India AI Regulation Overview

Overview

India’s national regulatory posture on artificial intelligence is grounded in a principle-based, techno-legal framework that balances the promotion of innovation with individual rights and public safety. Anchored in the flagship vision of '#AIforAll', first articulated by NITI Aayog in its 2018 National Strategy for Artificial Intelligence, India frames AI technology primarily as a powerful tool for inclusive economic growth and large-scale social transformation across key public sectors.

Over recent years, governance has evolved through coordinated policy frameworks issued by the Ministry of Electronics and Information Technology (MeitY)—most notably the India AI Governance Guidelines published in November 2025 under the IndiaAI Mission—and data-centric initiatives like the Draft National Data Governance Framework Policy. In parallel, financial regulators such as the Reserve Bank of India (RBI) and the Securities and Exchange Board of India (SEBI) have issued domain-specific frameworks, while Parliament considers statutory proposals such as the Artificial Intelligence (Ethics & Accountability) Bill, 2025.

Regulatory Approach

India's overarching strategy combines principle-based soft-law directives, sector-specific rules, and the strategic deployment of Digital Public Infrastructure (DPI) and national compute facilities. Rather than relying on a single rigid horizontal law, national guidance emphasizes an 'innovation over restraint' posture. The governance landscape is guided by seven core principles known as 'sutras': Trust as the Foundation, People First, Innovation over Restraint, Fairness & Equity, Accountability, Understandable by Design, and Safety, Resilience & Sustainability.

Regulatory enforcement relies predominantly on domain-specific oversight and existing statutory instruments, including the Information Technology Act, 2000 (and its IT Rules, 2021) and the Digital Personal Data Protection Act, 2023. However, legislative proposals such as the Artificial Intelligence (Ethics & Accountability) Bill, 2025 signal an incremental move toward enforceable accountability, mandatory bias audits, right-to-explanation safeguards, and heightened scrutiny for high-risk applications in critical decision-making domains.

Key AI Legislation

• India AI Governance Guidelines (2025): Official MeitY policy framework under the IndiaAI Mission establishing a whole-of-government governance model, seven guiding sutras, an action plan, and practical guidelines for developers and regulators.
• Artificial Intelligence (Ethics & Accountability) Bill, 2025: Proposed legislation introduced in the Lok Sabha to establish a statutory Ethics Committee for AI, prohibit algorithmic discrimination, require bias audits, and limit AI surveillance.
• Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021: In-force subordinate legislation under the IT Act, 2000 prescribing due diligence obligations for online intermediaries, enhanced compliance for Significant Social Media Intermediaries (SSMIs), and a three-tier grievance architecture.
• FREE-AI Framework (RBI Committee Report, 2025): Advisory framework by the Reserve Bank of India providing 7 sutras and 26 recommendations across six strategic pillars for responsible AI adoption by financial regulated entities.
• SEBI Proposed AI Responsibility Amendments (2024): Proposed regulatory amendments assigning clear responsibility and accountability to Market Infrastructure Institutions and registered intermediaries for AI tools used in securities markets.
• Draft National Data Governance Framework Policy (2022): MeitY draft policy creating an India Data Management Office (IDMO) and the 'India Datasets' programme to enable access to non-personal and anonymized datasets for research and startups.
• AIRAWAT Approach Paper (2020): NITI Aayog strategy paper detailing the multi-tenant architecture and governance for a national AI-specific cloud computing infrastructure.
• National Strategy for Artificial Intelligence (2018): Foundational NITI Aayog blueprint establishing the '#AIforAll' initiative and prioritizing five high-impact social sectors.

Governance & Enforcement Bodies

Governance of AI in India involves a whole-of-government institutional structure working alongside specialized sectoral regulators. Under the MeitY India AI Governance Guidelines (2025), strategic coordination is driven by the inter-agency AI Governance Group (AIGG), supported by the Technology & Policy Expert Committee (TPEC) for advisory input, and the AI Safety Institute (AISI) for technical testing, safety research, and standards validation. NITI Aayog serves as the central policy think tank driving overarching AI strategy and public compute initiatives.

Sector-specific execution remains distributed among domain authorities. MeitY acts as the main ministry for digital policy, computing infrastructure, and intermediary regulation. Financial sector adoption and risk management are governed by the Reserve Bank of India (RBI) for banking entities and fintechs, and by the Securities and Exchange Board of India (SEBI) for capital markets. The proposed Artificial Intelligence (Ethics & Accountability) Bill, 2025 envisages establishing an independent statutory Ethics Committee for AI with investigative and regulatory authority across sensitive domains.

Enforcement Bodies

Agency	Mandate	Key Powers	Website
Ministry of Electronics & Information Technology (MeitY)	Central ministry for electronics, IT policy, IndiaAI Mission, and digital public infrastructure.	Formulating national policies, issuing IT guidelines, administering IT Rules 2021, overseeing IDMO and AISI initiatives.
Reserve Bank of India (RBI)	Central bank regulating commercial banks, cooperative banks, NBFCs, and financial regulated entities.	Issuing Master Directions, supervisory expectations, circulars, and financial AI governance standards under the FREE-AI framework.
Securities and Exchange Board of India (SEBI)	Regulatory authority governing the Indian securities markets, intermediaries, and market infrastructure.	Proposing and enforcing binding compliance amendments, supervising market participants, and assigning AI tool responsibility.
NITI Aayog	National policy think tank directing strategic planning and public sector AI integration.	Developing national strategy (#AIforAll), publishing approach papers (AIRAWAT), and recommending cross-sectoral policy standards.
Ethics Committee for Artificial Intelligence (Proposed)	Proposed statutory oversight body under the 2025 Ethics & Accountability Bill.	Investigating algorithmic bias, conducting audits, approving AI surveillance, issuing penalties, and ordering system withdrawals.
AI Safety Institute (AISI) (Recommended)	Technical body recommended under the 2025 MeitY Guidelines for AI safety research.	Developing evaluation metrics, benchmarks, test suites, and conducting safety testing and risk assessments.

Penalties & Enforcement

Under existing policy frameworks and non-binding guidelines (such as the MeitY 2025 Guidelines and the RBI FREE-AI report), compliance relies primarily on board-level governance, self-audits, voluntary commitments, and supervisory guidance. However, statutory compliance for web platforms and intermediaries is enforced through the Information Technology Act, 2000 and the IT Rules, 2021. Intermediaries that fail to observe due diligence obligations risk losing statutory safe-harbour protection under Section 79 of the IT Act, exposing them to legal liability under central laws.

Under proposed statutory frameworks—most notably the Artificial Intelligence (Ethics & Accountability) Bill, 2025—enforcement mechanisms become significantly more formal. The proposed Ethics Committee for AI is granted investigative authority to examine complaints regarding algorithmic bias, discriminatory outcomes, or unauthorized surveillance. The Bill empowers the Committee to recommend remedial measures, mandate the withdrawal of biased AI systems from the market, and impose financial penalties for ethical and compliance breaches.

Data Protection Framework

India’s AI regulatory posture is closely integrated with its national data governance laws, particularly the Information Technology Act, 2000 and the Digital Personal Data Protection Act, 2023 (DPDP Act). Official guidelines from MeitY and sector-specific frameworks from the RBI require all AI developers and deployers to align their data handling practices with statutory privacy requirements, ensuring transparency, user consent, and data protection by design.

To safely unlock data for AI model development without violating individual privacy, the government relies on non-personal and anonymized data frameworks. The Draft National Data Governance Framework Policy (2022) sets out standards for non-personal datasets, creating an India Data Management Office (IDMO) to run the 'India Datasets' platform using standardized metadata and anonymization toolkits. Furthermore, national infrastructure blueprints like NITI Aayog's AIRAWAT emphasize privacy-enhancing technologies, differential privacy controls, access logs, and data localization considerations for sensitive processing.

Sector-Specific Rules

Sectoral tailoring is a defining characteristic of AI oversight in India. In the financial domain, the RBI published its FREE-AI Committee Report in August 2025, detailing 26 actionable recommendations for banks, NBFCs, and fintechs. It prescribes board-approved AI policies, model risk management (MRM), independent validation, incident reporting protocols, and an AI innovation sandbox. Similarly, SEBI proposed amendments in November 2024 to assign direct accountability to Market Infrastructure Institutions (MIIs) and financial intermediaries for their operational AI tools.

Five priority social sectors were initially targeted under NITI Aayog’s 2018 National Strategy (#AIforAll) and the AIRAWAT platform blueprint: healthcare (radiology, diagnostics, early disease detection), agriculture (yield prediction, crop disease analysis), education (personalized learning platforms), smart cities (predictive maintenance, service delivery), and smart mobility (traffic optimization). Across these areas, sectoral regulators and line ministries are tasked with integrating AI explainability, safety standards, and procurement guidelines into their existing regulatory domains.

International Alignment

The corpus holds no record of formal adoption of foreign regulatory regimes such as the EU AI Act, direct transposition of OECD AI principles, or formal bilateral AI agreements with international jurisdictions.

However, Indian policy papers explicitly reference global harmonization as a strategic goal. The Artificial Intelligence (Ethics & Accountability) Bill, 2025 notes the objective of aligning domestic regulatory efforts with international best practices while addressing India-specific challenges. Furthermore, infrastructure projects like NITI Aayog's AIRAWAT explicitly benchmarked international facilities—such as the U.S. Summit supercomputer and Japan’s ABCI—to inform national architectural standards and operational benchmarks.

Future Developments

Future AI regulation in India involves a combination of pending legislative actions and structured policy execution plans. The Artificial Intelligence (Ethics & Accountability) Bill, 2025, introduced in the Lok Sabha in December 2025, represents a potential move toward statutory oversight via a national Ethics Committee for AI. In financial markets, SEBI is finalizing proposed regulatory amendments to establish binding accountability standards for market intermediaries using AI tools.

Concurrently, the MeitY India AI Governance Guidelines (2025) outline a staged Action Plan spanning short-, medium-, and long-term phases. Immediate priorities include constituting the AI Governance Group (AIGG), Technology & Policy Expert Committee (TPEC), and AI Safety Institute (AISI), alongside establishing national AI incident databases and risk classification tools. Subsequent phases envisage operating regulatory sandboxes, integrating AI into Digital Public Infrastructure, and introducing targeted legislative amendments if domain-specific regulatory gaps persist.