Spotify Won't Label AI Music, So Volunteers Are Doing It

Spotify said last year it would flag AI generated tracks. It hasn't, at least not on the dozens of tracks 404 Media reviewed for its latest report. So a handful of independent developers are building the disclosure layer themselves.

The two projects doing the work are SoullessMusic, run by Graeme Fulton, and SlopTracker. Fulton told 404 Media he could not tell what was AI and what was not, and that he watched real artists complain that someone had 'popped up with their song and just copied it.' His site pitches itself as a database for 'AI artists hiding on Spotify. No bands, no studios, no soul, just machines and melody.' SlopTracker takes the opposite side of the same problem, letting anyone paste a Spotify link or upload a file and get back a confidence score from open-source detectors.

The money attached is what turns this from an aesthetic complaint into a business story. SlopTracker estimates a small set of tracked AI artists is generating an estimated $5.7 million a year, with the top one, 'mikeeysmind,' at $1.5 million on its own. The site frames the transfer as $0.1188 per second draining from real artists. For scale, rival streamer Deezer, which does label AI, reported that 44 percent of all new music uploaded to its platform is now AI generated.

The honest caveat is that these detectors are themselves AI, and their creators concede false positives. The reporting also does not give a comparable Spotify-side upload figure, or a timeline for the labeling Spotify promised, and Spotify did not respond to a request for comment. What the story does show is where the pressure is coming from now: not from regulators or the majors, but from volunteers with a scraper and a domain name, publishing lists Spotify would rather nobody keep.