YouTube is expanding access to its AI-powered deepfake detection tool

In the coming weeks, YouTube will make a specialized tool for detecting faces in AI-generated videos available to all content creators aged 18 and older. Previously, the technology was available only to a limited group of people, including journalists, politicians, and participants in the partner program.

This was reported by The Verge. According to company representative Jack Maylon, expanding access is intended to provide every user with the same level of protection against the unauthorized use of their image by brands or third parties in manipulative videos.

The technology was first introduced in 2024 and launched in late 2025 for owners of monetized channels with at least 1,000 subscribers, but now the security feature will reach a much wider audience. To activate protection, users must complete a verification process in YouTube Studio on a computer, which includes scanning a QR code, providing a copy of an ID, and recording a short selfie video. Once identity is confirmed, the platform’s algorithms will automatically scan all newly uploaded videos for visual matches with the applicant’s face.

All matches detected by the system will appear in a special “Similarity” section of the “Content Detection” menu, where the profile owner can personally review the video and submit a formal request for its removal. In the request form, the user must describe in detail the nature of the unauthorized use of their likeness. Although the platform’s interface already includes a separate question regarding potential voice cloning, the current tools do not yet have the technical capability to independently detect audio matches, given the general development of AI technologies following Russia’s full-scale invasion of Ukraine.