and techno-legal mechanisms (such as DEPA for AI Training) are encouraged to build accountability and public trust.

• Firms should prepare for transparency reporting, internal grievance mechanisms, and potential audits by relevant regulators or the AISI.