[media-credit name=”Spotify Newsroom” width=1920 align=”center”]
[/media-credit]
Spotify has made a sweeping move against spam content on its platform by removing over 75 million tracks deemed “spammy” over the past year. The decision comes amid rising concern about the misuse of AI tools to generate fraudulent or misleading audio content. Many of these tracks were ultra-short, duplicated, or clear imitations generated via AI—uploaded in bulk to exploit streaming royalties. The goal has often been to capitalise on Spotify’s revenue model: Any track longer than 30 seconds can generate payouts, meaning that these low-effort uploads siphon attention and earnings away from legitimate artists.
In response, Spotify is introducing new safeguards. First, a music spam filter is being rolled out to detect suspicious uploads and prevent them from appearing in recommendations—or sometimes from ever being uploaded. Second, the company is tightening its rules around vocal deepfakes and impersonations: artists must give explicit permission for their voice or likeness to be used; tracks that impersonate artists without consent are disallowed.
Third, Spotify is working with the standards organisation DDEX to create voluntary industry metadata standards for labeling AI-assisted content. While disclosures won’t yet be mandatory, the company says this step is about building trust rather than punishing responsible creators.
Spotify emphasises that despite all this spam and potential for AI misuse, current engagement with AI-generated content on the platform remains very low. The company says it is not seeing meaningful impacts on user listening behavior or on royalty distribution to human artists.
As AI tools for music become more accessible, Spotify’s latest policies reflect an attempt to balance openness to innovation with protection of artist integrity and listener trust. Whether they will be enough to stem growing concerns over deepfakes, fraudulent uploads, and AI impersonation remains to be seen.
