Calabi Labs · Guide · 2026-05-29
Govt asks social media platforms to label take down ai generated deepf
Government Asks Social Media Platforms to Label & Take Down AI-Generated Deepfakes
What is the new government directive? The Indian government has directed social media platforms—including Facebook, Instagram, YouTube, X (Twitter), and others—to identify, label, and remove AI-generated deepfake content from their platforms. This directive falls under the IT Rules and aims to curb the spread of synthetic media that can mislead, defame, or manipulate the public.
What Exactly Are Platforms Required to Do?
| Requirement | Details |
|---|---|
| Label AI-generated content | All posts, videos, images, or audio created or significantly modified using AI must carry a visible label |
| Take down deepfakes | Content identified as deepfakes—especially those involving impersonation, defamation, or misinformation—must be removed within 24–72 hours of a valid complaint |
| Report compliance | Platforms must submit periodic compliance reports to the government |
| User grievance mechanism | Each platform must have a visible mechanism for users to report deepfake content |
| Tracing origin | Platforms are required to help trace the originator of problematic deepfake content when required by law enforcement |
Why Is This Happening Now?
Deepfake technology has become increasingly accessible and dangerous. Recent incidents include:
- Political deepfakes — fabricated videos of leaders making false statements
- Celebrity impersonation scams — AI-cloned voices and faces used in financial fraud
- Non-consensual intimate imagery — AI-generated explicit content using real people's faces
- Enterprise impersonation — deepfake audio used to authorize fraudulent bank transfers
The government's move is a direct response to the exponential rise in AI-generated synthetic media that threatens individual privacy, democratic processes, and public safety.
Which Laws Govern This?
- IT Act, Section 66E — penalizes publishing or transmitting obscene material including morphed images
- IT Rules 2021 (Intermediary Guidelines) — mandate due diligence and grievance redressal
- BNS Section 66C/66D — penalizes identity theft and cheating through digital means
- DPDP Act 2023 — protects digital personal data and addresses synthetic media misuse
What Has the Government Said Exactly?
> Platforms must ensure that any content that is synthetically created using AI or any other tool is prominently labeled. Failure to comply will result in the platform losing its safe harbour protection under Section 79 of the IT Act.
Multiple ministry-level advisories have been issued to major platforms, with explicit timelines for compliance.
How Are Platforms Responding?
| Platform | Status |
|---|---|
| Meta | Added "AI-generated" labels on Facebook & Instagram; expanded deepfake detection |
| Google/YouTube | Running AI content classifiers; requires creators to disclose AI-generated content |
| X (Twitter) | Updated policy to label and remove manipulative synthetic media |
| Others | Under active compliance review with the Ministry of Electronics & IT (MeitY) |
What Can You Do Right Now?
- If you see suspected deepfake content — report it using the platform's built-in reporting tool
- If you're a victim — file a complaint on the National Cyber Crime Reporting Portal (cybercrime.gov.in) or your local police station
- If you are a content creator — always disclose when content is AI-generated using platform label tools
- Stay informed — follow MeitY advisories for updated compliance requirements
How to Check If Content is AI-Generated
- Look for labels — Meta, Google, and X now show "AI-generated" badges on detected content
- Check metadata — legitimate AI-generated content often carries C2PA content credentials
- Use detection tools — tools like Calabi can analyze whether video or image content is synthetically generated
- Look for tells — unnatural blinking, blurred hands, flickering edges, or audio-video sync issues
The Bottom Line
The government has drawn a clear line: AI-generated deepfakes are not exempt from content moderation. Platforms must label synthetic media, act on complaints fast, and provide traceability. Non-compliance carries real legal consequences.
For individuals, the best defense is awareness, reporting, and using reliable AI-detection tools to verify suspicious content before sharing it.
Try Calabi free at calabilabs.com — 3 cleans, no card.