YouTube rolls out an AI feature to protect creators from deepfakes

We’re yet to see AI solving the greatest issues of humanity, but we’re seeing more and more AI-generated videos, including elaborate deepfakes. Considering that, more people may want access to tools that would help them protect their likenesses. Today, YouTube has officially launched such a tool.

YouTube launches a likeness detection tool to fight deepfakes

YouTube officially launched its new AI likeness detection tool today. The feature allows creators to detect AI-generated or altered videos that feature their facial likeness without authorization. The creators will be able to manage and request the removal of the detected videos.YouTube’s goal is to help creators safeguard their identities and prevent audiences from being misled by AI-generated deepfake videos. The tool will be available to YouTube creators who are part of the YouTube Partner Program. They will start getting access to the features in the coming weeks, and by January 2026, all monetized creators will have access.

How can you use the likeness detection tool?

The new feature is available in the content detection tab in YouTube Studio. To use it, creators need to complete an identity verification process that requires a photo ID and a selfie video. Once the process is completed, creators will receive an alert about all the AI-generated videos that use their likenesses.

YouTube Studio will feature a list of all unauthorized videos, showing information about the channel, title, and views of every video. The tool will highlight the segment of the videos where the creator’s likeness is used, and it’ll allow them to submit a request for the removal of the video.

Only the beginning

With the proliferation of apps like Sora, deepfakes are going to be a growing problem for everyone. YouTube is making a good first step, but famous people are not the only ones who could suffer from AI slop with their likenesses. Not to mention that providing a photo ID and my biometric data to keep myself from bad AI slop doesn’t sound good.I’d prefer to see more restrictions for AI-generated videos, like a separate feed, or at least a very clear warning that a clip is likely fake. That won’t stop deepfakes, but it could spare people from explaining that a video is fake.

“Iconic Phones” is coming this Fall!

Relive the most iconic and unforgettable phones from the past 20 years! Iconic Phones is a stunningly illustrated book we’ve been crafting for over a year—and it’s set to launch in just a couple of months!Iconic Phones: Revolution at Your Fingertips is the ultimate coffee table book for any phone enthusiast. Featuring the stories of more than 20 beloved devices, it takes you on a nostalgic journey through the mobile revolution that transformed our world. Don’t miss out—sign up today to lock in your early-bird discount!

Visible+ Pro – up to $135 savings on Verizon’s fastest 5G

We may earn a commission if you make a purchase

Check Out The Offer

View Full Bio

Ilia, a tech journalist at PhoneArena, has been covering the mobile industry since 2011, with experience at outlets like Forbes Bulgaria. Passionate about smartphones, tablets, and consumer tech, he blends deep industry knowledge with a personal fascination that began with his first Nokia and Sony Ericsson devices. Originally from Bulgaria and now based in Lima, Peru, Ilia balances his tech obsessions with walking his dog, training at the gym, and slowly mastering Spanish.

Read the latest from Ilia Temelkov

Trending Products