The producer had been making music for eight months with tools that generated complete songs from text prompts, and the process had become routine: type a description, wait a minute, upload the result. The tracks had generated streams, and the streams had generated royalties, small but real. In late June, the producer learned that the royalties were about to stop, at least on one platform. TIDAL, the streaming service owned by Square, announced that music it determines to be wholly generated by artificial intelligence will no longer participate in royalty payments.
The policy, announced June 29 and effective July 15, makes TIDAL the first major streaming service to cut fully AI-generated music out of the revenue system entirely. Such tracks will remain available to listeners, but they will carry an AI label and will not earn royalties or participate in TIDAL’s direct-to-fan sales. The service also said it will remove AI-generated music that impersonates real artists or misleads listeners. The rules apply to independent uploads through TIDAL’s artist platform as well as to catalog content.
The move is the sharpest response yet to a problem that has been building for two years. The cost of making a song has fallen to nearly zero: prompt-to-song tools can produce finished tracks in minutes, and the volume of AI music flowing into streaming services has become a flood. Deezer, a rival platform, has said that fully 44 percent of the new tracks uploaded to its service each day are AI-generated, and the percentage is rising across the industry.
The economics explain the stakes. Streaming royalties are paid from a pool that all plays share, so an influx of AI tracks dilutes the earnings of human artists, even if the AI tracks are not fraudulently labeled. A smaller but darker corner of the problem is outright fraud: operators have used bot accounts and AI-generated songs to farm streaming payouts, and federal prosecutors in the United States brought the first criminal case over the practice this year, accusing one operator of pulling in around $8 million with fake streams and synthetic music.
TIDAL’s approach differs from its competitors’ in an important way. Spotify and Apple Music have focused on labeling AI content and filtering spam, keeping such tracks in their catalogs while managing how they are presented. Deezer has gone further, removing fully AI-generated tracks from its recommendations. TIDAL’s policy targets the money directly, denying royalties to AI music as a class, a choice that creates a financial disincentive for the mass production that has been clogging the pipelines.
The service frames the decision as a matter of fairness to human creators. Its stated policy is that streaming revenue should reward music that is created, written and performed by people, and that royalties should not flow to works with no human author. The company acknowledges that its definition, wholly AI-generated, leaves gray areas, and that tracks with any meaningful human input remain eligible, but it says the line is worth drawing even if it is imperfect.
The music industry is watching to see whether other platforms follow. The major labels have spent years negotiating over AI and have pushed streaming services to take a harder line, and rights organizations have called for clearer rules on what counts as a human-made work. TIDAL’s position is not binding on Spotify or Apple Music, but it raises the cost of inaction: if the most artist-friendly streaming service refuses to pay for AI music, the others face pressure to explain why they do.
The commercial calculation is more complicated than it looks. AI music is cheap to make and easy to upload, which makes it a source of catalog depth and engagement for platforms that tolerate it. Cutting it off removes that volume, and TIDAL, a smaller service, has less to lose than its larger rivals. The bet is that the service’s credibility with artists, its brand, is worth more than the stream counts.
The labels have reacted with a mix of support and caution. The major record companies have spent two years negotiating over AI, and they have pushed streaming services to be more transparent about synthetic content, but they also have their own AI ambitions, and their stance on TIDAL’s policy has been careful. Independent artists, the group TIDAL’s policy is designed to protect, have been the most vocal supporters, seeing the move as a rare example of a platform choosing artists’ interests over catalog growth.
The practical effects will take months to measure. TIDAL has said it will use automated tools to identify wholly AI-generated tracks, and the accuracy of those tools will determine whether the policy protects human artists or punishes them, if a human-made track is mislabeled. The company says its policy is a living document and will evolve, a hedge that suggests the line between human and machine creation will be argued over for years.
For the producers of AI music, the policy is a line in the sand. Some will move their uploads to platforms that still pay; others will add enough human input to qualify under TIDAL’s rules. The debate over where the line sits, and who draws it, is only beginning. TIDAL has drawn its own line, and the rest of the industry now has to decide where theirs runs.
