In what’s being touted as the first partnership of its kind, CAA is collaborating with YouTube on a program promising to let actors, athletes and other talent fight back against AI-generated fakes uploaded to the video platform.
YouTube’s message to Hollywood is that it’s trying to be a good partner to creative industries by taking “proactive steps to build responsible AI.” Under the partnership, CAA clients, “many of whom have been impacted by recent AI innovations,” will have access to YouTube’s early-stage likeness management technology, which is designed to identify and manage AI-generated content featuring faces on YouTube “at scale,” according to the companies.
In early 2025, YouTube said, it will begin testing its newest likeness-management technology with “leading celebrity talent, including award-winning actors and top athletes from the NBA and NFL.” At this stage, the companies aren’t revealing specific names of CAA talent expected to participate in the program.
YouTube’s tool will identify unauthorized AI replicas and let talent submit requests to remove them. “Importantly, this is the first step of a larger testing effort,” YouTube said in a blog post Tuesday. “Over the next few months, we’ll announce new testing cohorts of top YouTube creators, creative professionals and other leading partners representing talent.” YouTube said the initial participants will provide “critical feedback” to help it build its detection systems and refine the controls, prior to the video platform making the technology widely available.
Bryan Lourd, CAA’s CEO and co-chairman, said he began talking with YouTube CEO Neal Mohan earlier this year “about the importance of creating a responsible AI ecosystem that protects artists and their IP rights, while unlocking new possibilities for creative expression.”
“At CAA, our AI conversations are centered around ethics and talent rights, and we applaud YouTube’s leadership for creating this talent-friendly solution, which fundamentally aligns with our goals,” Lourd said in a statement. “We are proud to partner with YouTube as it takes this significant step in empowering talent with greater control over their digital likeness and how and where it is used.”
News of the CAA partnership follows YouTube’s announcement in September that it was developing a new system to detect and manage AI-generated content showing faces. In addition, YouTube has developed synthetic-singing identification technology within Content ID that will let partners automatically detect and manage AI-generated content on YouTube that simulates singing voices. On Monday, YouTube announced that it will roll out an update to let creators and rights holders choose to allow third-party companies to use their content to train AI models.
“At YouTube, we believe that a responsible approach to AI starts with strong partnerships. We’re excited to collaborate with CAA, an organization that shares our commitment to empowering artists and creators,” Mohan commented. “In the days ahead, we’ll work with CAA to ensure artists and creators experience the incredible potential of AI while also maintaining creative control over their likeness. This partnership marks a significant step toward building that future.”
Last year, CAA launched theCAAvault, which scans, captures and securely stores clients’ digital likenesses (face, body and voice). According to the agency, the service addresses issues relating to consent, credit and compensation by providing artists “complete ownership, control and permission over their digital identities.” The company has formed strategic partnerships for theCAAvault with multiple technology companies, including Clear Angle Studios, Metaphysic, Deep Voodoo and Veritone. CAA said it also actively engages with policymakers and others stakeholders to help shape AI policies and initiatives to protect creators and their intellectual property, and most recently, publicly supported the No Fakes Act legislation recently introduced in Congress.
YouTube said CAA’s “expertise and commitment to responsible innovation make them a strong first partner” for its AI-generated detection and management system.
RELATED: ‘Clean’ AI Video Model to Launch in Early 2025 Targeting Hollywood Clients
{Categories} _Category: Implications{/Categories}
{URL}https://variety.com/2024/digital/news/caa-youtube-talent-ai-deepfakes-remove-1236251470/{/URL}
{Author}Todd Spangler{/Author}
{Image}https://variety.com/wp-content/uploads/2024/12/CAA-YouTube-AI-Deepfakes.jpg?w=1000&h=563&crop=1{/Image}
{Keywords}News,CAA,YouTube{/Keywords}
{Source}Implications{/Source}
{Thumb}https://variety.com/wp-content/uploads/2024/12/CAA-YouTube-AI-Deepfakes.jpg{/Thumb}