Chubo / Shutterstock
The legal battle over artificial intelligence training data has taken a significant contractual turn. Tech giant Google has formally moved to dismiss a proposed class action copyright lawsuit brought against it by a group of independent musicians regarding its Lyria 3 artificial intelligence music model.
In a legal motion filed in the United States District Court for the Northern District of Illinois, Google urged the court to throw out the complaint with prejudice. The company’s core defense rests on a strict interpretation of YouTube’s Terms of Service. Google argues that because the independent creators chose to upload and monetize their songs on the video platform, they automatically granted the service and its corporate affiliates a worldwide, royalty free, and transferable license to reproduce, distribute, and prepare derivative works using that content. According to the filing, this existing contractual agreement fully covers the exact platform conduct the artists are now attempting to sue over.
Furthermore, Google’s legal team at Quinn Emanuel Urquhart and Sullivan asserted that the musicians are relying on an entirely unsupported hypothesis that the system specifically trained on their individual songs. The tech firm noted that despite the extensive length of the initial complaint, the plaintiffs failed to identify any actual infringing audio output or prove concrete harm.
Google also targeted separate aspects of the lawsuit, arguing that the musicians lack proper legal standing under the Digital Millennium Copyright Act, failed to establish false endorsement claims under the Lanham Act, and relied purely on speculation regarding state biometric privacy violations. While other prominent generative AI developers continue to fight high profile music industry lawsuits using traditional fair use copyright arguments, Google’s strategy marks a major defensive shift by leveraging explicit user platform agreements to shield its ongoing corporate technology development.
About the Author
