Universal Music Group, Sony Music and Warner Records have sued AI music synthesis companies Udio and Suno for allegedly committing mass copyright infringement by using recordings owned by the labels to train music-generating AI models, Reuters reports. Udio and Suno can generate novel song recordings (e.g., “a dubstep song about Linus Torvalds”) based on text-based music descriptions.
The lawsuits, filed in federal courts in New York and Massachusetts, allege that the AI companies’ use of copyrighted material to train their systems could result in AI-generated music that directly competes with, and potentially devalues, the work of human artists.
Like other generative AI models, both Udio and Suno (which we reported on separately in April) rely on a wide selection of existing human-made artworks to teach a neural network the relationship between words in a written prompt and musical styles. The record labels rightly point out that these companies have intentionally kept the sources of their training data vague.
Until generative AI models went mainstream in 2022, it was common practice in machine learning to extract and use copyrighted information without asking permission. But now that the applications of these technologies have become commercial products themselves, rights holders are knocking on the door to claim their rights. In the case of Udio and Suno, the record labels are demanding damages of up to $150,000 per song used in training.
In the lawsuit, the record companies cite specific examples of AI-generated content that allegedly recreates elements of well-known songs, including “My Girl” by The Temptations, “All I Want for Christmas Is You” by Mariah Carey and “I Got You (I Feel Good)” by James Brown. They also claim that the music synthesis models can produce singing voices similar to those of famous artists such as Michael Jackson and Bruce Springsteen.
According to Reuters, this is the first case of lawsuits specifically targeting AI for music generation, but music companies and artists have been preparing for the challenges this technology could bring for some time.
In May, Sony Music sent warning letters to over 700 AI companies (including OpenAI, Microsoft, Google, Suno, and Udio) and music streaming services, banning AI researchers from using its music to train AI models. In April, over 200 musicians signed an open letter urging AI companies to stop using AI to “devalue the rights of human artists.” And last November, Universal Music filed a copyright infringement lawsuit against Anthropic for allegedly including artists’ song lyrics in its Claude LLM training data.
Much like the New York Times’ lawsuit against OpenAI over its use of training data, the outcome of the record labels’ new lawsuit could have profound implications for the future development of generative AI in creative fields, for example, requiring companies to license any musical training data they use to create music synthesis models.
Compulsory licensing of AI training data could make the development of AI models economically impractical for small startups like Udio and Suno – and judging by the open letter mentioned above, many musicians would welcome this potential outcome. However, such a development would not prevent major labels from eventually developing their own AI music generators themselves, so only large companies with lots of money would retain control of generative music tools for the foreseeable future.
