Meta’s independent Oversight Board has urged the company to overhaul its policies on AI-generated deepfakes, particularly during conflicts and crises when misinformation can spread quickly. The recommendation came after the board reviewed a fake AI-generated video posted on Facebook during the 2025 Israel-Iran conflict that falsely showed large-scale destruction in Israel. Fact-checkers later debunked the video, but it had already circulated widely online.
The board criticized Meta for failing to properly label or address the misleading video. It argued that the platform should have marked the content with a “High Risk AI” label because it could mislead the public during a sensitive geopolitical event. According to the board, the spread of synthetic media during wartime can influence public perception, political responses, and even international tensions if not handled quickly and transparently.
To address these risks, the Oversight Board recommended several major changes. These include stronger AI-detection tools, clearer labeling of AI-generated media, and better transparency about the origins of manipulated content through content provenance standards. The board also suggested that Meta create clearer rules specifically for AI-generated content and enforce them more consistently across platforms like Facebook and Instagram.
The case highlights a growing global concern about AI-driven misinformation during wars and political crises. As generative AI makes it easier to create realistic fake videos, experts warn that social media platforms must develop faster detection systems and stricter policies to prevent synthetic media from distorting public understanding of real-world events.
