Spotify is taking a stand against low-quality content, removing over 75 million songs identified as Spotify AI slop or spam from its platform. This massive cleanup follows a one-year update on the streaming giant’s initiative to implement AI protections for artists, songwriters, and producers. The removed tracks were flagged for serious or repeated violations of platform rules, marking a significant shift in how the service handles generative content.
In addition to the outright removals, an additional 50 million songs have been filtered out of user recommendations following repeated complaints regarding quality. Spotify analyzed tracks that listeners were consistently skipping or ignoring, finding a clear pattern: these tracks were almost exclusively identified as slop or spam. By reducing the visibility of these tracks, the platform aims to refine the user experience and ensure that human-made music remains at the forefront of discovery.
Prioritizing music made by authentic, human artists has become a central pillar for the service over the last six months. Editors are now held to the same standard, ensuring they do not program content identified as spam. To achieve this, Spotify utilizes a combination of listener engagement, upload patterns, metadata analysis, and information from multiple third-party AI detection tools. The company acknowledges that the battle against low-quality content will only intensify as generative AI evolves, citing challenges such as mass uploads, SEO manipulation, and artificially short tracks.
Spotify competitor Tidal announced plans in 2024 to identify and tag music found to be almost completely AI-generated. While these labeled releases remain on the platform, they are ineligible for monetization. Spotify and Apple Music have previously implemented features to identify AI-generated content, though those efforts relied heavily on transparency from artists and distributors.
The platform also reports that 300,000 artists are now verified, featuring a badge that signals a “real, identifiable person” is behind the music. In contrast, music uploaded by what the service calls “AI Personas” is marked with a separate badge and excluded from editorial or algorithmic recommendations by default. Additionally, the “Artist Profile Protection” beta program is helping creators prevent impersonation and stream-farming, with plans to expand the feature by the end of the year. On the listener side, a new feature now lists AI contributions to a song in its credits, providing full transparency directly within the app.