Google Defends AI Training Rights Through YouTube Terms in Music Copyright Lawsuit

Google has defended its right to potentially use music uploaded to YouTube for artificial intelligence training, citing the platform’s terms of service in an ongoing copyright lawsuit involving its AI music model, Lyria 3.

The legal dispute stems from a lawsuit filed earlier this year by a group of independent artists, songwriters, and producers who allege that Google trained its AI-powered music generator on copyrighted songs without providing compensation to creators.

Lyria 3, introduced through Gemini in February, is capable of generating songs based on user prompts. Soon after its launch, concerns emerged over the data used to train the model, leading to allegations that artists’ works had been exploited without permission.

In a recent court filing, Google argued that creators who uploaded music directly to YouTube had already granted the company extensive rights to use that content. According to the filing, YouTube’s terms of service provide a “broad licensee” allowing Google to utilise uploaded material, including for purposes such as training artificial intelligence systems.

However, Google stopped short of confirming whether it actually used YouTube-uploaded music to train Lyria 3. Instead, the company’s legal argument focuses on the permissions granted through YouTube’s user agreement.

Google’s motion, filed by litigation firm Quinn Emanuel reads, “Even accepting their untested allegations as fact, the complaint cannot stand. Plaintiffs each granted YouTube, and Google — which provides the service — a broad license to use the uploaded content. That license, present in YouTube’s terms of service, authorised the conduct alleged in the complaint.”

In essence, Google maintains that even if such content were used for AI training, creators had already authorised that possibility by accepting YouTube’s terms.

The filing further references a specific clause from the platform’s agreement, stating: “By providing content to the service, you grant to YouTube a worldwide, non-exclusive, royalty-free, sublicensable and transferable license to use that content (including to reproduce, distribute, prepare derivative works, display and perform it) in connection with the service and YouTube’s (and its successors’ and affiliates’) business.”

The company’s defense differs from strategies adopted by other AI firms facing copyright-related lawsuits. Companies such as Anthropic, Suno, and Udio have largely argued that training AI systems on copyrighted content falls under fair use because the resulting technologies are transformative.

Google, by contrast, is relying on the claim that it already possesses a valid license through YouTube’s terms for content uploaded by users. If courts accept this interpretation, it could significantly strengthen the legal foundation for AI training using user-generated content hosted on major platforms.

The lawsuit also argues that Google’s role extends beyond merely developing a music-generation model. Plaintiffs claim the company benefits from its control over YouTube and Content ID, giving it unique access to creative works. As the complaint states, “Google didn’t just have access to Plaintiffs’ music; it operated the infrastructure through which much of that music reached the world.”

Adding to its legal challenges, Google is also facing another lawsuit brought by journalists, podcasters, and audiobook narrators who allege their voices were used to train AI models without authorisation.

The implications of the current case may not apply equally to all music available on YouTube. Many recordings distributed through major record labels and publishers are governed by separate licensing agreements that may contain distinct provisions regarding AI training. Some newer agreements reportedly include AI-specific protections.

During an investor call in October, Universal Music Group chief executive Lucian Grainge said the company’s renewed deal with YouTube had secured “really important guardrails and protection for our artists and writers around Gen AI content."

Google has not publicly revealed the exact datasets used to train Lyria 3. When unveiling the model, a company spokesperson stated that it had been trained on music that Google had “a right to use under our terms of service, partner agreements, and applicable law.”