Meta's deepfake moderation isn't good enough, says Oversight Board

"The Board's findings highlight that Meta's current system to properly label AI content is overly dependent on self-disclosure of AI usage and escalated review and does not meet the realities of today's online environment. The case also highlights the challenges with cross-platform proliferation of such content, with the content appearing to have originated on TikTok before appearing on Facebook, Instagram, and X."

"Meta's methods for identifying deepfakes are not robust or comprehensive enough to handle how quickly misinformation spreads during armed conflicts like the Iran war. Access to accurate, reliable information is vital to people's safety amid the heightened risk of AI tools being used to spread misinformation during massive military escalations throughout the Middle East."

The Meta Oversight Board determined that Meta's current deepfake identification and labeling systems are inadequate for addressing misinformation during military escalations. An investigation into a fake AI video of Israeli building damage revealed that Meta's approach relies too heavily on self-disclosure of AI usage and lacks robust detection capabilities. The content spread across multiple platforms, originating on TikTok before appearing on Facebook, Instagram, and X. The Board recommends Meta improve misinformation rules for deceptive deepfakes, establish separate community standards for AI-generated content, develop better detection tools, increase transparency about policy violations, and expand AI content labeling efforts including High-Risk AI labels and Content Credentials adoption.

Read at The Verge