New Delhi: In a major move to combat deepfakes and misinformation, the Government of India has officially notified amendments to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021. These new regulations, aimed at governing Synthetically Generated Information (SGI), will come into effect on February 20, 2026.
Under the new mandate, social media giants like Instagram, YouTube, and Facebook, as well as individual content creators, must adhere to strict transparency standards regarding Artificial Intelligence (AI).
Key Highlights of the New AI Regulations:
1. Mandatory AI Disclosure Labels
Every piece of content—be it audio, video, or image—that has been artificially created or algorithmically altered to look authentic must carry a clear and prominent label. This label must be visible on the content itself, ensuring users know they are viewing AI-generated media before they interact with or share it.
2. 3-Hour Content Removal Deadline
The government has significantly tightened the timeline for removing illegal content. For specific lawful orders, platforms now have only three hours to take down content, down from the previous 36-hour window. Other compliance timelines have been slashed from 15 days to seven, and 24 hours to 12.
3. User Declarations and Metadata
Significant Social Media Intermediaries (SSMIs) are now required to ask users if their content is AI-generated during the upload process. Furthermore, platforms must embed persistent metadata and unique identifiers into AI content to ensure traceability, making it harder for labels to be removed when files are downloaded and re-uploaded.
4. Automated Blocking of Illegal Media
Platforms are now legally obligated to use automated tools to proactively block specific categories of harmful AI content, including:
- Child Sexual Abuse Material (CSAM).
- Deepfakes intended to deceive or misrepresent real people.
- Content involving explosives or weapons.
- Obscene or pornographic synthetic media.
5. Legal Consequences for Users
The rules aren’t just for big tech. Users who misrepresent AI-generated content or use it for deceptive purposes could face penalties under the Bharatiya Nyaya Sanhita (BNS) or the POCSO Act. Platforms are also required to send “due diligence” warnings to users at least once every three months to remind them of these penalties.
What Stays Exempt?
The government has clarified that routine digital edits do not fall under these strict rules. This includes:
- Color correction and noise reduction.
- Language translation and transcription.
- Illustrative AI graphics used in professional presentations (like PowerPoint) or research papers.
The Bottom Line
With these rules, India joins a select group of nations taking a firm stand against the unregulated growth of AI-generated misinformation. For the average user, the most visible change will be the “AI-generated” tags on reels and videos, providing a much-needed layer of transparency in the digital age.
