YouTube is rolling out its AI-powered likeness detection tool to politicians, journalists, and government officials, helping them flag and remove unauthorized deepfakes from the platform. This comes as AI-generated content, especially deepfakes, grows more convincing and potentially disruptive-just as major elections approach.
The tool allows verified users to request takedowns of videos that use their likeness in AI-generated content. It functions like Content ID, but targets faces instead of copyrighted music or video. “As AI-generated content evolves, the individuals at the center of these conversations need reliable tools to protect their identities,” said Amjad Hanif and Leslie Miller, YouTube VPs, in a company blog post.
Who gets access-and who decides?
YouTube hasn’t disclosed exactly which politicians or journalists can use the tool, sidestepping questions about high-profile figures like Donald Trump. The company only confirmed a “broad international rollout” planned over the coming weeks and months, keeping specifics under wraps for now.
Eligible users must verify their identity with a video selfie and government ID. YouTube insists this data is used solely for verification-not to train Google’s AI models. Once verified, users can scan for videos featuring their likeness and submit takedown requests. But flagged videos aren’t automatically removed; YouTube weighs public interest, parody, and satire before deciding. “We’ll continue to carefully evaluate these exceptions when we receive requests for removal,” the company said.
Why it matters for creators and viewers
AI deepfakes are easier than ever to create, thanks in part to tools like Veo 3 and new AI editing features built into YouTube itself. This raises the risk of fake videos going viral-especially those targeting public figures. For creators, the new tool means increased scrutiny if their videos use politicians’ or journalists’ faces, even in parody or commentary.
For viewers, this could reduce exposure to misleading or manipulated videos but also lead to more content being flagged or removed if it crosses YouTube’s evolving guidelines. The platform says it’s balancing identity protection with free expression, though how those decisions are made remains unclear.
The bottom line
- YouTube’s deepfake tool is expanding to cover politicians, journalists, and officials-just in time for election season.
- Access remains limited, with no confirmation on who’s included in the initial rollout.
- Verification is required, and flagged videos aren’t guaranteed removal.
- Creators using public figures’ likenesses should expect tighter oversight.