Skip to main content

Musical Artists' Warning about AI Use

Dark music

This week, more than 200 musical artists — including Billie Eilish, Nicki Minaj, Katy Perry, and Stevie Wonder — signed an open letter, urging tech platforms to stop devaluing music. The letter asks companies to stop doing “irresponsible AI,” or to stop producing content that feels like it's been stolen from them. The artists don't want AI platforms and developers to be able to use their content without their permission. 

While the letter acknowledges that “AI has enormous potential to advance human creativity,” it argues that some platforms and developers are using AI to sabotage human creativity. But think for a moment about how humans learn to create music. If someone followed a band religiously and was very creative or musical, they would learn the troupes that define the voice or style. If they then started producing their own music based on what they learned, we wouldn't call that plagiarism or a copyright violation. We’d call that a style. 

Generative AI works in a similar way. We're able to train a model on the music of a particular musician, band, or even a style and have the model reproduce things in that style, band, or artist. 

We don't want to stop the research into these technologies. We want to stop irresponsible business practices. Using an artist’s work to train an AI system and then selling what is produced by that system must include payment for the artist. This is about good business sense. 

In the long run, we want systems to be musical and interesting and work with us. The only way we can make that happen is if we allow the technology to move forward. But we've got to make sure that the business is aligned with the people who are actually making the business possible. Not just the technology, but the artists themselves. 

Kristian Hammond
Bill and Cathy Osborn Professor of Computer Science
Director of the Center for Advancing Safety of Machine Intelligence (CASMI)
Director of the Master of Science in Artificial Intelligence (MSAI) Program

Back to top