Spotify Removes 75 Million AI-Generated Spam Tracks to Protect Artist Royalties
Key Takeaways
- ▸Spotify purged 75 million spam AI tracks over 12 months to protect artist royalty pools, but legitimate AI music remains welcome
- ▸Enforcement targets fraud through voice cloning bans, machine learning spam detection, and DDEX metadata standards requiring disclosure of AI involvement
- ▸AI music is allowed on Spotify when creators hold commercial rights, show genuine creative intent, and comply with authorization standards
Summary
Over the past 12 months, Spotify removed more than 75 million AI-generated spam tracks from its catalog to protect legitimate artists' royalties. The platform clarified that it is not banning AI music entirely, but rather targeting fraudulent submissions characterized by mass uploads, artificial metadata, unauthorized voice cloning, and algorithmically-gamed short tracks designed to siphon royalties without genuine creative intent. Spotify's enforcement strategy includes voice cloning bans, machine learning spam filters, DDEX metadata standards for transparency, anti-scraping defenses, and a dedicated monitoring team. The company is actively working with labels to legitimize authorized AI-created content, while industry groups like IFPI and RIAA develop voluntary labeling standards to distinguish AI-generated from AI-assisted music.
- The platform's approach signals a distinction between creative AI use (authorized remixes, AI-assisted mixing) and catalogue-stuffing spam
