Deepfake
Also called: Synthetically generated information, AI-generated media · Topic: Transparency, Accountability and e-Governance · NCERT: Beyond NCERT
Meaning
A deepfake is a fake video, audio clip or image made or changed with AI (artificial intelligence). It shows a real person saying or doing something they never said or did. Indian law calls it "synthetically generated information". In February 2026, MeitY amended the IT Rules, 2021. Platforms must now label AI-made content and embed provenance, a hidden digital record of where the content came from. They must also remove unlawful content within three hours instead of 36. A platform that fails to act risks losing safe harbour under Section 79 of the IT Act, which is its legal protection from liability for what users post. Deepfakes matter because they can mislead voters and damage reputations. They also erode trust in public institutions.
Example
In October 2026, the Supreme Court heard a case on the misuse of live-streamed court hearings, including manipulated and deepfake clips. Advocate Prashant Bhushan proposed a verifiable digital signature or watermark on every official clip. This would let viewers tell genuine court footage from faked versions.
Don't confuse with
- Misleading edited clip: a real 20-second clip of a judge's remark, cut away from the argument around it, is genuine footage. It is not a deepfake, and a watermark cannot stop it from misleading people.
- Misinformation: this is any false or misleading content, including plain text. A deepfake is one specific kind, made with AI.
Recent developments
- July 2025: A MeitY release (issued through PIB, the government's press office) clarified that Sections 67A and 67B of the IT Act cover obscene material made with deepfake technology. Section 67A punishes publishing or sending sexually explicit content, and Section 67B covers child sexual abuse material. This matters because existing law could already punish obscene deepfakes before any AI-specific rules came in.
- November 2025: The India AI Governance Guidelines (5 November 2025) set out a risk-based, proportionate framework, which means stricter control where the possible harm is greater. The framework builds on existing laws. This matters because India handles deepfakes through IT Act rules and guidelines, not through a separate AI Act passed by Parliament.