MUNICH, GERMANY — A court in Munich has ruled that OpenAI infringed German copyright law by including song lyrics in the training data for its language models and ordered the company to pay compensation to the writers.
Judges said on Tuesday that both the internal storage of the texts and their later appearance in ChatGPT outputs violated the rights of the composers. “Both the memorisation in the language models and the reproduction of the song lyrics in the chatbot’s outputs constitute infringements of copyright law,” the court wrote.
The claim was brought in November 2024 by GEMA, the German collective that represents more than 100,000 composers, songwriters and music publishers. The case covered nine German-language songs whose lyrics appeared, unlicensed, in the chatbot’s answers.
OpenAI, which is based in San Francisco, argued that its systems do not copy or store specific works but adjust internal settings to reflect patterns learned from vast datasets. The company also contended that users, not OpenAI, are responsible for any infringing output. The court rejected both arguments and held the start-up liable on all counts.
In a brief statement after the verdict, OpenAI said: “We disagree with the ruling and are considering next steps. The decision is for a limited set of lyrics and does not impact the millions of people, businesses and developers in Germany that use our technology every day.” The company added that it is “having productive conversations with many organisations around the world” on licensing.
GEMA described the judgment as the first major European ruling on the use of copyrighted material to train generative-AI systems. Law firm Raue, which acted for GEMA, said the decision “sets an important precedent for the protection of creative works and sends a clear signal to the global tech industry”. Kai Welp, head of GEMA’s legal department, said authors “must receive remuneration for the commercial exploitation of their works so that they can make a living”.
The German Journalists’ Association welcomed the ruling as “a milestone victory for copyright law”. The verdict could influence separate lawsuits filed in the United States by authors and media companies who allege that OpenAI trained ChatGPT on their work without permission.