ARTICLE AD BOX
![]()
New authorities rules present mandate wide labelling for each AI-generated content, including deepfakes and synthetic audio, starting February 20. Social media platforms indispensable verify idiosyncratic declarations connected AI contented and embed traceable metadata. Takedown timelines person been drastically reduced to arsenic small arsenic 3 hours for definite violations, with platforms besides required to pass users astir penalties.
The Central authorities has notified amendments to the IT intermediary rules that formally bring AI-generated content—including deepfake videos, synthetic audio and algorithmically altered visuals—under a structured regulatory model for the archetypal time.The Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026, volition travel into effect from February 20. Under the updated framework, societal media platforms and integer intermediaries indispensable guarantee each synthetically generated accusation (SGI) is intelligibly and prominently labelled truthful users tin instantly archer it isolated from existent content.Platforms are besides required to embed persistent metadata and unsocial identifiers to marque specified contented traceable backmost to the source.
Crucially, intermediaries cannot let the removal oregon suppression of these labels oregon metadata erstwhile applied.
Bigger platforms, bigger responsibilities
Significant societal media intermediaries—think Instagram, YouTube, Facebook—face tighter obligations. Before immoderate upload, they indispensable get a idiosyncratic declaration connected whether the contented is synthetically generated and deploy automated tools to verify those claims. If contented is flagged arsenic AI-made, it indispensable transportation a disposable disclosure earlier it goes live.
Notably, the authorities dropped an earlier connection from the October 2025 draught that would person required disposable watermarks covering astatine slightest 10% of surface abstraction connected AI-generated visuals. Industry groups, including IAMAI, had pushed backmost hard, calling the regularisation excessively rigid and technically impractical crossed formats. The last mentation inactive requires wide labelling—just not a fixed-size watermark plastered crossed the content.The amendments besides sharply compress takedown timelines. In circumstantial cases, platforms present person conscionable 3 hours to enactment connected lawful orders, down from 36 hours earlier. Other effect windows person been chopped from 15 days to 7 and from 24 hours to 12.The rules besides mandate that platforms pass users astatine slightest erstwhile each 3 months astir penalties for violating the caller provisions, including misuse of AI-generated content.
