India Tightens IT Rules on Deepfakes, Mandates Labelling & Faster Takedowns Under IT Rules

In a decisive move to curb the misuse of AI and deepfake technologies, the Centre has amended India’s IT Rules to bring synthetically generated content firmly under regulatory oversight—placing stricter responsibilities on social media platforms, content hosts, and AI-driven intermediaries.

On February 10, the Ministry of Electronics and Information Technology (MeitY) notified the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026, which will come into force from February 20. The amendments expand the scope of the existing IT Rules to explicitly cover AI-generated audio, visual, and audio-visual content, including deepfakes.

For the first time, the rules formally define “synthetically generated information” as content that is artificially or algorithmically created or altered in a manner that appears real or indistinguishable from an actual person or real-world event. 

At the same time, the government clarified that routine editing, accessibility enhancements, educational material creation, and good-faith formatting will not fall under this category.

Under the amended framework, intermediaries that enable or facilitate the creation or sharing of AI-generated content will be required to deploy reasonable and appropriate technical measures, including automated tools, to prevent the dissemination of unlawful synthetic content. 

This includes material involving child sexual abuse, non-consensual intimate imagery, false electronic records, impersonation, deceptive portrayals of individuals or events, and content linked to explosives or weapons.

A key provision mandates that all permissible AI-generated content must be clearly labelled. Visual content must carry prominent disclosures, audio content must include prefixed audio warnings, and platforms must embed permanent metadata or provenance markers, where technically feasible, to identify the tool or service used to generate the content.

The rules also significantly tighten response timelines. 

Intermediaries must now act on lawful takedown directions within three hours, down from the earlier 36-hour window. Grievance redressal timelines have also been shortened, with user complaints requiring acknowledgement within two hours and resolution within seven days.

Platforms have been directed to inform users every three months about the consequences of violating platform policies or the law, including account suspension, content removal, penalties, and criminal liability.

In cases involving serious offences, intermediaries must notify the appropriate authorities without delay.

Importantly, MeitY clarified that proactive content removal or access disabling by intermediaries, including through automated moderation tools, will not jeopardise their safe harbour protections under Section 79 of the Information Technology Act.

The amendments signal the government’s intent to move beyond advisory frameworks and establish hard compliance obligations for AI-era platforms, as concerns over deepfakes, election interference, identity manipulation, and non-consensual content continue to rise.