Spotify says it removed more than 50 million songs considered AI-generated slop or spam from user recommendations over the past year. Alongside those removals, the streaming service has changed how it recommends music, aiming to give more visibility to work made by human artists.
Reducing unwanted tracks in recommendations
The removals were described in an announcement on Tuesday. Spotify says its editors follow the same standard, and that they do not program content identified as slop or spam. The company also reports a marked decrease in listener concerns about this material appearing in recommendations since it made the changes.
The distinction matters because Spotify’s announcement concerns what listeners are recommended, rather than a blanket removal of all AI-made music from the service. The company’s approach combines screening for content it considers spam or slop with adjustments intended to make human artists more prominent in recommendations.
How Spotify assesses music
Spotify says it uses several kinds of information to determine whether a track may be AI-generated slop. These include listener engagement and feedback, upload patterns, metadata, and third-party AI detection tools.
The company describes these as inputs to its assessment. The announcement does not say that any single signal automatically determines whether a track is removed from recommendations. Instead, the stated approach draws on activity around a track, information included with uploads, and detection tools.
Spotify has not mandated labels on all AI tracks. It has introduced an “AI Persona” badge for artists who use an AI identity that does not represent a real person. Music from AI personas will not be recommended, according to the announcement.
Different services are taking different approaches
Other music platforms have adopted policies or features that address AI music in other ways. YouTube Music is putting labels on AI songs, while Deezer gives users an option to filter out AI slop. Bandcamp has banned AI music entirely.
Those differences show that the services are not responding with one shared rule. Spotify’s stated measures focus on recommendations, an AI Persona badge, and identifying tracks considered slop or spam. The other examples cited in the announcement involve labels, a user filtering option, or a platform-wide ban.
Spam enforcement and AI music tools
The announcement also places the latest removals alongside Spotify’s earlier action against spam. Last year, the company said it took down more than 75 million songs identified as spam. That figure describes a previous spam takedown, while the more than 50 million figure covers songs considered AI-generated slop or spam removed from user recommendations over the past year.
Spotify is also partnering with Universal Music Group and the independent music publisher Believe to launch a tool for creating AI music remixes. Taken together, these details show the company addressing AI in music through both recommendation policies and a planned remix tool. The announcement does not provide further details about how that tool will work.
For listeners, the practical change Spotify describes is that recommendations should feature more music made by human artists and less material the service identifies as slop or spam. Its approach relies on a combination of signals and does not amount to labeling every AI track or excluding all AI music from the platform.