TikTok Tests AI Likeness Detection Tool

TikTok began testing an opt-in tool designed to scan for AI-generated likenesses and enable creators to report them to the platform, as observed by social media consultant Matt Navarra. This new feature is currently undergoing initial testing with a limited group of US-based creators, according to TikTok US spokesperson Zachary Kizer. The tool aims to address concerns surrounding the use of artificial intelligence to create deepfakes or impersonate individuals without consent.

The development follows a broader industry trend where social media and content platforms are grappling with the proliferation of AI-generated content. YouTube, for instance, has been developing a comparable tool to help its creators manage the use of their likeness in AI-generated videos. The specifics of TikTok's detection mechanism and its reporting process have not yet been fully disclosed, but the company's spokesperson confirmed the ongoing testing phase. The opt-in nature of the tool suggests that creators will need to actively enable it for their content to be scanned.

This initiative by TikTok underscores the growing need for robust mechanisms to identify and manage synthetic media. As AI technology advances, the ability to generate highly realistic but artificial content becomes more accessible, posing challenges for authenticity and user trust. By providing creators with a means to report potential AI impersonations, TikTok is attempting to empower its user base and maintain a more controlled environment for content creation and consumption. The company has not provided a timeline for a wider rollout beyond the current testing phase with select US creators.