Can you make money on Spotify with AI-generated music?

In September 2026, Argentine media entrepreneur Mario Pergolini launched Zquizo, an entertainment and technology company with a line of work in AI-generated music. The news revived a question that's been circulating for months in forums and social media: can you build a real business by uploading AI-generated music to Spotify? We investigated the real cases, the royalty math, and the legal risk almost nobody mentions when talking about this.

A real example: the AI song that hit #1 in the United States

In late 2025, "Walk My Walk," by "virtual artist" Breaking Rust, hit #1 on Billboard's Country Digital Song Sales chart — the first time in history an AI-generated song led a Billboard chart. The track first went viral on Instagram, with AI-generated videos of lone cowboys and open highways, then jumped to Spotify.

What matters here: the song was not penalized for being AI. It earned royalties completely normally. Spotify doesn't downrank or block music for being AI-generated — what it does penalize (with demonetization, exclusion from editorial playlists, or outright catalog removal) is failing to disclose AI use, cloning voices without permission, or mass-uploading content to artificially inflate streams.

How much money you can make with AI-generated music

The most cited case is Velvet Sundown: a supposed folk-rock band that released three albums in a matter of weeks, all made with Suno (vocals included), and reached 1.4 million monthly Spotify listeners at its peak, even landing on algorithmic playlists like Discover Weekly. All of this before Rolling Stone confirmed the project was 100% AI-generated. Despite that peak, the estimated total earnings were just USD 1,000 to 3,000 — the number that most changes expectations about this business.

But Velvet Sundown isn't the ceiling. A report analyzing the ten highest-earning AI artists on Spotify and YouTube found that, combined, they've made USD 6.1 million from 2.3 billion combined plays. Some concrete examples, with figures estimated from their real stream counts:

• Enlly Blue (a 1950s-style AI vocalist): the highest-earning AI artist on Spotify alone — around USD 380,800 from 95.2 million streams.
• mikeeysmind: their top 10 tracks add up to 514.7 million streams, with estimated earnings of USD 2.06 million on Spotify alone.
• The Devil Inside: a "band" with ~700,000 verified monthly Spotify listeners, whose biggest track (Bones In The River) has 1.6 million streams. Released 13 albums in 2025 — a release pace impossible for a human band, one of the typical tells for spotting AI projects.
• Breaking Rust (the country song that hit #1 on Billboard): now has close to 30 million streams on Spotify alone.
• Don Bnnr (producer): no public earnings figure, but their tracks sit on 1,852 playlists with a combined 46.7 million followers — proof you can also build reach operating as a track producer for playlists, not just as a "band."

Smaller individual cases also circulate, told by their own creators on blogs and social media (for example, producers claiming to have made between USD 1,000 and 8,500 generating lo-fi tracks or background music for YouTube over a few months). Those figures are self-reported and not independently verified like the ones above — take them as a scale reference, not a confirmed data point.

Simulation: how much Spotify pays based on stream count

Using Spotify's real rate (between USD 0.003 and USD 0.005 per stream, before the distributor's cut) and the real monetization floor (a minimum of 1,000 streams per track in 12 months to earn anything), here's the full scale:

• 1,000 streams: USD 3 to 5 — right at the minimum for the track to start earning anything. Below this number, that track generates USD 0.
• 10,000 streams: USD 30 to 50
• 100,000 streams: USD 300 to 500
• 1,000,000 streams (1M): USD 3,000 to 5,000
• 10,000,000 streams (10M): USD 30,000 to 50,000
• 30,000,000 streams (30M) — Breaking Rust's real level today: USD 90,000 to 150,000
• 95,000,000 streams (95M) — Enlly Blue's real level: USD 285,000 to 475,000 (lines up with the real estimated figure of USD 380,800 reported for that artist)
• 100,000,000 streams (100M): USD 300,000 to 500,000

The simulation lines up reasonably well with the real cases above, which confirms that the USD 0.003-0.005 per-stream rate is a reasonable assumption to project with. The practical takeaway: this business needs real volume (tens of millions of streams, not thousands) to generate meaningful income — which explains why the top earners are the ones publishing huge catalogs consistently, not the ones uploading a single song hoping it goes viral.

The legal risk of uploading AI-generated music

In April 2026, a US federal court confirmed that tracks generated primarily by artificial intelligence cannot receive copyright protection, even if a human heavily curated and prompted them. The US Copyright Office had already signaled this before: simply typing a prompt and hitting "generate" doesn't count as sufficient human authorship. The result is that this music is born into the public domain.

In practice this means three things: you can't sell or exclusively license that catalog as intellectual property, anyone can take the same track and redistribute it, and the business doesn't build a valuable long-term catalog asset — it looks more like streaming arbitrage than building a record label. If the goal is for the catalog to have resale or future licensing value, a human should get substantially involved in each track (arrangement, editing, structure, manual mixing), not just generate and upload.

Which AI platforms to use, and which to avoid

For instrumental or electronic music without vocals — faster to produce at volume, which is exactly the variable that matters most in this model — these are the real options in 2026:

• Suno: USD 10 to 30 a month depending on the plan. The most generalist and highest-quality option, with full commercial rights on paid plans.
• Udio: up to USD 30 a month, with a more experimental sonic territory. Watch out: following a legal settlement with Universal Music, it disabled downloads, making it impractical for actually distributing music.
• Mubert: built for continuous ambient/electronic streams by "mood," with an API — the closest option to a continuous DJ set, though with less control over traditional song structure.
• Soundraw: fast and built for volume, with good coverage of electronic and pop genres, though with less creative depth than Suno or Udio.
• AIVA: the reference for orchestral and classical music — not the right choice for electronic.

How to upload AI-generated music to Spotify

Spotify doesn't accept direct uploads from artists: you need a digital distributor. Spotify's official monetization eligibility policy toward AI varies a lot from one distributor to the next — from those that accept it with no friction to those that ban it outright. Here's what to know about each before choosing:

• DistroKid — USD 24.99/year (Musician plan), unlimited uploads, 100% royalties: the cheapest option for anyone publishing at volume (exactly this business model). Delivers to stores in 24-72 hours. Pro: fast and cheap per track if you upload consistently. Con: runs its own "AI Provenance Check" ahead of the external filter — one of the strictest at verifying AI is properly disclosed.
• TuneCore — USD 24.99/year (Rising Artist plan), 100% streaming royalties: includes YouTube Content ID out of the box, but keeps 20% of social platform earnings. Pro: strong store coverage. Con: the most demanding distributor about AI disclosure — it asks you to specifically flag whether AI was involved only in the instrumental, only in the vocals, or both.
• RouteNote — free (with a cut) or paid plan with 100% royalties: the most permissive of the group with AI music, with less friction in review. Pro: lower barrier to entry. Con: being looser also means it's often among the first to tighten rules when a platform like Spotify tightens its own.
• UnitedMasters — free (keeps 10%) or Debut+ USD 19.99/year (100% royalties): the Select plan (USD 59.99/year) adds access to brand and sync deals. Pro: low entry price. Con: doesn't publish an AI policy with the same level of detail as DistroKid or TuneCore — worth confirming directly with support before uploading a large catalog.
• Ditto Music — from USD 19/year: accepts AI music with a disclosure requirement, and scans the catalog for typical AI-generation patterns during moderation. Pro: low entry price, clear AI policy. Con: active scanning means misdeclared metadata has a higher chance of triggering a rejection.
• Amuse — free (keeps 10%) or paid plan with more control: has human review in the upload process. Pro: good free entry point to test the waters. Con: allows AI but with specific store exclusions and no AI "covers" — human review can filter out more content than an automated scanner.
• Symphonic — Starter USD 19.99/year, 100% streaming royalties: however, splits 70/30 (they keep 30%) on user-generated content and social media revenue. Pro: solid mid-price option. Con: that UGC/social split can add up if your distribution strategy leans on TikTok or Reels.
• ONErpm — commission model, ~15% of royalties generated: no upfront cost, only takes a cut once the catalog actually generates streams. Pro: no fixed initial cost, useful for testing without risking capital. Con: at high stream volume, that 15% ends up costing more than a flat subscription like DistroKid's.
• AWAL — 15% commission (Core tier): no formally published AI policy, and human review under the Sony Music umbrella — in practice, not very accessible for AI-generated music catalogs.
• CD Baby — USD 9.99 per single or USD 14.99 per album (one-time fee), keeps 9% for life: bans AI-generated music entirely as of October 2025. Rule it out directly for this kind of project.

For a volume model (many tracks, consistent releases) like the one this business requires, DistroKid or Ditto Music are the most sensible starting points: low annual price, no upload limit, and an AI policy clear enough not to risk the whole catalog over a mislabeled metadata field.

So, is it worth it?

It's a real business, but not the one usually told in interviews. It's not "one viral song makes you rich" — it's a volume-and-patience business, where each track needs to clear the 1,000-stream floor before it starts adding up, and where the most successful documented case so far — 1.4 million listeners — generated just a few thousand dollars. On top of that there's a real legal gap: without substantial human involvement, that catalog isn't defensible as intellectual property. Anyone getting into this should go in with both of those numbers clear from day one, not after the fact.

The same underlying principle — that AI (and platforms) reward content with real presence and authorship, not content generated on autopilot — is exactly what we measure in our own guide to measuring your brand's AI presence and share of voice. If your business isn't music but a brand that wants to show up when ChatGPT, Perplexity, or Gemini answer questions about your industry, that's the place to start. And if music isn't your thing either, we ran the same exercise on another format: can you make money with AI-generated cartoons on YouTube?