Beatport will no longer accept tracks that are fully or majority AI-generated, the digital music store announced, while releases created with AI assistance will be tagged to increase transparency. The policy shift is supported by an expanded partnership with fraud detection specialist Beatdapp and updated systems to identify automated and generative music more efficiently.
Under the new rules, music produced entirely or mostly by AI will be rejected and rightsholders will be notified. Tracks where AI plays a supporting role will still be accepted, provided the finished work remains majority human-made. Those releases will be labelled during ingestion, giving Beatport’s curation team clearer context for reviewing and featuring content.
Community survey findings
The platform said the decision reflects feedback from its core audience of DJs. A recent survey found:
- 60% of respondents would not play AI-generated tracks.
- 77% cited a firm preference for supporting human creators.
- 8% expressed openness to playing AI-generated music.
- 13% would consider it if artists and labels were properly compensated.
“Beatport exists to serve DJs and empower artists and labels. This means partnering with our community and providing confidence in what our platform offers,” said CEO Matt Gralen. “While new production techniques have driven electronic music forward since the beginning, there is a difference between a tool that assists human creation and a system that replaces it entirely. Beatport is built on the former.”
Industry context
Beatdapp co-founder and co-CEO Andrew Batey added that the partnership will “bring greater transparency to generative AI while protecting human creativity.”
The move arrives amid broader scrutiny of AI in music. Last month, Paris-based streaming service Deezer confirmed that roughly 50% of daily uploads, around 90,000 tracks, are now AI-generated. In August, Spotify began labelling artist profiles that use an ‘AI persona’. A June investigation revealed that AI developers had used an estimated 21 million copyrighted tracks to train generative models.
