Key points
- The Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026 were notified on 10 February 2026 and took effect on 20 February 2026.
- They define “synthetically generated information”: audio, images or video made or altered by computer so as to appear real. Routine editing, good-faith documents and accessibility uses are excluded, and AI-generated text is not covered.
- Services that let users create such content must block unlawful synthetic content and label the rest prominently, with permanent metadata or a unique identifier where technically feasible.
- Takedown deadlines are far shorter: three hours after a court order or government notice (previously 36), and two hours for intimate or morphed images (previously 24).
Deepfakes have moved from novelty to everyday risk faster than the law. India’s answer is an amendment to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, notified by the Ministry of Electronics and Information Technology (MeitY) on 10 February 2026 (G.S.R. 120(E)) and in force from 20 February 2026. It applies to intermediaries: social media platforms, messaging services, hosting providers and other services through which users create, upload or share content. An intermediary that does not follow the Rules risks losing the “safe harbour” in section 79 of the Information Technology Act, 2000, which protects it from liability for what its users post.
What counts as synthetically generated information
The new definition covers audio, visual or audio-visual information that is artificially or algorithmically created, generated, modified or altered using a computer resource so that it appears real, and that depicts a person or event in a way that is, or is likely to be seen as, indistinguishable from a real person or event (rule 2(1)(wa)). Three things are excluded:
- routine or good-faith editing, formatting, enhancement, technical correction, colour adjustment, noise reduction, transcription or compression that does not materially alter the substance, context or meaning;
- the routine or good-faith preparation of documents, presentations, PDFs, educational or training materials and research outputs, including illustrative or template-based content, so long as no false document or false electronic record results; and
- uses solely to improve accessibility, clarity, quality, translation, description, searchability or discoverability, without altering any material part of the content.
Text produced by AI falls outside this definition, though unlawful text remains covered by the existing Rules.
Duties of services that offer AI tools
An intermediary whose service lets users create or alter synthetic content must (rule 3(3)):
- use reasonable technical measures, including automated tools, to prevent unlawful synthetic content, such as child sexual abuse material, non-consensual intimate imagery, false documents or records, material about explosives or arms, and content that falsely depicts a real person or event in a way likely to deceive;
- label all other synthetic content prominently, so that it is easily noticed (for audio, with a disclosure at the start), and embed permanent metadata or another provenance mechanism, including a unique identifier, to the extent technically feasible; and
- not allow the label or the metadata to be modified, suppressed or removed.
Extra duties for large platforms
A significant social media intermediary (a platform with at least 50 lakh registered users in India) must, before content is published, ask users to declare whether it is synthetic, use technical measures to check the declaration, and label content that is declared or found to be synthetic. A platform that knowingly allows, promotes or fails to act on synthetic content in breach of the Rules is treated as having failed to exercise due diligence (rule 4(1A)).
Faster takedowns
| Situation | Before | Now |
|---|---|---|
| Unlawful content, on a court order or a reasoned notice from an authorised government officer (rule 3(1)(d)) | 36 hours | 3 hours |
| Intimate images, nudity, sexual content or impersonation, including morphed images, on the complaint of the person shown (rule 3(2)(b)) | 24 hours | 2 hours |
| Resolving a user’s grievance (rule 3(2)(a)(i)) | 15 days | 7 days |
| Certain complaints seeking removal of content (proviso to rule 3(2)(a)(i)) | 72 hours | 36 hours |
The amendment also makes clear that removing content in compliance with the Rules, including by automated tools, does not cost an intermediary its safe harbour (rule 2(1B)).
What it means
- Platforms and apps with AI features, such as image, voice or video generators, filters and avatar tools, need labelling, metadata and abuse prevention built in, and a takedown process that can act within hours, at any time of day.
- Brands and agencies using AI-generated visuals, voices or presenters in advertising should expect platforms to ask for a declaration and to label the content. Plan for it, and keep records of what was generated and how.
- People and companies targeted by a deepfake can complain to the platform’s grievance officer; for intimate or morphed images the platform must act within two hours. Courts also grant urgent injunctions to protect a person’s name, image and voice: see our article Who Owns Your Face?
- Watch for further changes. In April 2026 MeitY consulted on a further draft amendment, including labels that stay visible for the whole of a piece of synthetic content. Check the current text before relying on any detail.
See our technology, media and entertainment and data protection pages.
Sources
This article is for general information and states the law as at 2 November 2026. It is not legal advice. For advice on your situation, contact the firm. Published by BRB Legal.
