The short video platform TikTok is launching a pilot project to protect the likeness of its creators. The new feature, called Likeness Detection, is being tested on a limited group of creators from the US. This tool is designed to automatically search for AI-generated content that uses a specific blogger's face.
The mechanism is based on proactive scanning of uploaded videos. As soon as the system detects a deepfake or a fake account creating the illusion of a real person's presence, it immediately notifies the owner of the original face. The creator can then file a complaint about the violation, significantly simplifying the process of combating fraudulent schemes and identity theft.
Activating the feature requires passing strict verification through a third-party service, Jumio. The procedure includes two stages: a real-time selfie and scanning an identity document. It is important to note that, according to the company, TikTok does not store copies of the provided documents. Biometric data is used solely to create a digital template, allowing the system to subsequently identify matches across hundreds of thousands of videos.
Why This Is Critically Important for the Market
The market for deepfakes and AI clones is growing exponentially, and platforms are forced to adapt. TikTok, being one of the largest content platforms, becomes a prime target for scammers who create fake accounts in the name of popular bloggers. Likeness Detection is not just a security tool but a direct threat to the fake content industry. If the system proves effective, we will witness a standard for all social networks.
My professional commentary: At this point, it is only a test on a small sample, meaning incomplete protection for most users. However, the very fact of implementing an algorithmic facial detector is a powerful signal to regulators and investors. In the next 12 months, we will see how such solutions become a mandatory requirement for token listings of projects related to video and audio synthesis. Platforms that do not implement deepfake protection risk losing audience trust and facing lawsuits from creators.