The Meta Oversight Board has complained that Meta’s methods for detecting deepfakes are not robust enough to address the rapid spread of disinformation during armed conflict. The semi-independent committee that guides Meta’s content curation practices is calling on the company to reshape how it identifies and labels AI-generated content on Facebook, Instagram , and Threads.
The call for action stemmed from an investigation into a fake AI video depicting alleged damage to buildings in Israel that was shared on Meta’s social platforms last year. The Oversight Board stresses that its recommendations are particularly relevant now, given the “massive military escalations” across the Middle East this week.
According to the Board's announcement, access to accurate and reliable information is vital for people's safety amid the increased risk of using AI tools to spread misinformation.
See also: Meta will enable competitive AI chatbots on WhatsApp
“The Board’s findings highlight that Meta’s current system for appropriately flagging AI content relies too heavily on self-disclosure of AI usage and tiered review and does not meet the realities of today’s online environment,” the Meta Oversight Board.
Challenges in Deepfakes Detection and Cross-Platform Propagation
The case also highlights the challenges with cross-platform dissemination of such content, as the content appears to have originated on TikTok before appearing on Facebook, Instagram , and X.This phenomenon demonstrates the need for a coordinated approach to tackling deepfakes across all social media platforms.
See also: Lawsuit against Meta for Ray-Ban Meta glasses
The recommendations issued by the Meta Oversight Board include pressuring Meta to improve its existing misinformation rules to address misleading deepfakes, and establish a new, separate community standard for AI-generated content. Meta is also being asked to develop better AI detection tools, be transparent about penalties for AI , and scale up efforts to flag AI content.
Technical Solutions: C2PA and Advanced Signage
The latter includes ensuring that “High-Risk AI” labels are added to synthetic images and videos more frequently, and improving the adoption of C2PA (also known as Content Credentials) so that information about AI is “clearly visible and accessible to users.”
C2PA of media manipulation, allowing users to verify the origin and authenticity of digital content. However, the Board is concerned about reports that Meta “inconsistently applies” the C2PA “tools AI.”
See also: Meta: Lawsuits against advertisers for celeb-bait scams
Broader Implications and Future Challenges
Deepfakes are increasingly being used as a tool for gender-based harassment, targeting women to intimidate them both online and offline. The Oversight Board highlighted that victims of non-consensual deepfake erotic images – particularly those without public profiles – face the burden of constantly reporting images, which is both traumatic and resource-intensive.
