GEMA CEO Tobias Holzmüller, whose society is suing OpenAI in Germany for alleged training-related copyright infringement. Photo Credit: GEMA
It turns out GEMA’s push to rein in generative AI giants isn’t limited to a comprehensive royalties framework, as the German society has filed a copyright infringement lawsuit against OpenAI.
GEMA formally announced the high-stakes legal action today, a little over one week after unveiling its “AI Charter.” A collection of 10 “ethical and legal principles for dealing with generative” AI, said Charter followed the October rollout of an aggressive royalties-model proposal for artificial intelligence platforms.
Now, the Berlin-based entity has upped the ante once again, this time with a complaint submitted to the Munich Regional Court. Naming as defendants U.S.-based OpenAI LLC and OpenAI Ireland Ltd., the lawsuit maintains that ChatGPT was trained on protected song lyrics (a portion written by GEMA’s approximately 95,000 members) without authorization, per the filing party.
Also according to GEMA, which has launched a dedicated webpage summarizing the suit, ChatGPT produces some of the lyrics when responding to user prompts. As many know, rightsholders are alleging the same type of infringement in multiple stateside cases against leading AI businesses.
But with these domestic legal battles plodding along as the underlying technology continues to advance by leaps and bounds, GEMA is taking the opportunity to tout itself as “the first collecting society worldwide to file a lawsuit against a provider of generative” AI systems.
And while it perhaps goes without saying given the size of the defendant – reportedly valued at an eye-watering $157 billion – GEMA isn’t hesitating to emphasize the case’s significance for the broader AI sector in Europe.
“GEMA will also consider whether to initiate legal action against other AI providers in the future,” the previously mentioned webpage spells out.
“The lawsuit is a test case to clarify numerous legal issues,” the text proceeds. “The aim is to specifically refute the AI system providers’ contention that training with and subsequent use of the generated content is free of charge and possible without the rights holders’ authorisation. … In addition, the lawsuit’s filing and the resulting media attention are intended to initiate a public discussion on copyright and AI.”
Running with the public-discussion point, GEMA has further teed up “a digital background discussion” on generative AI’s use of protected media, featuring CEO Tobias Holzmüller and additional “experts,” for November 19th.
In other words, the case and the corresponding conversations could have a material impact on the AI landscape in Europe, where the AI Act is already going into effect. DMN reached out to OpenAI for comment but didn’t receive a response in time for publishing.