Spotify announced in October 2026 that it has removed more than 50 million AI-generated spam songs from its user recommendation system so far this year, and completely deleted more than 75 million seriously violating spam tracks.
Facing the impact of massive AI spam on the streaming ecosystem, Spotify has deeply adjusted its recommendation system over the past six months, combining multi-dimensional signals such as listener feedback, upload patterns, metadata, and third-party AI detection tools, focusing on combating new types of violations such as bulk uploading, SEO manipulation, and artificially shortened track lengths.
This crackdown comes as the music industry grapples with the rise of AI music generation tools such as Suno and Udio, with giants like Universal Music and Sony Music engaged in litigation with relevant startups over copyright issues. To curb fraud using bots to inflate play counts at the source, Spotify has previously set an annual play count threshold and launched an AI artist identification labeling feature, having verified more than 300,000 real human artists whose works account for more than 99% of the platform's search traffic.
Spotify's systematic governance actions have not only cleaned up the platform's recommendation ecosystem, but also provided an important reference for the streaming industry in dealing with the copyright and trust crisis brought by generative AI.