Two numbers, two different actions
The coverage of this update has tended to merge two figures that describe different things, over different periods.
The 50 million measures visibility. In its 20 October update, Spotify says it has taken more than 50 million slop and spam tracks out of recommendations during 2026. Separately, it says it made numerous updates to its recommendation system over the past six months, so flagged tracks no longer surface in algorithmic picks and its editors do not programme them. The announcement does not establish that all 50 million exclusions happened within those six months. What the figure counts is exclusion from recommendations. It is not a count of tracks removed from Spotify's catalogue, and the post does not establish the current status of every track it covers.
The 75 million is about deletion, and it is not new. Spotify's post presents it as the total of spammy tracks removed from the platform altogether for serious or repeated rule violations, but gives it no date. Asked by Music Business Worldwide, a Spotify spokesperson confirmed that it is the figure first given in September 2025, covering the previous 12 months, and that the company is not sharing an updated total. Placing an old deletion count beside a new recommendation count makes the current action look larger and harder than the update actually claims.
There is a third number, and it is missing: the AI share. The 50 million bundles AI-generated material together with other spam. Spotify's own list of the tactics it hunts includes mass uploads, duplicates, SEO manipulation, spammy playlist networks and artificially short tracks. Some of that material is AI music. Some is ordinary spam that predates generative tools. Without a split, nobody can use this announcement to say how much AI music Spotify is dealing with.
What Spotify actually found
The most useful part of the update is not the headline count. Earlier this year, Spotify says, it studied the tracks listeners were skipping, ignoring or telling it they did not want to hear, and found those tracks were consistently ones its systems had already identified as slop or spam. That is a reasonable basis for a recommendation change: if listeners are already voting against this material, stop recommending it. Spotify reports a marked decrease in listener concerns about this content showing up in recommendations since the changes. It gives no figure for that decrease, and the observation is Spotify's own.
Spotify also concedes the limit of its own method. No single signal catches slop or spam on its own, so it combines listener engagement and feedback, upload patterns, metadata and information from multiple third-party AI detection tools, and it says AI detection technology has limitations, and that identifying AI-generated content will only get harder as generative AI evolves. Deezer, by comparison, said on 13 October that it receives nearly 90,000 fully AI-generated tracks a day and already keeps them out of algorithmic recommendations and editorial playlists. That is a daily inflow figure, a different kind of quantity from Spotify's cumulative exclusion count. But it underlines the point: a cumulative count of what has been pushed out of recommendations says little about whether the share listeners encounter next month will be lower.
The practical effect
For listeners, the intended effect needs no deletion to matter: fewer flagged tracks in algorithmic playlists and radio, and none in editorial programming, if the system works as Spotify describes. Whether it works that way is, for now, Spotify's own report.
For artists, the sharper tools are elsewhere in the same update. AI Personas, artist identities Spotify believes are AI-generated, are now labelled, and are excluded from editorial and algorithmic recommendations by default unless a listener follows them. AI Credits, launched in April, let artists disclose exactly which roles and instruments on a track were made with AI; Spotify says artists are submitting tens of thousands of them every day. Artist Profile Protection, in beta since March with thousands of artists opted in, lets artists review releases that list them as an artist before those releases reach their profile. Spotify says it will make the tool available to all artists by the end of the year: a promise, not a completed rollout.
None of that requires believing the 50 million is a count of AI songs. Spotify does not present it as one, and it should not be read as one elsewhere.
Sources
- [1] Spotify, "One Year In: Putting Human Artistry First on Spotify", Spotify Newsroom, 20 October 2026 — the primary announcement: more than 50 million slop and spam tracks taken out of recommendations this year, the separate 75 million platform removals, the six months of recommendation-system changes, the detection method and its stated limitations, the 1,000-stream royalty threshold and the 10x figure, AI Credits, AI Persona labels, Artist Profile Protection, and the marked decrease in listener concernsRead source
- [2] Music Business Worldwide, "Spotify has pulled more than 50m 'AI slop and spam' tracks out of its recommendations this year", 20–21 October 2026 — the Spotify spokesperson's confirmation that the 75 million platform-removal figure is from September 2025, covers the previous 12 months and has not been updated; also the source of the Deezer context (nearly 90,000 fully AI-generated tracks received per day, per Deezer's 13 October statement) and of the March beta date for Artist Profile ProtectionRead source