- India cuts content removal time from 36 to 3 hours after government or court notice
- AI-content labelling is mandatory, including clear tags for deepfake material
- Large platforms must use automated detection, publish reports, and support law enforcement
India has reduced the timeline for platforms to remove unlawful content from 36 hours to just 3 hours of receiving a government or court notice, and within 2 hours in sensitive cases involving nudity or impersonation.
The government has also made AI-content labelling mandatory, bringing deepfakes directly within the country's digital regulation framework.
While existing laws already cover deepfake crimes, the Centre has now strengthened the regulatory framework by making it a must for platforms to ensure clear labelling and traceable metadata for AI content. This would help people in identifying synthetic content.
Apart from the drastically reduced timeline - from 36 to 3 hours - to remove unlawful content, the government has made it mandatory for platforms to address complaints within 36 hours, down from 72 hours earlier.
For sensitive categories such as nudity, impersonation and related offence, the action timeline has been reduced from 24 hours to 2 hours.
The government made these announcements in a statement today, amid the strong message it sent to Facebook's parent Meta and its owner Mark Zuckerberg over removing a video of Prime Minister Narendra Modi.
The government said large social media platforms with over 50 lakh users must use automated detection tools to catch unlawful content, publish compliance reports, appoint grievance officers, support law-enforcement agencies and enable user verification and appeal systems.
The Information Technology (IT) Act, 2000 already has provisions to deal with identity theft, impersonation, privacy violations and making or sharing obscene content. The Bharatiya Nyaya Sanhita (BNS) 2023 has sections to punish organised cybercrime involving deepfakes.
Some projects supported by the government under the IndiaAI Mission are Saakshya (IIT Jodhpur and IIT Madras) for deepfake detection, AI Vishleshak for audio-visual forgery detection, and IIT Kharagpur's realtime voice deepfake detection system.
