MeitY Proposes IT Rules to Label and Trace AI-Generated and Deepfake Content

• The Ministry of Electronics and Information Technology (MeitY) has released draft amendments to the IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 , introducing mandatory requirements for social media platforms to label and trace AI-generated and deepfake content . The move aims to curb misinformation, impersonation, and manipulation caused by synthetic media.

Key Points

Mandatory Labelling:

• All content created or modified using AI tools must carry a visible or audible label or a unique metadata identifier . Labels must cover at least 10% of visual content or the first 10% of audio duration . Metadata cannot be altered, removed, or suppressed.

Scope for Significant Platforms:

• Social media platforms in India with over 5 million users —including Facebook, YouTube, and Snapchat—are classified as Significant Social Media Intermediaries (SSMIs) . These platforms are responsible for implementing labelling mechanisms and ensuring transparency of AI-modified content.

Transparency and Accountability:

• Platforms enabling AI content creation must deploy verification tools to detect synthetic media. Users must declare when their uploads use AI tools. The measures aim to maintain digital spaces that are safe, open, and accountable , while supporting responsible AI use.

GK Facts

• IT Rules, 2021 impose obligations on large platforms to protect user safety and data accountability.

• Deepfakes can be used for impersonation, misinformation, and election interference.

• Countries like the EU, US, and China are introducing AI content labelling and watermarking standards.

• Safe-harbour protection under the IT Act may be revoked if platforms fail to meet due diligence standards.

Public Consultation:

• MeitY has invited feedback on the draft amendments until 6 November 2025 at itrules.consultation@meity.gov.in . The final rules will set standards for AI content transparency and accountability in India’s digital ecosystem.

In summary:
The proposed amendments aim to make AI-generated content traceable and clearly labelled , holding platforms and users accountable and reinforcing trust in online information.