India's Deepfake Crackdown: Mandatory AI Labels and Permanent Metadata Tags

The proposed changes introduce a clear definition of “synthetically generated information” and require platforms, especially major social media intermediaries (SSMIs), to label such content through metadata and visible or audible markings.

Under IT rules, SSMIs are those platforms that have more than 5 million registered users in India, such as Facebook, YouTube, Snapchat, etc.

Proposed rule mandates that social media platforms facilitating AI content creation must ensure that the information is “prominently labelled or embedded with permanent unique metadata or identifier”.

The identifier must be visible or audible, “covering at least ten per cent of the surface area of the visual display or, in the case of audio content, during the initial ten per cent of its duration,” the rules stated.

The draft rules also said that the metadata or identifier must not be alterable, suppressed, or removed.

If a platform knowingly allows unlabelled or falsely declared AI-generated content, it will be deemed to have failed in exercising due diligence under the IT Act.

Platforms that enable the creation or modification of synthetic content would also be required to adopt technical measures to verify and declare if uploaded content is AI-generated, the ministry said.

The ministry added that the move is part of its broader push to maintain an “open, safe, trusted and accountable Internet” while addressing growing risks of misinformation, impersonation, and election manipulation driven by generative AI.

Mahesh Makhija, Partner and Technology Consulting Leader, EY India, said the proposed rules mark a clear step toward ensuring authenticity in digital content.

"Labelling AI generated material and embedding non-removable identifiers will help users distinguish real content from synthetic. This will serve as the foundation for responsible AI adoption, these measures will give businesses the confidence to innovate and scale AI responsibly," he mentioned.

The ministry has invited stakeholder feedback on the draft by November 6, 2025.

The government had earlier this month selected five projects under the programme launched by IndiaAI to advance real-time deepfake detection, strengthen forensic analysis and further other AI-related security issues.

- IANS