Spotify has removed 75 million AI-generated music tracks from its platform in a single year, marking one of the largest content purges in streaming history. The move targets a surge in spam accounts that flooded the service with algorithmically produced songs designed to collect royalties without human oversight.
Behind the Purge
As Spotify's Sam Duboff, senior director and global head of marketing, policy, and music business, revealed in a recent interview, the platform identified and removed 75 million AI-generated tracks in 2025. Duboff noted that while the volume is staggering, the underlying spam tactics are not new. The same patterns used by bots to exploit ad revenue on other platforms have migrated to music streaming.
Spotify has invested in automated detection tools and human review teams to catch these tracks before they reach listeners. The company emphasizes that not all AI-generated music violates its policies, only content created solely to manipulate streaming counts and collect royalties without genuine artistic effort.
Why This Matters
The purge has direct consequences for independent musicians who rely on streaming income. When spam tracks inflate the total pool of streams, revenue per legitimate play shrinks, and recommendation algorithms can surface low-quality content over original work. For Spotify, the challenge is ongoing: as detection improves, spam operators refine their methods using more advanced AI.
Beyond Spotify, the music industry faces a reckoning. Record labels, distributors and rights organizations must develop standards for what constitutes acceptable AI use. The line between AI-assisted creativity and automated spam will shape the economics of streaming for years to come.
How Spotify Identifies Problem Tracks
The Broader AI Content Challenge
Spotify is not alone in grappling with AI-generated spam. Social media platforms, e-commerce sites and video streaming services all face similar floods of synthetic content designed to game algorithms. The difference on Spotify is the direct financial incentive: each play can generate fractions of a cent, meaning even low-volume fraud can yield significant returns at scale.
As AI generation tools become cheaper and more accessible, the cost of producing fake music drops toward zero. This forces streaming platforms to invest heavily in moderation infrastructure, a cost that may eventually flow back to users through higher subscription fees or reduced royalty payouts.
What Comes Next
Spotify has not disclosed the full cost of its anti-spam operation, but the scale of removals suggests a dedicated internal team. The company is likely to implement more stringent upload requirements, such as identity verification for artists and content fingerprinting for all new tracks. Industry observers expect regulators in Europe and the United States to examine the role of AI in music streaming abuse, potentially leading to new rules on synthetic content disclosure.



