Deepfakes Are Racking Up Views — Social Media Barely Labels Them
Only 1 in 3 fake videos gets an AI label — and they reach 160,000 views.
Researchers audited AI labelling on Instagram, TikTok, X and YouTube, drawing on the European Commission's July 2026 guidelines on deepfakes. All four platforms now apply labels automatically, and a greater share of labels are platform-applied than in earlier audits. But only 33% of expert-identified deepfakes in systemic risk contexts carried a platform-applied AI label — while reaching a median of 160,000 views. In controlled uploads of AI-generated content carrying standard AI provenance signals, platforms labelled only 61% of uploads, and commonly strip those signals after uploading. The labels exist, but coverage remains incomplete where labelling arguably matters most.
- Instagram, TikTok, X and YouTube all label AI content automatically now, but only 33% of expert-flagged deepfakes got a label.
- Those unlabelled deepfakes reached a median of 160,000 views, so plenty of people saw them without any warning.
- Platforms stripped the hidden "made by AI" digital fingerprints from 39% of uploads, making fakes harder to trace later.
Why It Matters
You can't trust a missing AI label — check before you share or believe what you see.