The government has amended the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, to explicitly bring AI generated content—termed “synthetically generated information”—under India’s intermediary framework.
The Ministry of Electronics and Information Technology (MeitY) notified the changes on February 10 with enforcement beginning February 20. This tightens platform obligations around labelling synthetic media and accelerates takedown timelines for unlawful or harmful content.
From a regulatory design standpoint, the objective is unmistakable: Curb the risks posed by deepfakes, impersonation, and digitally supercharged misinformation, while ensuring intermediaries that follow due diligence requirements continue to benefit from safe harbour protections.
“This is one of the first instances in India where AI-generated content is directly addressed within a binding regulatory framework. While the rules do not regulate AI systems per se, they effectively regulate AI outputs at the distribution layer—a pragmatic step in the absence of a standalone AI law,” says Supratim Chakraborty, partner at Khaitan & Co.
At the same time, the rules make it clear that ordinary, good faith edits are not treated as synthetic content. Simple actions like trimming a video, fixing captions, translating text, improving accessibility, or creating educational material are allowed—as long as they don’t mislead viewers or create false electronic records.This separation helps regulators focus on harmful deepfakes and deceptive AI generated media without penalising harmless everyday editing.
Platforms that allow creation or sharing of AI content must clearly label synthetic media so users can immediately identify it as such, and, where technically feasible, embed permanent provenance markers or metadata that help trace origin. Intermediaries are prohibited from enabling removal or suppression of these disclosures once applied.
The government has moved away from the earlier draft’s prescriptive “10 percent of the frame” watermark idea; the final rules use a principle based standard that the label must be “clear” and “prominent”, giving companies flexibility on design while keeping user transparency front and centre.Also Read: DPDP Rules enactment: What it means for companies and citizens
For serious violations—such as non consensual intimate deepfakes, deceptive impersonation, child sexual abuse material, or other unlawful synthetic content—platforms are expected to act within three hours of being notified, a sharp tightening from earlier norms around 36 hours.
Some complaint categories under Rule 3 also see shorter internal timelines, signalling a broader push towards rapid response on high risk content. Supporters say speed is essential to limit viral harm, while critics warn the compressed window could force automated removals with limited human review.
For platforms—global giants and Indian startups alike—the rules translate into a materially higher compliance burden. The operational implications are significant, particularly given the short transition window. “Platforms will need to embed metadata and labelling tools, deploy automated detection systems, and recalibrate grievance and takedown workflows within compressed timelines,” adds Chakraborty. The amendments also leave interpretive questions, he says. “The obligations extend to intermediaries that ‘offer a computer resource’ enabling synthetic content, which could raise scope issues for different categories of AI service providers.”
First Published: Feb 11, 2026, 17:05
Subscribe Now