An AI model like ChatGPT learns to speak by taking words directly from popular songs without paying for them. That is exactly what happened, and now there are consequences. The Munich Regional Court in Germany ruled that OpenAI, the company behind ChatGPT, infringed copyright by using protected song lyrics to train its language models. This means that the AI "learned" from artists' works without their consent, and the court ordered OpenAI to pay unspecified damages. The ruling follows a lawsuit filed by the German organization GEMA, which manages the rights of composers, lyricists, and music publishers.
What exactly happened?
The dispute began in November 2024, when GEMA filed a lawsuit against OpenAI. GEMA claims that ChatGPT used lyrics from nine famous German hits to improve its text-generation capabilities. These songs included Herbert Grönemeyer's 1984 synth-pop parody of masculinity titled "Männer" (Men), as well as Helene Fischer's song "Atemlos Durch die Nacht" (Breathless Through the Night), which became the unofficial anthem of the German soccer team during the 2014 World Cup. The court found that OpenAI had indeed used these lyrics for training without permission.
OpenAI defended itself by arguing that its models merely absorb vast amounts of data, do not store specific songs, and that the users who enter prompts are responsible for the outputs. However, the court rejected this argument and said that OpenAI is responsible for what its system produces. The court also emphasized that even short excerpts, such as just 15 words, constitute copyright infringement because they could not have appeared by chance. OpenAI also could not claim that it was "nonprofit" or that innovation justified such conduct—the court rejected all of these arguments.
Who is GEMA, and why is this a victory for it?
GEMA is an organization with approximately 100,000 members that protects the rights of music creators in Germany. Following the ruling, its CEO, Tobias Holzmüller, said that the internet is not a self-service store where people can take human creative works for free. According to him, the decision protects musicians' livelihoods and proves that even operators of AI tools such as ChatGPT must comply with copyright law. The law firm Raue, which represented GEMA, added that the ruling creates legal certainty for creators, publishers, and platforms across Europe and sends a signal to the global technology industry.
GEMA now hopes to hold talks with OpenAI about how rights holders could receive fair compensation. This case is considered the first major European ruling in the fight against AI "stealing" creative content and could also affect other fields, such as writing or art.
OpenAI's response
OpenAI, founded by Sam Altman, disagrees with the ruling and is considering an appeal. In a statement, the company said that the decision concerns only a limited set of song lyrics and will not affect the millions of people, businesses, and developers in Germany who use its technology every day. It added that it respects creators' rights and is holding discussions with many organizations around the world to ensure that everyone can benefit from the opportunities offered by AI.
The court also rejected the argument that the text and data mining exception applied to this case because GEMA had reserved the rights and the reproductions were substantial. OpenAI was found negligent because it could not claim to have been unaware of the risks. OpenAI is facing similar lawsuits in the US from authors and media groups that claim ChatGPT was trained on their works without consent.
The entire case serves as a warning to technology giants that European courts take copyright seriously.
Sources: reuters.com and theguardian.com
