YouTube is expanding its likeness-detection technology, which identifies AI-generated deepfakes, to a pilot group of government officials, political candidates and journalists. Members of the pilot programme will gain access to a tool that detects unauthorised AI-generated content and allows them to request its removal if they believe it breaches YouTube’s policies.
The technology was first introduced last year to approximately four million creators in the YouTube Partner Program after earlier testing phases.
Similar to YouTube’s existing copyright detection system, the tool analyses uploaded videos to identify AI-generated faces that imitate real people. Such technology is increasingly used to create deepfakes that attempt to spread misinformation or manipulate public perception by portraying well-known figures—including politicians and government officials—saying or doing things they never did in reality.
With the new pilot programme, YouTube says it aims to strike a balance between protecting freedom of expression and addressing the risks posed by AI systems capable of producing convincing digital replicas of public figures.
“This expansion is really about the integrity of the public conversation,” said Leslie Miller, YouTube’s Vice President of Government Affairs and Public Policy, during a press briefing ahead of the launch. “We know that the risks of AI impersonation are particularly high for those in the civic space. But while we are providing this new shield, we are also being careful about how we use it,” she said.
Miller added that not every detected match would automatically be removed after a request. Instead, YouTube will assess each case under its existing privacy policy guidelines to determine whether the content qualifies as parody or political commentary, which are protected forms of expression.
The company also said it supports stronger legal protections in the United States. In particular, it has backed the proposed NO FAKES Act in Washington, which aims to regulate the use of AI to create unauthorised recreations of an individual’s voice or visual likeness.
To access the tool, eligible participants in the pilot programme must first verify their identity by uploading a selfie alongside a government-issued ID. They can then create a profile, review potential matches detected by the system, and choose whether to request the removal of the content.
YouTube said it plans to expand the feature over time. Future versions could allow users to prevent content that violates their rights from being uploaded in the first place, or potentially enable them to monetise videos featuring their likeness, similar to how its existing copyright detection systems allow rights holders to claim revenue from uploaded material.
The company declined to disclose which politicians or public officials are participating in the initial tests but said its long-term aim is to make the technology widely available.
Your reaction
Discussion
What's your take on this story?
No comments yet — be the first to share your view.