Tidal Cuts AI Music Royalties in Crackdown on Fake Songs

Tidal isn’t banning fully AI-generated music. It’s doing something sharper: cutting off Tidal AI music royalties for tracks it identifies as wholly machine-made.

The streaming platform said that as of Monday, June 29, it will not knowingly attribute royalties to music it determines is 100% AI-generated, according to PYMNTS. Beginning in mid-July, Tidal also plans to identify and tag music it classifies as fully AI-generated, while blocking or removing AI tracks used for fraud, deception, or impersonation.

That’s the tension at the center of the policy. AI tools are increasingly part of music production, but Tidal is drawing a hard commercial line between AI-assisted work and tracks produced entirely by generative systems. For a shorter news version of the same policy shift, see XOOMAR’s AI Music Loses Royalties under Tidal's New Pay Rules.

Why Tidal AI music royalties are being cut, not just labeled

The assumption was that streaming platforms would handle AI music mainly through disclosure. Put a badge on it. Let listeners decide. Tidal’s move goes further by changing who gets paid.

Under the new policy, the platform says it won’t knowingly attribute royalties to music it identifies as wholly AI-generated. That matters because labeling is informational. Demonetization is financial.

“We are committed to protecting and rewarding organic creativity to avoid compromising an artist's ability to connect with and build their fandom from Tidal subscribers,” Tony Gervino, executive vice president and editor-in-chief at Tidal, said.

Gervino said the policy was sparked by listener comments that they don’t want to be exposed to wholly AI-generated music, as well as a flood of completely AI-generated tracks that often impersonate existing artists for financial gain.

The policy does not say all AI use is disqualifying. Gervino also said Tidal is not trying to “bash technological advancement,” noting that technology can help artists understand fans, build recording and engineering tools, automate workflow, and build instruments.

The split is deliberate:

• Before: A fully AI-generated track could appear in a catalog and potentially participate in monetization if not otherwise blocked.
• After: If Tidal identifies the track as wholly AI-generated, it will not knowingly attribute royalties to it.
• Next step: Beginning in mid-July, Tidal will tag music it identifies as 100% AI-generated.

The policy targets machine-made tracks, not every AI-assisted song

Tidal’s AI policy is built around one threshold: whether a track is wholly AI-generated.

That distinction is critical. A producer using AI for workflow, engineering, or creative support is not the same as an anonymous uploader submitting machine-made tracks at scale. Tidal’s public policy, as described in the source material, focuses on the second category.

The company also says it will block or remove AI-generated music it determines is used for fraudulent purposes. Examples in the source include deceiving listeners or exploiting the music, name, or likeness of musicians.

A narrower version of this distinction matters for artists and labels because AI is not one thing. It can mean an AI mastering tool, a generative instrumental bed, an AI-cloned voice, or a fully synthetic song with no human performance. Tidal is starting with the clearest category: 100% AI-generated music.

For more context on the broader upload pressure around this issue, XOOMAR has also covered how TIDAL Cuts Off AI Music Royalties as Uploads Explode.

How could Tidal separate fully AI-generated songs from AI-assisted music?

Tidal has not disclosed a full technical recipe for detection. That’s a key gap.

The source material does say Tidal will identify and tag music it classifies as 100% AI-generated, and that it will expand tagging to “substantially AI-generated” music when AI detection methods become more reliable. Related reporting also says Tidal will use automated tools to identify and remove AI-generated music that attempts to impersonate an artist or group.

So the practical challenge is clear, even if the exact tooling is not. Tidal has to separate these cases:

Analysis: the hardest part won’t be the obvious spam. It will be the gray zone. If a track uses AI in composition but includes human vocals, human lyrics, or human instrumentation, Tidal will need a defensible way to decide whether it crosses the line from assisted creation into machine-generated output.

That matters because a wrong classification could hit royalties, reputation, and listener trust at the same time.

A 500-track example shows why the money rule matters

Consider a hypothetical uploader who submits 500 instrumental tracks generated by an AI music tool, gives them generic titles, and routes them through a distributor.

Under Tidal’s new policy, if the platform identifies those tracks as wholly AI-generated, it can tag them as AI-generated and stop assigning royalties to them. If Tidal determines they are tied to fraud, deception, or impersonation, it can also block or remove them.

Now compare that with a human artist who uses AI to sketch drum patterns, then writes lyrics, records vocals, plays guitar, and finishes the song. Based on the policy described, that case may not meet the same “wholly AI-generated” threshold.

That contrast explains why Tidal AI music royalties are the pressure point. A label alone tells listeners what they’re hearing. A no-royalties rule attacks the incentive to upload large volumes of machine-made tracks for payout.

The supplied industry context shows why platforms are paying attention. PYMNTS reported that Deezer invested in AI detection infrastructure and found that fully AI-generated tracks it received daily rose from 10,000 in early 2025 to over 60,000 in March. It also reported that 85% of the tracks in 2025 were fraudulent ones used to game royalty payouts.

That Deezer data does not prove why Tidal acted. Tidal’s stated reasons are listener concerns and a flood of AI-generated music that often impersonates existing artists for financial gain. But the Deezer numbers show the kind of scale streaming services are trying to manage.

Who benefits, and who now has more work?

Artists get the clearest benefit if the policy works as intended. Tidal is trying to keep royalties tied to original works made by people, while reducing incentives for fully synthetic catalog flooding and artist impersonation.

Labels and distributors get a new burden. Tidal expects AI-generated music to be properly identified before it reaches the platform. That puts more pressure on upstream checks, especially when a track sits between human-made and machine-made.

AI music companies get a mixed signal. Tidal is not rejecting AI tools outright. Gervino explicitly defended technology’s role in helping artists build, record, engineer, and manage their work. But the platform is rejecting the idea that fully machine-generated output should automatically earn royalties like human-made music.

Listeners get more visibility. Beginning in mid-July, the planned AI tags should make it easier to know when a track has been classified as 100% AI-generated.

The next break point is reliability. Tidal says it will update the policy as technology evolves, and will tag substantially AI-generated music when detection improves. Watch whether the platform explains appeal processes, distributor obligations, and how it handles borderline songs. Those details will decide whether this becomes a workable standard for Tidal AI music royalties, or just the first fight over where human authorship ends.

Impact Analysis

• Tidal is moving beyond disclosure by changing whether fully AI-generated tracks can earn money.
• The policy draws a financial line between human-led music creation and wholly machine-made output.
• It signals growing pressure on streaming platforms to address AI impersonation and royalty abuse.