India Is Done Waiting on AI Rules. Its Consultation Paper Lands Next Month.

On October 8, 2026, Union IT Minister Ashwini Vaishnaw stood at the India launch of the World Bank's World Development Report 2026: The Promise of Artificial Intelligence and told the room what many in New Delhi had been circling for months: the time has come for India to regulate artificial intelligence. The Ministry of Electronics and Information Technology, known as MeitY, will publish a regulation-focused consultation paper on AI within a month, a document he said would "form the base for the regulation which has to come."

Until now, New Delhi's posture has rested on voluntary guidelines, existing intermediary rules, and the IndiaAI Governance Guidelines, which set expectations on safety, transparency, and accountability without new enforcement machinery. The October 8 announcement signals a turn toward something harder: formal regulation, drafted after public consultation, built on what Vaishnaw called a "techno-legal" approach. As of October 11, the paper itself has not been published, so every detail below describes the government's stated intentions, not enacted law.

What the paper will cover

Vaishnaw laid out four stated elements the consultation will address: AI safety, a human-first approach, skilling, and a commitment to leave no person behind. He added a fifth focus with more urgency than the others: deepfakes. "Deepfake quality is improving," he said, describing how convincing synthetic media sets a false narrative and lures people into doing wrong things. The consultation is also expected to consider psychological harms associated with AI use, a category that has received far less regulatory attention than bias or misinformation but is increasingly hard to ignore as AI companions and generative tools reach millions of users.

The techno-legal turn

India has already been testing the technical half of that formula: MeitY has consulted on amendments to the IT Rules, 2021 addressing synthetically generated information, and a separate ministry advisory urged social media platforms to curb "manipulated or out-of-context content," going beyond deepfakes to cover misleading material presented as current or authentic.

The existing enforcement rails matter because the consultation paper will be layered on top of them. The IT (Intermediary Guidelines and Digital Media Ethics Code) Rules were notified in 2021 under the IT Act, 2000. Section 69A allows the government to order the blocking of public access to online information, and Section 79 governs the safe-harbour protections shielding intermediaries from liability for user content. In February 2026, the IT Rules amendments cut takedown compliance timelines to 1 to 3 hours for certain flagged content, forcing platforms operating in India to maintain rapid-response systems. A future AI regulation is likely to extend similar expectations to AI developers, not just social media intermediaries.

"Bulk of the responsibility will have to be taken by the industry." The industry, not the state, is expected to build the safeguards first.

Sovereignty runs underneath everything

The safety discussion is only half the story. At the same event, MeitY Secretary S Krishnan raised the strategic half: privately owned AI models carry a risk because Indian data can travel abroad while those models learn from whatever Indian users feed them. Krishnan argued India needs a judicious mix of approaches to protect its strategic autonomy, and reporting from the event indicates the country is turning toward open-source and open-weight models that can run inside India without sending data out.

That ambition is backed by hardware spending on a scale few countries match. The IndiaAI Mission, launched in March 2024 with a budget of Rs 10,371.92 crore over five years, has made more than 38,000 GPUs available, and Mission CEO Abhishek Singh said in February the country should have roughly 100,000 by the end of 2026. The ministry has also issued a request for empanelment of 10,000 GPUs, with an initial 5,000. On the manufacturing side, the government approved Semicon 2.0 in July 2026 with an outlay of Rs 1,27,500 crore for chip design, fabrication units, and advanced packaging. Most of India's GPUs still come from abroad through approved compute providers, which the government itself acknowledges leaves the country exposed to supply disruptions and high import costs.

The deepfake fight is already live

While the consultation paper is still a month away, India's deepfake crackdown has already started moving. The MeitY advisory on manipulated content warns that false, misleading, or out-of-context content presented as current or authentic can cause public confusion, disturb public order, and create law-and-order concerns, and it asks platforms to prioritize complaints from authorized government agencies. The timing is telling: the advisory arrived ahead of a planned October 10 protest and followed a viral deepfake video involving Telugu film stars, the exact kind of incident that makes abstract policy debates feel urgent.

How India fits between Washington and Brussels

The October 8 announcement lands in a week when the contrast between global approaches could hardly be sharper. In the United States, the White House has declared AI security-incident reporting mandatory after Anthropic's disclosure of rogue-agent behavior. In the EU, the AI Act's transparency duties for AI-generated content have applied since August 2026, and the Commission has information requests out to more than 30 AI companies. India's consultation paper is likely to be more prescriptive than Washington's voluntary frameworks and more operationally detailed than Brussels', with an emphasis on what platforms and developers must actually build: detection systems, reporting workflows, and human review.

There is also a distinctive infrastructure play. At the India AI Impact Summit earlier in 2026, Vaishnaw outlined plans for a UPI-like public digital infrastructure for AI, an interoperable framework letting multiple AI models operate on shared public rails. Combined with the 10,000-GPU empanelment push and the open-weight turn, the picture is of a country trying to regulate AI and industrialize it at the same time.

What to watch

The consultation window should open by early November, counting a month from October 8. That is when the real debate begins: which risks the paper names, whether deepfake rules become hard law or stay advisory, and how industry responds to the demand that it shoulder the bulk of the responsibility. For startups and platforms operating in India, the practical advice in the interim is unglamorous but clear: map every AI system you deploy, prepare model cards and data-provenance notes, and watch the MeitY announcements page. In India's techno-legal model, the paperwork is the regulation.