TikTok creators may soon have a new way to protect themselves from fake AI videos. The social media company is testing a tool that helps creators find and report unauthorized copies of their faces and voices on the platform.
Currently, some creators in the United States are testing this new feature. To use it, creators must first prove who they are. They do this by sharing a photo of their ID and taking a real-time selfie scan through a security company named Jumio. TikTok stated that it does not keep these ID documents. It only uses the facial data to look for copies of the creator.
Once a creator is verified, TikTok's system automatically scans the platform for AI-generated videos that look or sound like them. If the tool finds anything, the creator can review the videos and report any unauthorized posts or accounts to TikTok.
The threat of "deepfakes"—realistic but fake videos made by artificial intelligence—has become a major concern for online creators. TikTok is not the only company trying to solve this. YouTube recently released a similar tool that is available to all adult users, showing that platforms are under pressure to protect people's identities in the age of AI.