The boss of Spotify has said he would not completely ban content generated by artificial intelligence from the music streaming service.
Daniel Ek told the BBC he thought there were legitimate use cases for the technology in music, but that it should not be used to impersonate real artists without their consent.
He said there were three “buckets” of AI use in music: tools such as auto-tune, which he said was acceptable; software which impersonated artists, which was not; and a more controversial middle ground where AI-generated music was inspired by a specific artist but did not directly mimic them.
But he said the issue would likely be debated for “many, many years”.
“You can imagine someone uploading a song, claiming to be Madonna, even if they’re not. We’ve seen pretty much everything in the history of Spotify at this point with people trying to game our system,” he said.
“We have a very large team that is working on exactly these types of issues.”
Spotify does not allow content from its platform to be used to train AI models.
Last month, Irish singer Hozier said he would consider striking over the threat of AI to the music industry, and a number of other artists have spoken out about their concerns about the use of technology in creating music.
Regulators worldwide are stepping up their scrutiny of AI, given its explosion into general use worldwide and fears over its impact on jobs, industry, copyright, the education sector and privacy – among many other areas.
We've seen pretty much everything in the history of Spotify at this point with people trying to game our system
Daniel Ek, SpotifyThe UK will host a safety summit on the potential opportunities and threats posed by AI in November.
Read MoreSpotify makes AI voice clones of podcasters and uses them to speak other languages
Spotify CEO Daniel Ek denies existence of alleged artist streaming scheme
Streaming service Deezer signs Universal Music royalty deal to prioritise musicians over AI
