Meta’s Oversight Board rebuked the social-media giant for failing to label an AI-generated video falsely depicting wartime damage in Haifa, Israel, urging the company to overhaul how it handles manipulated media. The 21-member body said Meta’s reliance on user self-disclosure and complaint-driven moderation isn’t adequate amid a surge of deceptive AI videos tied to global conflicts, which it warned risks eroding trust in information. The video in question amassed nearly one million views before the board intervened; Meta initially said it didn’t meet the threshold for imminent harm and declined to label it. Meta now says it will label the clip within seven days and will follow the board’s guidance for “identical” content in the same context. The dispute highlights long-running tensions over the board’s influence as Meta loosens some moderation approaches and grapples with the scale and speed of AI-generated content.
Related article:
