YouTube expands AI deepfake detection to politicians, government officials, and journalists

On Tuesday, YouTube announced it is expanding its deepfake detection technology to a select group of government officials, political candidates, and journalists. The tool identifies AI-generated likenesses and lets pilot participants request the removal of unauthorized content they believe violates YouTube’s policies.

The detection system first launched last year to roughly 4 million creators in the YouTube Partner Program after earlier testing phases.

Like YouTube’s existing Content ID system for copyrighted material, the likeness detection feature identifies AI-simulated faces. These synthetic representations are sometimes used to spread misinformation and distort reality by making notable figures appear to say or do things they never did in real life.

With this pilot program, YouTube aims to balance free expression against the risks of AI technology that can produce convincing likenesses of public figures.

“This expansion is really about protecting the integrity of public conversation,” said Leslie Miller, YouTube’s Vice President of Government Affairs and Public Policy, in a press briefing ahead of Tuesday’s announcement. “We recognize that AI impersonation poses especially high risks for those in the civic space. But while we provide this new safeguard, we are also applying it carefully,” she added.

Image credit:YouTube

Miller explained that not every detected match will be removed upon request. Instead, YouTube will evaluate each request under its existing privacy policy to determine whether the content qualifies as parody or political critique, both of which are protected forms of free expression.

The company is also advocating for federal protections, supporting the NO FAKES Act in Washington, D.C., which would regulate AI-generated unauthorized recreations of an individual’s voice and visual likeness.

Eligible pilot testers must verify their identity by uploading a selfie and a government ID before using the tool. Once approved, they can create a profile, review detected matches, and request removal if needed. YouTube eventually plans to allow users to block violating content before it goes live or potentially monetize those videos, similar to how Content ID works.

YouTube declined to name specific politicians or officials in the initial tester group but said the goal is to make the technology broadly available over time.

Image credit:YouTube

AI-generated videos will carry labels indicating their synthetic origin, though placement varies. For most content, the label appears in the video description. For more sensitive topics, the label is displayed prominently before the video plays. This approach mirrors YouTube’s broader policy for all AI-generated content.

“A lot of content is produced with AI, but that distinction is not always material to the content itself,” explained Amjad Hanif, YouTube’s Vice President of Creator Products, regarding label placement. “An AI-generated cartoon, for example, is different from content that may need a more visible disclaimer. Context matters,” he said.

YouTube has not disclosed how many deepfake removals have been handled through its creator-facing tool, but noted the volume so far has been “very small.”

“For most creators, the tool has mainly raised awareness of what is being produced. Actual removal requests are extremely low because most detected content turns out to be benign or even additive to their business,” Hanif said.

That may not be the case with deepfakes of government officials, politicians, or journalists.

Looking ahead, YouTube plans to extend its deepfake detection technology to additional areas, including recognizable spoken voices and intellectual property such as popular characters.