Spotify is tightening its approach to low-quality and spam content while exploring ways for artists to earn money from AI-powered remixes and covers, reflecting the growing challenge of managing artificial intelligence on music-streaming platforms.
In an update published on October 20, the streaming company said it had removed more than 50 million tracks identified as low-quality AI-generated content or spam from its recommendations this year. The tracks have not all been deleted from Spotify, the company says they are no longer surfaced through its recommendation systems.
The figure is separate from more than 75 million spam tracks Spotify says it removed entirely from the platform for serious or repeated violations of its rules. The company outlined the figures in its “official update on protecting human artistry”
Why Spotify is targeting AI-generated content
Generative AI has made it easier to create and upload large volumes of music, raising concerns about impersonation, duplicate uploads, misleading metadata, and attempts to manipulate streaming recommendations.
Spotify says it uses several signals to identify likely spam and low-quality content, including listener engagement, upload patterns, metadata, and information from third-party AI detection tools. It has also introduced measures to help artists protect their profiles from releases falsely associated with their names.
However, identifying AI-generated music is not straightforward. Spotify acknowledges that detection technology has limitations, and not every track created with AI is necessarily spam or low quality.
The distinction matters: the company’s stated policy is directed at spam, deception, and content it considers unsuitable for recommendations, rather than a blanket removal of all music involving artificial intelligence.
A new route for licensed AI remixes
At the same time, Spotify is developing a system that would allow fans to create AI-powered remixes and covers of existing songs under a licensing arrangement.
The company has signed an agreement with music distributor Believe, including its TuneCore service, to let participating artists and rights holders opt into the planned tool. According to “Spotify’s announcement of the partnership“, streams of eligible remixes would generate royalties for participating artists and the rights holders behind the original recordings.
Participation is intended to be voluntary. Artists would first opt in to consideration for AI-related programmes and then decide whether to take part in the partnership.
The proposed tool is designed to create a licensed environment for fan remixes and covers, an activity that already takes place across social media and other online platforms, sometimes without permission or compensation for the original creators.
The feature is still upcoming, however, and should not be described as a service that is already available to all Spotify users.
Can AI innovation protect artists’ livelihoods?
The competing moves illustrate the tension facing music platforms. AI tools could give fans new ways to engage with music and create additional revenue streams, but they can also make it easier to copy an artist’s voice, reproduce their work, or flood platforms with content designed to attract streams.
For artists, the central questions are whether they can meaningfully consent to the use of their work, receive fair compensation, and retain control over how their music is represented. For streaming services, the challenge is to develop systems that protect those rights without automatically excluding legitimate experimentation.
Spotify’s approach combines tighter recommendation controls with a proposed licensing framework. Its effectiveness will depend on how clearly the rules are applied, how reliably AI-related content is identified, and whether artists consider the new arrangements fair.
What this means for the music industry
The developments could influence how other streaming services respond to AI-generated music. As platforms introduce more disclosure and rights-management measures, record labels, independent distributors, and artists will need to decide how they want their work used in AI-powered products.
The wider issue is no longer simply whether AI belongs in music. It is whether the industry can establish practical rules for permission, attribution, and payment as the technology becomes more widespread.
