Universal Music Group, Sony Music, and Warner Records have filed lawsuits against AI music-synthesis companies Udio and Suno, accusing them of widespread copyright infringement. The companies allegedly used recordings owned by the labels to train their AI models, enabling these AIs to generate new music based on text descriptions, such as “a dubstep song about Linus Torvalds.” The lawsuits, filed in federal courts in New York and Massachusetts, argue that this unauthorized use of copyrighted material could result in AI-generated music that competes directly with and potentially diminishes the value of human artists’ work.
Like other generative AI models, Udio and Suno rely on a diverse array of existing human-created works to train their neural networks to interpret text prompts and create music styles accordingly. The record labels criticize these companies for being vague about the sources of their training data.
Until recently, using copyrighted material without permission was common in machine learning, but the commercialization of AI technologies has prompted rightsholders to seek compensation. In this case, the record labels are seeking damages of up to $150,000 per song used in training.
The lawsuits highlight specific instances where AI-generated content allegedly mimics elements of well-known songs by artists like The Temptations, Mariah Carey, and James Brown. Furthermore, the music-synthesis models are claimed to replicate vocals resembling those of famous artists such as Michael Jackson and Bruce Springsteen.
This legal action is seen as the first of its kind targeting music-generating AI specifically, although the industry has been preparing for such challenges for some time. Sony Music, for instance, has already sent warning letters to over 700 AI companies and music-streaming services, prohibiting the use of its music for AI training. Moreover, over 200 musical artists signed an open letter urging AI companies to respect the rights of human artists.
The outcome of these lawsuits could have significant implications for the future development of generative AI in creative fields. It may lead to requirements for AI companies to license all musical training data, potentially making AI model development financially unfeasible for smaller startups like Udio and Suno. Conversely, major labels could leverage their resources to develop their own AI music generators, consolidating control over this emerging technology.
