Meta Urged to Boost Oversight of Fake AI Videos
The Oversight Board recommends Meta adopt a separate AI content policy, improve automated detection, and use digital watermarks to better handle deceptive AI content.
- On Tuesday, the Oversight Board urged Meta to create a dedicated AI content rule separate from its misinformation policy and recommended more frequent 'high risk' AI labels with stronger automated detection.
- A fake Haifa clip posted from the Philippines last year falsely claimed damage, users reported it, but Meta declined to add a 'high risk' label or remove it until the Oversight Board intervened.
- The board found Meta's system relies heavily on self-disclosure and rare escalated reviews, while also flagging dependence on third-party fact-checkers, trusted partners, and inconsistent use of digital watermarks.
- Meta said it welcomed the board's decision, will act on identical content when operationally possible, and has 60 days to respond after disabling three accounts linked to the page.
- Looking beyond Meta, the board suggested generative AI companies align on metadata and detection standards, warning AI misinformation spreads fastest during wars and urging wider use of Content Credentials.
15 Articles
15 Articles
Oversight Board urges Meta to toughen rules on AI-generated content and deepfakes
The latest criticism centers on a fake AI-generated video shared on Meta platforms last year that purported to show damage to buildings in the Israeli city of Haifa. It was posted by a user in the Philippines posing as a news source.Read Entire Article
Meta failed to flag AI video during 2025 Israel-Iran war, Oversight Board says
The Oversight Board rules Meta must do more to label AI-generated war slop as industrial-scale disinformation from the Israel-Iran conflict overwhelms current social media detection tools.
The Meta Group operates an independent, international body called the Oversight Board, comprised of representatives from various organizations and scientists. The board regularly reviews different aspects of Meta's platforms to provide suggestions for improvement. Now, in connection with... [See article: Meta and AI: Oversight Board calls for greater accountability]
Coverage Details
Bias Distribution
- 80% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
