Spotify has reportedly pulled more than 75 million AI-generated songs from its platform, according to Sam Duboff, the company’s executive overseeing global artists, marketing, and policy.
The flood of AI-generated music isn’t new. For the past couple of years, AI content has been creeping into every corner of the internet, and social media in particular has felt the weight of it. Music streaming platforms haven’t been spared either, with thousands of tracks, many of dubious quality, uploaded daily.
Spotify has always walked a fine line when it comes to AI music — encouraging AI-generated content on the one hand, while trying to maintain quality on the other. Duboff said this ongoing tension is largely what pushed the company to remove those 75 million tracks.
He also admitted the situation puts Spotify in an awkward position, pointing to what he called “blurred lines” around what actually counts as AI-generated music these days. Without clear-cut criteria, quality control becomes a messy, subjective exercise.
And it’s not just about weeding out obvious AI slop. Plenty of legitimate artists now use AI tools as part of their everyday studio workflow, a trend that’s only likely to grow as major tech companies keep pushing artists toward these tools.
To put the scale of the problem in perspective, Duboff noted that Spotify receives around 100,000 song uploads every single day. Industry estimates suggest that as much as 44% of music uploaded to streaming platforms is AI-generated, meaning tens of thousands of AI tracks could be landing on Spotify daily, even if that figure is only partly accurate.
Part of what’s driving this surge is how easy AI music creation has become. Tools like Suno let anyone type a prompt and get a full song back in seconds. Google’s Lyria model offers similar capabilities and is built into Gemini, available to basically anyone, whether they’re a professional musician or not. That ease of access is exactly why so much of what’s coming in raises quality concerns.
