TikTok Tests AI Tool to Detect Likeness Fakes
▼ Summary
– TikTok is testing an opt-in tool that scans for AI-generated likenesses and lets creators report them, initially available to some US creators.
– The tool requires creators to verify their identity through Jumio using a real-time selfie scan and ID check, though TikTok does not retain ID documents.
– After verification, TikTok’s system scans for AI content potentially using a creator’s likeness, allowing creators to review and report unauthorized posts or accounts.
– YouTube has been developing a similar tool and recently made it available to all adult users.
– The test was spotted by social media consultant Matt Navarra and confirmed by TikTok US spokesperson Zachary Kizer.
TikTok has begun testing a new opt-in tool designed to help creators detect and report unauthorized AI-generated content that mimics their likeness. The feature was first spotted by social media consultant Matt Navarra, and TikTok US spokesperson Zachary Kizer confirmed to The Verge that the initial rollout involves “some” creators in the United States. This move follows a similar initiative from YouTube, which recently extended a comparable tool to all adult users on its platform.
To access the feature, eligible creators must first verify their identity through a third-party service called Jumio. The process requires a real-time selfie scan and an ID check. Kizer emphasized that TikTok does not store ID documents and only uses facial data for the purpose of likeness matching and identifying potential unauthorized uses of a creator’s image.
Once verified, TikTok’s system automatically scans the platform for AI-generated content that may feature the creator’s likeness. Creators can then review the flagged content and choose to report any posts or accounts they believe are using their image without permission. This test represents a significant step in addressing the growing challenge of deepfakes and synthetic media on social platforms.
(Source: The Verge)
