A German court ruled that AI music company Suno infringed copyrights by using music from GEMA's catalog without a license for both AI model training and music generation. Suno disagrees with the ruling and is considering an appeal.
This ruling establishes a precedent in Germany requiring AI companies to secure licenses for copyrighted material used in both training AI models and generating new content, impacting the broader AI industry's use of existing creative works.
A German court has ruled against AI music startup Suno in a copyright lawsuit brought by music rights organization GEMA. The Munich Regional Court found that Suno violated copyright law by using music from GEMA's catalog to train its AI models and by reproducing protected works without a license.
The ruling requires AI companies to obtain licenses for the commercial use of GEMA's repertoire, covering both AI model training and the generation of music. The lawsuit centered on six songs, including "Daddy Cool," “Rasputin,” “Forever Young,” and “Mambo No. 5.” GEMA accused Suno of training its AI on copyrighted songs without permission.
In a statement, Suno said it built its platform to help people create new music, not reproduce existing songs. The company stated the ruling is based on a misunderstanding of how its technology works and that it is considering its legal options. "Our tools give people the ability to create new songs... From the beginning, we trained our models to create new songs, not reproduce existing ones, and built protections into our platform. We disagree with today’s ruling—which rests on a fundamental mischaracterization of how Suno’s technology works, how it is used and how U.S. law applies—and are evaluating all available options, including an appeal,” Suno said.
The decision comes about eight months after GEMA won a separate copyright case against OpenAI, in which a Munich court ruled that ChatGPT unlawfully reproduced copyrighted song lyrics. The ruling also comes as Suno faces mounting legal scrutiny over how it built its AI models. Earlier this month, leaked source code showed the company's training data included more than 113,000 hours of YouTube Music, 62,000 hours from Pond5, and 12,000 hours from Deezer, supporting claims by the music industry that Suno trained its AI on copyrighted recordings without permission, allegations that Suno continues to fight in court.