TL;DR
Spotify says it removed more than 50 million tracks it identified as AI-generated slop or spam from its recommendations this year, following updates to its recommendation system. The figure is in addition to 75 million spammy tracks the company says it removed in recent years; Spotify has not published a detailed breakdown of the removals or the detection criteria.
Spotify says it removed more than 50 million tracks it classified as AI-generated “slop and spam” from its recommendations this year, after listeners complained about the material appearing in their feeds. The company says the removals followed changes to its recommendation system and are separate from 75 million spammy tracks it had removed in recent years.
Music Business Worldwide reported Tuesday on Spotify’s disclosure, which followed the service’s announcement of AI-related protections a year ago. Spotify says it studied tracks that listeners skipped, ignored or flagged as unwanted earlier this year. It said the tracks repeatedly identified as slop or spam appeared to share that pattern of negative listener response.
“Over the past six months, we’ve made numerous updates to our recommendation system to remove slop and spam, and surface more music made by authentic, human artists,” Spotify said in a post on its website. The company reported a “marked decrease” in listener concerns about this content appearing in recommendations after the changes. It did not provide figures for the decrease or specify the period and measurement used.
Spotify says it does not rely on a single test to identify suspect tracks. Its stated approach combines listener engagement and feedback, upload patterns, metadata and information from multiple third-party AI detection tools. The company also says it looks for tactics including mass uploads, duplicate tracks, search-engine manipulation, spammy playlist networks and artificially short recordings.
How Recommendation Filters Affect Listeners
The scale Spotify reported signals that the company sees spam and AI-generated uploads as a recommendation-quality problem, not just a question of whether a track is technically available on the service. Removing tracks from recommendations can reduce how often listeners encounter them, even if the announcement does not establish that every removed track was deleted from Spotify’s catalog.
For listeners and artists, the distinction matters. Recommendation systems influence what people discover and what receives plays; filtering suspected spam could make it harder for mass-uploaded material to crowd out other music. At the same time, decisions based on automated signals can affect legitimate releases, and Spotify has not explained how artists can challenge a classification or how it checks for errors.
The disclosure also highlights a tension in Spotify’s approach to AI. The company says it is taking action against unwanted AI-generated music in recommendations, while it has also pursued AI uses elsewhere on its platform. In 2025, Spotify partnered with ElevenLabs on AI-narrated audiobooks, according to the source report. That does not show that the company’s position is uniformly against AI; the stated crackdown is aimed at content it identifies as spam or slop.
As an affiliate, we earn on qualifying purchases.
Spotify’s Earlier Spam Removals
The more than 50 million tracks described in this disclosure are additional to 75 million “spammy tracks” Spotify says it removed in recent years. The figures cover different periods and use different descriptions, so they should not be added together as a single total of AI-generated tracks. Spotify has not said how many of the newer removals were confirmed as AI-generated, as opposed to being classified more broadly as spam or slop.
The issue has extended beyond recommendation spam. The source report says critics have faulted Spotify for not doing enough to remove AI fakes of deceased artists and music impersonating real acts. King Gizzard & the Lizard Wizard left Spotify in 2025 in protest of CEO Daniel Ek’s investment in AI drone technology, according to the report. That dispute concerned the company’s investment, not the removal figures announced in this update.
“Earlier this year, we looked closely at the tracks listeners were skipping, ignoring, or telling us they didn’t want to hear.”
— Spotify, in a post on its website
As an affiliate, we earn on qualifying purchases.
What Spotify Has Not Disclosed
Spotify has not released a track-by-track breakdown showing how many of the more than 50 million removals were generated by AI, how many were classified as spam for other reasons, or how many were removed from recommendations rather than taken off the service entirely. The company’s description does not establish that every flagged track was fraudulent or that all AI-generated music was targeted.
The service also has not published the detection tools’ accuracy rates, the number of false positives, or details of an appeal process for affected artists and distributors. Spotify’s claim that listener concerns fell is not accompanied in the cited disclosure by a numerical baseline, time series or independent assessment. The extent to which the changes altered listeners’ exposure to unwanted tracks therefore remains unclear.
As an affiliate, we earn on qualifying purchases.
Spotify’s Next Filtering Updates
Spotify says it will continue updating its systems as spammers change tactics, including attempts to exploit bulk uploads, duplicates, metadata and playlist networks. The company has not announced a date for another report or provided a schedule for future recommendation changes.
Further disclosures on removal categories, measurement methods and safeguards for artists would help clarify how the policy is working. For now, the confirmed development is Spotify’s account of a large-scale reduction in tracks it classified as slop or spam within recommendations; the company’s public figures do not show the precise makeup or overall effect of those removals.
As an affiliate, we earn on qualifying purchases.
Key Questions
How many tracks did Spotify say it removed?
Spotify said it removed more than 50 million tracks it classified as AI-generated slop or spam from recommendations this year. It said this was in addition to 75 million spammy tracks removed in recent years.
Were all of the removed tracks confirmed to be AI-generated?
No such breakdown was provided. Spotify described the tracks as AI-generated slop and spam, but did not say how many were verified as AI-generated versus classified as spam through other signals.
Does removal from recommendations mean the tracks were deleted?
The disclosure describes removing tracks from recommendations. It does not clarify whether all of them were also removed from Spotify’s catalog or were simply excluded from recommendation surfaces.
How does Spotify identify suspected spam?
Spotify says it combines listener engagement and feedback, upload patterns, metadata and information from third-party AI detection tools. It also looks for tactics such as mass uploads, duplicates and artificially short tracks.
What remains unknown about the results?
Spotify has not shared a detailed category breakdown, accuracy rates, a numerical measure of the reported decline in listener concerns or details of how artists can appeal a classification.
Source: rss
