Google DeepMind's latest advancements in AI music generation arrive at a critical juncture, showcasing impressive new creative tools while simultaneously navigating a legal minefield where competitors face massive copyright lawsuits from the recording industry.
The company recently enhanced its Music AI Sandbox platform with its updated Lyria 2 model and new RealTime features, expanding access for creators. The move comes just months after the music industry took decisive legal action against other AI music startups over alleged widespread copyright violations.
Legal battle lines drawn: The core conflict intensified in June 2024 when the Recording Industry Association of America, representing major labels like Sony Music Entertainment, UMG Recordings, and Warner Records, filed landmark lawsuits against AI music generators Suno and Udio.
- The suits allege mass copyright infringement through the unlicensed copying and use of vast quantities of sound recordings to train their AI models, seeking damages of up to $150,000 per infringed work.
- Both Suno and Udio formally responded, admitting their models were trained on copyrighted materials but arguing their actions constitute lawful "fair use"
Google charts a careful course: Against a litigious backdrop, Google appears to be treading carefully. The company says its history of working with the music industry, stating, "Ever since launching the Magenta project, in 2016, we’ve been exploring how AI can enhance creativity... always hand-in-hand with the music community." All music generated by Lyria 2 and Lyria RealTime models is watermarked using SynthID technology, to identify AI-generated audio even after modifications. Access to the enhanced Sandbox remains limited, currently requiring US-based creators to sign up via a waitlist.