India is moving to make social media companies respond faster to deepfakes and other AI-generated impersonations. The amended rules shorten compliance timelines, require clearer identification of synthetic audio and visual content, and place new pressure on platforms that operate in one of the world’s largest and fastest growing markets for internet services.
What the amended rules require
The changes were published as amendments to India’s 2021 IT Rules and will come into effect on February 20. They bring deepfakes into a formal regulatory framework, with obligations that apply to social media platforms that allow users to upload or share audio-visual content.
Platforms must require users to disclose whether audio-visual material is synthetically generated. They must also use tools to verify those claims, clearly label deepfakes, and embed traceable provenance data in synthetic content.
The rules also set out categories of synthetic content that are not allowed. These include deceptive impersonations, non-consensual intimate imagery, and material linked to serious crimes.
The practical effect is a more structured compliance burden for platforms. A company cannot simply wait for users or authorities to identify problematic material after it spreads. The rules expect platforms to build systems that can identify, label, verify, and block certain synthetic media before or shortly after it appears.
Shorter takedown windows raise the stakes
The most immediate operational change is time. The amended rules include a three-hour deadline for official takedown orders and a two-hour window for certain urgent user complaints.
Those timelines matter because non-compliance can affect safe-harbor protections under Indian law. In plain terms, platforms that do not act within the required window may face greater legal liability for the content they host.
Rohit Kumar, founding partner at New Delhi-based policy consulting firm The Quantum Hub, described the amendments as more targeted than a broad attempt to regulate all AI content. But he also warned that the speed demanded by the rules will be difficult for companies to manage.
The amended IT Rules mark a more calibrated approach to regulating AI-generated deepfakes.
He added that the two- to three-hour takedown windows will raise compliance burdens and deserve close scrutiny because failure to meet them is tied to the loss of safe harbor protections.
Aprajita Rana, a partner at AZB & Partners, said the rules focus on AI-generated audio-visual content rather than all online information. She also noted that the framework carves out exceptions for routine, cosmetic, or efficiency-related uses of AI.
At the same time, Rana cautioned that requiring intermediaries to remove content within three hours once they become aware of it departs from established free-speech principles.
The law, however, continues to require intermediaries to remove content upon being aware or receiving actual knowledge, that too within three hours.
She said the labeling requirements would apply across formats to curb the spread of child sexual abuse material and deceptive content.
Automation becomes central to compliance
The amended rules depend heavily on automated systems. Platforms are expected to deploy technical tools to verify user disclosures, identify and label deepfakes, and prevent the creation or sharing of prohibited synthetic content.
That design reflects the scale of the problem as framed by the rules. With over a billion internet users and a predominantly young population, India is a critical market for companies such as Meta and YouTube. If global platforms adjust moderation products and internal workflows for India, those changes could influence practices beyond the country.
Automation, however, is also where some of the sharpest concerns appear. New Delhi-based digital advocacy group Internet Freedom Foundation said the rules risk accelerating censorship by drastically compressing takedown timelines. The group warned that short deadlines leave little room for human review and may push platforms toward automated over-removal.
These impossibly short timelines eliminate any meaningful human review.
Internet Freedom Foundation also raised concerns about the expansion of prohibited content categories and provisions that allow platforms to disclose the identities of users to private complainants without judicial oversight.
The tension is clear. Faster removal may reduce the spread of harmful deepfakes and deceptive content. But when platforms must decide quickly, especially under legal pressure, they may rely more heavily on automated judgments that can remove lawful or contested speech.
Why India’s decision matters for global platforms
India’s size as a digital market gives the amended IT Rules influence beyond a single jurisdiction. A platform that serves users in India may need to redesign disclosure prompts, detection tools, labeling systems, escalation processes, and takedown workflows to meet the new requirements.
Two industry sources told TechCrunch that the amendments followed a limited consultation process, with only a narrow set of suggestions reflected in the final rules. The Indian government appears to have accepted proposals to narrow the scope from all online material to AI-generated audio-visual content, but other recommendations were not adopted.
The same sources said the scale of changes between the draft and final rules warranted another round of consultation so companies could receive clearer guidance on compliance expectations.
The rules also arrive against an existing backdrop of disputes over content removal in India. Social media platforms and civil-society groups have criticized the breadth and opacity of takedown orders. X challenged New Delhi in court over directives to block or remove posts, arguing that they amounted to overreach and lacked adequate safeguards.
In October 2025, the Indian government reduced the number of officials authorized to order content removals from the internet in response to a legal challenge by X over the scope and transparency of takedown powers.
The rollout comes as India hosts the AI Impact Summit in New Delhi from February 16 to 20. The event is expected to draw senior global technology executives and policymakers, placing the country’s new deepfake rules in front of many of the companies and officials most affected by them.
Meta, Google, Snap, X, and the Indian IT ministry did not respond to requests for comments.