Spotify strips 50 million AI slop tracks from user suggestions

Spotify has removed more than 50 million songs classified as AI-generated slop or spam from its recommendation systems over the past year, according to an announcement on Tuesday. This action aims to ensure users see more music created by human artists rather than automated content.

The company clarified that these tracks remain on the platform but are no longer suggested to listeners by algorithms or editors. Spotify stated that complaints regarding this content appearing in recommendations have decreased significantly since the changes were implemented.

What we know

Spotify identified these tracks by analysing listener engagement, upload patterns, metadata, and third-party AI detection tools. The company noted that AI detection has limitations and will become more difficult as technology improves.

In addition to the 50 million tracks removed from suggestions, Spotify deleted more than 75 million songs outright last year for serious rule-breaking. The company did not specify how many of the 50 million tracks were purely AI-generated versus other forms of spam.

Spotify has introduced an "AI Persona" badge for artists using an AI identity that does not represent a real person. Music from these AI personas is excluded from recommendations by default, though the service has not mandated labels for all AI tracks.

Other platforms are adopting stricter measures. YouTube Music labels AI songs, Deezer allows users to filter out AI slop, and Bandcamp has banned AI music entirely. TIDAL has also stopped paying royalties for fully AI-generated tracks.

Spotify reported that tracks likely to be AI-generated are ten times more likely to earn fewer than 1,000 streams annually, which is the minimum threshold for royalty payments on the service.

Defender's checklist

• Monitor user feedback channels for complaints about irrelevant or low-quality content.
• Review algorithmic transparency reports to understand how content is categorised.
• Verify if third-party detection tools are being used for content moderation.
• Check if specific badges or labels are applied to synthetic media.

The background in plain language

Streaming services rely on recommendation algorithms to suggest content to users. These systems analyse listening habits to predict what people want to hear. When low-quality or automated content floods the platform, it can degrade the user experience. Companies must balance allowing diverse content with maintaining quality standards. Teams usually combine technical detection with human review to manage this. Users often skip or reject content that does not meet their expectations, providing valuable data for these systems to learn from and improve over time.

What to do and how to stay safe: Spotify

• Review your streaming service settings to see if you can filter specific content types.
• Use feedback buttons to skip or reject music that appears to be low quality.
• Check for verified badges on artist profiles to identify human creators.
• Monitor official announcements for updates on content moderation policies.

Step-by-step guide: AI Governance for Small Teams: Practical Controls and Cost

General security guidance from the TechWire Beat newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

Does Spotify delete the AI slop tracks completely?

No, the tracks remain on the service but are removed from algorithmic recommendations and editor suggestions.

How does Spotify identify AI-generated music?

It uses a combination of listener feedback, upload patterns, metadata, and third-party AI detection tools.

Are other platforms taking similar actions?

Yes, YouTube Music labels AI songs, Deezer offers filters, Bandcamp bans AI music, and TIDAL cuts off royalties for AI tracks.