Who Owns THIS? AI, Copyright and the Fight for Music’s Future
PanelAI has stopped feeling like a future problem. It is already inside the music business — in songwriting tools, production software, recommendation systems, marketing workflows, catalogues and increasingly in the music itself.
The legal fights around GEMA and generative AI are only the most visible part of a much bigger collision. Beneath them is a basic question the industry still hasn't answered: if a machine learns from music, who gets to decide what happens next?
For artists, the anxiety is immediate. Your voice, your writing, your catalogue and even the shape of your identity can become training material, imitation or competition. What does meaningful consent look like when the technology moves faster than contracts? What happens when an AI track can sound convincingly human, convincingly familiar — or convincingly like you?
For labels, the questions are just as complicated. How do you protect repertoire while still using new technology? What does licensing look like when training happens at enormous scale? Who carries responsibility when metadata disappears, attribution becomes unclear, or synthetic music enters the same discovery systems as human-made recordings?
And then there is the commercial reality. If AI-generated music starts competing directly for streams, playlist positions, sync placements and attention, the fight is no longer only about copyright. It becomes a fight over value: what music is worth, who gets paid, and whether the next generation of artists is entering an ecosystem designed to reward creativity — or simply content at scale.
For Europe, that question becomes even sharper. Smaller languages, local catalogues and underrepresented markets already struggle for visibility. If the next layer of music discovery is built on systems trained without context, transparency or accountability, those inequalities may simply be automated.
