The Kaipher Editorial Team 
Spotify announced sweeping changes to its artificial intelligence music policies on Thursday, revealing it has removed more than 75 million problematic tracks in the past year and will implement industry-wide disclosure standards to distinguish legitimate AI use from fraudulent content.
New AI Transparency Standards and Spam Filters
Spotify will adopt the forthcoming DDEX (Digital Data Exchange) standard for AI music disclosure, enabling artists and labels to specify precisely how AI contributed—whether in vocals, instrumentation, or post-production—rather than relying on a binary “AI or not AI” label. Sam Duboff, Spotify’s Global Head of Marketing and Policy, explained that AI integration occurs along a spectrum and that granular disclosures will foster greater accuracy and trust in the platform’s catalog.
In addition, Spotify is rolling out an enhanced AI-driven spam filter this fall to combat mass uploads, duplicate tracks, SEO manipulation, and ultra-short songs engineered to inflate royalty payments. Instead of outright deletion, flagged tracks will be removed from algorithmic recommendations—such as personalized playlists and radio stations—while remaining accessible via direct search. This approach aims to curtail fraudulent schemes without penalizing legitimate creators who may experiment with AI-based workflows.
Crackdown in Response to AI-Generated Fraud
The policy overhaul follows high-profile controversies in which entirely AI-generated “artists” achieved viral success without transparent attribution. The Velvet Sundown, for example, garnered over a million monthly listeners before Reddit users uncovered inconsistencies in profile photos and band biographies, exposing the platform’s vulnerability to deepfake-driven deception.
Spotify executives also addressed rumors that the company itself promotes AI-generated music to reduce royalty payouts. Duboff categorically denied these claims, stressing that Spotify neither produces nor owns any music on its platform; all tracks are uploaded and licensed by third parties. He cited a recent case where a North Carolina musician allegedly used AI to generate hundreds of thousands of songs, amassing over $10 million in fraudulent royalties—an abuse Spotify’s new safeguards are explicitly designed to prevent.
By combining detailed AI attribution with robust spam detection, Spotify aims to strike a balance between supporting innovative uses of artificial intelligence in music and protecting artists, rights holders, and listeners from deceptive practices.
