Europe may soon become the first territory to craft and pass new legislation on the use of artificial intelligence technologies that’s informed by the recent burst of developments around generative AI.
The European Union’s AI Act took its latest legislative step yesterday with a positive vote from a European Parliament committee for a bill whose overriding aims are “to ensure that AI systems are overseen by people, are safe, transparent, traceable, non-discriminatory, and environmentally friendly”.
This is about the whole gamut of applications for AI, most of which have little to do with music. For example, the act would ban intrusive uses of AI systems like predicting policing; real-time biometric identification; and emotion recognition systems in certain contexts.
However, in our ballpark, the new rules would include transparency requirements for generative AI models: “disclosing that the content was generated by AI, designing the model to prevent it from generating illegal content and publishing summaries of copyrighted data used for training”.
Music rightsholders and their representative bodies will approve. In WMG’s earnings call earlier this week, CEO Robert Kyncl pointed to the European legislation as a positive development that his company would support. European authors societies’ body GESAC was first out of the blocks yesterday to welcome the legislation in its current draft, meanwhile.
“Transparency obligation on foundation models and generative AI systems is key both for the public to understand how a content is generated by AI and for creators to see if their protected works are used without authorisation,” said general manager Véronique Desbrosses.
However, as she noted, this legislation isn’t finished: there are more votes, more amendments and almost certainly a lot more lobbying still to come before the act becomes law. Its progress will be followed closely by the music industry and by policymakers around the world alike.