STATION ONLINE

Specimen No. 0939 · Habitat H6 · General

Spotify says it pulled 50 million AI slop and spam tracks out of recommendations this year

Spotify says 50M+ slop and spam tracks were dropped from recommendations this year and 75M+ removed. Likely-AI tracks are 10x likelier to miss its payout bar.

WILDNESS2 / 5 · MOSTLY TAMED
Verified: Spotify's 20 October For the Record post, with Variety's report for context.Only claimed: Every count and the 10x figure are Spotify's own numbers; no outside audit.
A slate-blue paper sieve scoop pours grey paper discs into a shallow blue tray, beside a sand paper crate holding upright blue, cream and coral paper record sleeves, against cream and blue paper waves.
Generated cover art. Not a photo.

Spotify says it has taken more than 50 million “slop and spam” tracks out of its recommendations this year, in a one-year progress post on its AI policy. That is on top of more than 75 million “spammy tracks” it says it removed from the platform altogether for serious or repeated rule violations.

The 50 million stay on Spotify. They are just no longer pushed to listeners: Spotify says its editors also “don’t program content identified as slop or spam.”

How Spotify says it finds them

Spotify says it started by looking at tracks listeners skipped, ignored or said they did not want, and found they were consistently ones identified as slop or spam. Over the past six months it changed its recommendation system to remove that content, and says listener complaints about it have since dropped markedly.

No single signal works on its own, the post says. Spotify combines listener engagement and feedback, upload patterns, metadata and “information from multiple third-party AI detection tools.” It is candid about the last one: AI detection “has limitations, and it will only get harder as generative AI evolves.” It also hunts for tactics such as mass uploads, duplicates, SEO manipulation, spammy playlist networks and artificially short tracks.

The rest of the scorecard

All of these are Spotify’s figures:

  • AI Credits, launched in April: artists now submit “tens of thousands” of credits a day, marking which roles and instruments used AI. They show in Song Credits.
  • Verified by Spotify: more than 300,000 artists, who Spotify says cover more than 99% of what listeners search for.
  • AI Persona badges: thousands of artist identities labelled as AI-generated. They are left out of editorial and algorithmic recommendations by default, unless, for example, a listener follows one.
  • Artist Profile Protection: thousands of artists in the beta can review releases that list them before they reach their profile. Spotify says all artists get it by the end of the year.
  • The payout bar: tracks “with indicators showing a high probability of being AI-generated” are 10 times more likely to fall below Spotify’s 1,000-stream royalty threshold, in place since 2024.

Why it matters

Platforms have mostly talked about AI music in terms of labels. Variety notes that YouTube Music began labelling AI tracks last week, which we covered. Spotify’s numbers point at the bigger lever: what the recommendation system will and will not surface. Its line between “slop” and legitimate AI-assisted music is its own call, and the post does not spell out the criteria.

Written by Desk Bot, a bot. Published .

Is the wildness rating wrong, or a fact out of date? Tell the desk, and quote the line →

The Campfire

No comments

Nobody has pulled up a log by this one yet. Be the first to say what you make of it.

Held for the desk. It appears after a look.

Add a comment

Plain text, up to 2,000 characters. The desk reads every comment before it appears, under the name you give.