Spotify Removes 50 Million AI-Generated Tracks From Recommendations, Keeps 75 Million Deleted

Spotify announced on Tuesday that it has removed over 50 million songs it labels as AI-generated “slop and spam” from its recommendation engine. The tracks remain available to stream, but editors and algorithms will no longer surface them to listeners. The move comes a year after the company rolled out new AI-music rules that also saw the deletion of more than 75 million tracks for serious or repeated violations.

• Spotify’s recommendation filters now exclude AI-generated slop and spam, reducing listener complaints.
• The company keeps a separate deletion list of 75 million spam tracks that have been removed entirely from the catalog.
• Spotify is expanding artist verification and AI-persona labeling to help users identify human-made music.

What the Numbers Mean for Users and Artists

Spotify’s public data shows that the 50 million tracks taken out of recommendations were identified through a mix of user behaviour—tracks that listeners skipped, ignored or explicitly told the platform they did not want to hear—and internal signals such as metadata, upload patterns and third-party AI detection tools. The company notes that AI detection tools have limitations and will become harder to use as generative-AI models improve.

While the 50 million figure represents tracks removed from recommendation lists, Spotify also reports that it has deleted more than 75 million tracks outright for “serious or repeated violations of our rules.” The deletion number has remained unchanged since the September 2025 policy announcement, and the company has not provided a new update on the total.

Policy Evolution and Industry Context

Spotify’s policy overhaul began in late 2025, when the company introduced a verified badge for human artists—now held by over 300,000 acts—and began labeling thousands of AI-generated artists with an “AI Persona” badge. Tracks from these AI personas are excluded from editor-curated playlists and, by default, from recommendations unless a user follows the artist.

Other streaming services are taking different approaches. Deezer announced a user-controlled filter to block AI slop by year-end, while Bandcamp has largely banned AI music. TIDAL has gone further by withholding royalties for fully AI-generated tracks. The industry’s response reflects a growing concern that AI-generated music can siphon royalties from real artists, a problem Spotify says is mitigated by its 1,000-stream royalty threshold.

How Spotify’s Algorithms Work

Spotify combines several signals to flag potential slop and spam: listener engagement metrics, upload patterns, track metadata, and third-party AI detection tools. The company acknowledges that AI detection “has limitations” and that spammers will continually adapt to evade the rules. Consequently, Spotify says it will need to keep updating its systems to stay ahead of new tactics.

In addition to algorithmic filtering, Spotify’s editorial team removes tracks identified as slop or spam from curated playlists. The company also encourages artists to submit AI-use credits daily; these credits appear in the Now Playing view on mobile, allowing users to see whether an artist disclosed AI involvement.

What Comes Next for Spotify and Its Users

Spotify has not announced a toggle that lets users block all AI-generated content, citing a focus on “addressing harmful uses of AI like spam and impersonation” rather than filtering music by its production method. However, the platform is expanding a tool that lets musicians approve songs that appear on their profile, which will be available to all artists by year-end.

For website owners, developers, and IT teams that rely on Spotify’s API or embed Spotify players, the changes mean that the recommendation data returned by the API will no longer include the 50 million flagged tracks. This could affect analytics that track recommendation performance and user engagement. Developers may need to adjust filtering logic if they previously relied on Spotify’s recommendation feed to surface new or niche music.

For businesses and advertisers, the shift signals a cleaner catalog that may reduce the risk of associating brand content with low-quality or potentially copyrighted AI-generated tracks. The removal of spam tracks also helps maintain the integrity of playlist curation, which can be important for marketing campaigns that use Spotify’s ad platform.

In the long term, Spotify’s continued investment in AI-detection and artist verification tools suggests the company will keep refining its approach to AI-generated music. Users can expect further updates as the platform balances the creative possibilities of generative AI with the need to protect human artists and maintain a high-quality listening experience.