YouTube is taking a major step to protect creators and celebrities from unauthorized AI-generated content. The platform is teaming up with the Creative Artists Agency (CAA) to introduce new tech that will help identify and remove AI replicas of faces, voices, and likenesses.
The collaboration will begin with a test program for celebrities and athletes in early 2025. YouTube plans to expand this to top creators and other talent managers soon after. These tools will allow individuals to detect AI-generated versions of themselves on YouTube, giving them more control over their image and voice.
YouTube:
“We’re excited to work with CAA to ensure creators and talent have the tools to manage AI depictions of their likeness at scale.”
This development follows YouTube’s announcement in September about tools to handle AI-generated content featuring creators’ voices and images.
CAA’s CAAVault plays a key role in this partnership. The Vault stores digital versions of celebrities, including their faces and voices, which will be used to spot AI content on YouTube.
In addition to likeness protection, YouTube is also advancing technology to identify AI-generated singing voices. This “synthetic-singing identification” system will help detect AI content that mimics an artist’s voice.
YouTube has already allowed music labels to remove AI-generated music using artists’ voices and introduced a requirement for creators to label AI-generated content on their channels.
With these new tools, YouTube is setting a clear precedent for protecting the digital rights of creators, celebrities, and musicians in an era of rapidly advancing AI technology. The platform aims to provide a safer space for creators to thrive without the threat of AI exploitation.
