To Regulate or Not to Regulate: India’s AI Dilemma


On 8 October 2026, the Union Minister for Electronics and Information Technology announced that the government would release a consultation paper on artificial intelligence (AI) regulation within a month. The announcement is the latest in a series of developments reflecting India’s evolving approach to AI regulation. Over the past few years, the government’s position has shifted between opposing dedicated AI legislation, considering regulatory intervention, and adopting a techno-legal approach that relies primarily on existing laws and voluntary measures.

These developments raise questions about India’s regulatory direction, particularly whether existing laws are adequate, what additional safeguards may be needed, and when AI-specific legislation may be necessary.

India’s Evolving Position on AI Regulation

India’s early approach to AI regulation focused on developing policy frameworks and principles for responsible AI, rather than introducing a comprehensive legislative framework.

The 2018 Task Force on Artificial Intelligence recommended enabling policies for AI development and deployment, including data governance, tax incentives, and standard-setting. NITI Aayog’s National Strategy for Artificial Intelligence emphasised the need for frameworks addressing ethical concerns, data privacy, security, and intellectual property.

India’s early approach to AI regulation focused on developing policy frameworks and principles for responsible AI, rather than introducing a comprehensive legislative framework.

Subsequently, the Ministry of Electronics and Information Technology’s (MeitY) committee on cybersecurity, safety, legal, and ethical issues recommended guidelines on fairness, transparency, and accountability, alongside flexible self-regulatory mechanisms. In 2021, NITI Aayog released Towards Responsible AI for All and Operationalizing Principles for Responsible AI, outlining principles for responsible AI development and deployment. The latter also recommended a risk-based regulatory approach, with greater scrutiny for applications presenting higher risk. However, these recommendations had not translated into a comprehensive, AI-specific legal framework.

In April 2023, responding to a parliamentary question, the Union Minister for Electronics and Information Technology stated that the government was not considering legislation to regulate  AI’s growth. While acknowledging risks related to bias, discrimination, privacy, and accountability, the government viewed AI as a key enabler of economic growth and innovation. However, the government’s position was not entirely hands-off. The proposed Digital India Act, presented in March 2023, had already contemplated the regulation of high-risk AI systems, algorithmic accountability, and regulatory testing frameworks. In June 2023, the then Minister of State for Electronics and Information Technology also indicated that AI would be regulated through the lens of user harm. The government’s willingness to intervene became more apparent in March 2024.

On 1 March, the MeitY issued an advisory directing intermediaries and platforms to seek explicit government permission before making under-tested or unreliable AI models available to users in India. Following criticism from the technology industry, MeitY issued a revised advisory on 15 March, removing the prior-permission requirement and the requirement to submit compliance reports. It retained provisions addressing unlawful content, bias and discrimination, electoral integrity, labelling of unreliable AI outputs, along with identification of AI-generated misinformation and deepfakes. The episode also raised questions about the legal basis of such executive advisories and the predictability of regulatory intervention.

In April 2024, the Union Minister for Electronics and Information Technology indicated that the government was considering AI legislation after the general elections, particularly to address concerns relating to the rights and compensation of news publishers and content creators. The Union IT Minister also questioned the adequacy of industry self-regulation.

By October 2024, the emphasis had shifted again. Speaking at the ET Startup Awards, the Minister also cautioned against adopting heavily prescriptive approaches to AI regulation, referring to regulatory developments in the US and EU. Instead, he suggested following an approach comparable to India’s data protection and telecommunications laws, allowing space for innovation while addressing societal harms.

The broader discussion on AI regulation also reflected an emerging preference for regulating AI applications and their associated harms rather than the underlying technology.

These developments reflected the government’s continuing efforts to balance innovation and safety, while leaving open the question of which regulatory instruments would be appropriate.

Towards a Techno-legal Framework

India’s approach to AI governance was further articulated in 2025. In January, a MeitY subcommittee released its report on AI governance guidelines for public consultation. The report emphasised institutional coordination, a review of existing regulatory frameworks, and technical measures to address AI-related risks.

Subsequently, a separate drafting committee constituted by MeitY in July 2025 developed India AI Governance Guidelines, released in November 2025. The Guidelines concluded that a separate AI law was not required at that stage, as several AI-related risks could be addressed through existing legislation. Instead, they recommended relying on existing legislation, strengthening sectoral oversight, adopting voluntary commitments, and introducing targeted legal amendments where necessary. They also proposed an AI Governance Group to coordinate policy across ministries and regulators.

In parallel, MeitY released draft amendments to the Information Technology Rules in October 2025, proposing additional obligations concerning synthetically generated information, including deepfakes. This indicated that targeted regulatory intervention was being pursued alongside the development of a broader governance framework.

Central to the Guidelines was a techno-legal approach that sought to operationalise regulatory objectives through technical measures, including content authentication, privacy preservation, and bias mitigation. The effectiveness of such measures depends on clear legal obligations, enforceable standards, and institutional capacity. Technical safeguards alone cannot resolve questions of liability and accountability. The Guidelines, however, did not rule out additional regulatory intervention.

The Guidelines envisaged converting voluntary measures into mandatory obligations and introducing new legislation where emerging risks warranted such measures. The government reaffirmed its preference for this approach in Parliament in December 2025.

Renewed Interest in AI Regulation

While India continued to emphasise a techno-legal approach, the government also introduced binding obligations to address specific AI-related harms. In February 2026, the government amended the Information Technology Rules to introduce additional obligations for intermediaries concerning synthetically generated audio, visual, and audiovisual information, including deepfakes. The amendments introduced requirements relating to the labelling and identification of synthetic content, technical measures to address unlawful content, and additional due diligence obligations for relevant intermediaries.

While India continued to emphasise a techno-legal approach, the government also introduced binding obligations to address specific AI-related harms.

The amendments demonstrated how specific AI-related risks could be addressed through existing legislation. They also highlighted broader questions of accountability across the AI value chain. In particular, questions remain concerning the allocation of liability among model developers, deployers, and users, especially where AI systems operate with greater autonomy.

In April 2026, MeitY constituted the AI Governance and Economic Group (AIGEG) to coordinate AI policy across ministries and regulators, review existing regulatory frameworks, and identify the need for legal amendments. The group’s mandate also includes classifying AI use cases into ‘deploy’, ‘pilot’, and ‘defer’ categories, based on factors such as data availability, skills, legal frameworks, and labour adjustment capacity. The nature of these classifications, particularly whether they would be advisory or binding, remains unclear. The possibility of standalone AI legislation has also resurfaced.

In July 2026, media reports suggested that MeitY was examining a standalone AI law addressing consent, liability for AI-generated content, and the autonomy of AI agents. In September 2026, the Union IT Minister acknowledged the role of regulation in ensuring user safety, while highlighting the complexity of regulating AI. The October announcement signals continued deliberations on India’s regulatory approach. It remains unclear whether the consultation will lead to standalone legislation, amendments to existing laws, or further refinement of the techno-legal framework.

The Way Forward

India’s challenge is not simply whether AI should be regulated, but how it should be regulated. Existing laws and sectoral frameworks already address several AI-related risks, although gaps remain, particularly regarding liability and oversight of autonomous systems. At the same time, fragmented regulatory approaches and ambiguous obligations could increase compliance costs, create uncertainty, discourage innovation, and leave scope for discretionary or coercive enforcement. The costs of regulation must therefore be weighed against the risks of inadequate oversight, with adequate safeguards against arbitrary regulatory intervention.

As India seeks to accelerate AI development and adoption, the priority should be to establish a coherent, predictable regulatory framework that relies on existing laws where sufficient, introduces additional obligations where necessary, and ensures proportionate enforcement.

The proposed consultation provides an opportunity to assess existing regulatory gaps, harmonise sectoral approaches, and clarify institutional responsibilities.

As India seeks to accelerate AI development and adoption, the priority should be to establish a coherent, predictable regulatory framework that relies on existing laws where sufficient, introduces additional obligations where necessary, and ensures proportionate enforcement.


Basu Chandola is an Associate Fellow at the Observer Research Foundation.

The views expressed above belong to the author(s). ORF research and analyses now available on Telegram! Click here to access our curated content — blogs, longforms and interviews.



Source link

Related posts

Explained: Sarvam AI’s Token Factory and debate over AI sovereignty goals | Artificial Intelligence News

BHASHINI and NIELIT Partner to Expand Multilingual AI Education in India

Use Of Artificial Intelligence (AI) In Courts