Spotify removed over 75 million spam tracks in the past year and rolled out three new policies to protect artists from voice clones and content farms.
The platform announced these changes on September 25, 2025, after what executives called an “explosion of generative AI tools” flooding the music ecosystem with low-quality content. The new rules target voice impersonation, spam uploads, and require transparency labels for AI-generated music.
Mass Removal Shows Scale of AI Spam Problem
The 75 million removed tracks highlight how serious the AI spam issue has become for streaming platforms. Spotify revealed this figure shows the dramatic increase in artificial content since generative AI tools became widely available.
Rival platform Deezer now receives over 30,000 fully AI-generated tracks daily, up from 20,000 in April 2025. About 70% of streams for these AI tracks show fraudulent activity patterns.
Spotify’s total music payouts grew from $1 billion in 2014 to $10 billion in 2024. Sam Duboff, Spotify’s Global Head of Marketing and Policy, explained that “big payouts entice bad actors” who try to game the system with mass uploads and duplicate content.
Voice Clone Ban Protects Artist Identity
The first major change bans unauthorized AI voice clones completely. Spotify’s new impersonation policy states that vocal impersonation only works when the copied artist gives permission.
“Unauthorized use of AI to clone an artist’s voice exploits their identity, undermines their artistry, and threatens the fundamental integrity of their work,” Spotify explained in its announcement.
The platform also targets “content mismatches” where fraudsters upload music to other artists’ profiles. Spotify will work with distributors to stop these attacks before tracks go live.
Artists who want to license their voices to AI projects can still do so. But Spotify wants to ensure “the choice stays in their hands” rather than having their voices stolen without permission.
New Spam Filter Targets Mass Upload Schemes
Spotify’s second major change introduces a music spam filter rolling out this fall. The system will identify and tag accounts that engage in suspicious activities like mass uploads, creating excessive duplicates, and manipulating search results.
The filter specifically targets tracks just over 30 seconds long, which bad actors use to generate royalty payments. It also catches accounts that alter metadata to create fake duplicates and use SEO tricks to game recommendation systems.
“Our new spam filter will flag tracks and uploaders using these tactics and stop recommending them across Spotify programming,” Duboff said during a press briefing.
The platform plans to roll out the filter gradually to avoid penalizing legitimate artists. Spotify will add new detection signals as spam tactics evolve.
Industry Standard Labels Show AI Usage
The third change requires transparency about AI use in music creation. Spotify will support a new industry standard called DDEX that shows exactly how AI contributed to each track.
These labels will indicate whether AI helped with vocals, instruments, or post-production work. This gives artists credit for their creative choices while informing listeners about the technology involved.
“This industry standard will allow for more accurate, nuanced disclosures,” Duboff explained. “It won’t force tracks into a false binary where a song either has to be categorically AI or not AI at all.”
Fifteen labels and distributors have committed to adopting the new system, including Amuse, Believe, CD Baby, DistroKid, Downtown Artist & Label Services, and EMPIRE.
Platform Denies Creating Own AI Content
During the announcement, Spotify executives addressed rumors that the platform adds AI-generated tracks to editorial playlists to reduce royalty payments.
“Those rumors are categorically and absolutely false,” Duboff stated. “Spotify doesn’t generate any music. We don’t own any music. All the music on Spotify, 100% of it, is created, owned, uploaded by licensed third parties.”
The company emphasized it won’t penalize artists for legitimate AI use. “We’re not here to punish artists for using AI authentically and responsibly,” said Charlie Hellman, VP Global Head of Music Product.
Timeline for Implementation
The impersonation policy takes effect immediately. The spam filter began its gradual rollout and will expand over the coming weeks. The DDEX AI credits system needs more time for formal certification and supply chain updates before full implementation.
These changes is Spotify’s most comprehensive response yet to AI-generated content challenges. The platform plans to protect legitimate artists while still allowing creative AI use with proper disclosure.
Spotify’s approach focuses on transparency and artist control rather than blanket bans on AI music technology.
