AI-Generated Music Is Flooding Streaming Platforms — And Royalties Are Being Rewritten

Streaming services have spent the past two years managing an uncomfortable arithmetic problem. The total pool of money paid out to rights holders each month is roughly fixed, set by subscription revenue and advertising, while the number of tracks competing for a share of it has expanded sharply as AI generation tools have made it trivial to produce finished-sounding songs at almost no cost. Industry estimates of what proportion of uploads are now wholly or partly AI-generated vary widely and should be treated cautiously, but several major platforms have acknowledged the volume is large enough to affect how they calculate per-stream payouts.

From flat pools to tiered payouts

The traditional streaming royalty model divides a platform's royalty pool pro rata by share of total streams, meaning a lightly-streamed AI track and a lightly-streamed human artist's track are treated identically. Several platforms have begun moving away from this. Spotify, Deezer and others have introduced or trialled thresholds that exclude tracks below a minimum stream count from payouts altogether, partly to filter out low-effort AI uploads created purely to capture background-listening royalties through repetition and playlist gaming.

Deezer has gone further, introducing detection systems that flag suspected fully AI-generated tracks and routing their royalties separately from music understood to involve meaningful human authorship. The distinction between 'AI-assisted' and 'AI-generated' is doing a great deal of work in these policies, and platforms have been notably vague about how the line is actually drawn in borderline cases, such as AI-mastered vocals over human composition.

The platforms are trying to protect the value of scarcity in a system that AI has made abundant almost overnight.

Labels and collecting societies are split

Major labels have taken an ambivalent position. Several have licensing discussions under way with AI music companies, seeking payment for training their models on catalogue material, while simultaneously lobbying platforms to suppress unlicensed AI output that competes for listener attention. Collecting societies responsible for distributing royalties to songwriters and performers have generally been more cautious, warning that any detection-based filtering risks misclassifying legitimate independent artists who use AI tools for mixing, mastering or instrumental backing tracks without those tools substituting for the creative core of a song.

• Some platforms exclude tracks under a minimum stream threshold from royalty pools entirely, aimed partly at low-effort AI uploads.
• Detection tools attempting to flag fully AI-generated tracks remain imperfect and are not disclosed in detail by the platforms using them.
• Licensing negotiations between labels and AI music firms are proceeding separately from platform-level payout policy, creating two parallel disputes.
• Songwriter and performer collecting societies have pushed back on filtering approaches they see as too blunt.

What this means for working musicians

For independent and mid-tier artists, the immediate effect has been a further squeeze on already thin per-stream payouts, since any filtering that removes AI tracks from the pool denominator can raise the notional per-stream rate for surviving tracks, but only marginally, and only if enforcement is effective. Musicians' unions in several countries have called for mandatory labelling of AI-generated content on streaming platforms, similar to disclosure requirements now used in some advertising and political content contexts, arguing that listeners have a right to know what they are streaming even if the royalty question is settled separately.

There is no consensus yet on where this settles. What has changed is that platforms, labels and collecting societies have all accepted, within the last year, that the pre-AI royalty architecture cannot simply absorb the current volume of machine-generated output without materially devaluing every other track on the service. The negotiations now under way are effectively about how scarcity gets reconstructed inside a system built for it.