Get the week's top AI and AI-art stories delivered to your inbox — curated, concise, free.
Free. Unsubscribe any time.
Deezer has crossed a stark milestone: more than 50% of all tracks uploaded to the streaming platform each day are now AI-generated, with over 90,000 AI tracks arriving daily as of June 2025.
Deezer has seen AI-generated uploads surpass 50% of all daily new music as of June 2025.
Image: TechCrunch / TechCrunch AI
The scale is striking. Deezer's catalog already held tens of millions of tracks before generative audio tools became mainstream — and now a single month's AI output at this rate would add roughly 2.7 million new tracks. For context, Spotify's entire catalog sat around 100 million songs as recently as 2023. The math suggests AI is compressing what once took decades of human recording into months of automated generation.
The practical consequence for anyone earning from music — or competing for listener attention — is royalty dilution. Streaming platforms typically distribute a fixed pool of revenue across all streams. When AI tracks flood a catalog and accumulate even modest play counts, they draw from the same pot as human-made music. Deezer has publicly pushed for a tiered or artist-centric royalty model that weights streams differently based on whether a track is AI-generated or human-performed, though no such system is yet in place industry-wide.
That policy gap is exactly why the 50% threshold matters beyond the headline number. It is not just a curiosity about volume — it is evidence that the current upload-and-stream infrastructure was not designed for this ratio. Platforms built around human creative output are now majority pipelines for machine output, and the monetization rules have not caught up.
According to TechCrunch, Deezer has been among the more vocal streaming services on the need for structural change, but the company has not announced a hard cap on AI uploads or a separate AI-only catalog tier.
The acceleration maps closely onto the mainstream availability of tools like Suno and Udio, which let users generate complete, production-ready songs from a text prompt in seconds. The legal exposure around those tools is real — Sony Music Entertainment has filed suit against Udio over alleged copyright infringement across more than 30,000 songs, as covered in our earlier report on the Sony-Udio lawsuit — but the litigation has not slowed upload volume on platforms.
For AI-art creators who also work in audio, or who build AI characters and companions with voice and music components, the Deezer data signals that the same content-flood dynamic hitting image platforms is already further along in music. Discoverability is eroding fast. A track uploaded today competes not just with the human-made catalog but with tens of thousands of AI outputs released the same day.
Deezer's disclosure puts pressure on Spotify, Apple Music, and YouTube Music to publish comparable figures — or to explain why they aren't tracking them. YouTube has already moved to restrict monetization for low-effort AI-generated video content, signaling that at least one major platform is willing to act structurally rather than just rhetorically.
For creators building AI-generated music as part of a broader creative practice — whether scoring AI video, adding ambient sound to AI art, or releasing tracks under their own name — the window where AI music is a novelty advantage is closing. Volume is no longer a differentiator. Curation, style specificity, and human editorial judgment on top of AI generation are becoming the actual competitive edge.
