The ease with which artificial intelligence can now generate music is astonishing: an estimated 60,000 fully AI tracks are uploaded to streaming services daily, representing over a third of all new music.
It’s a capability that feels almost magical, even for those of us who can’t carry a tune. In seconds, AI can compose entire songs—pop, soul, metal, anything you can imagine—lyrics included. The output is often indistinguishable from human-created music, even to trained ears.
But this technological leap comes at a cost. An investigation reveals a widespread attempt at fraud, with criminals leveraging AI tools to siphon billions of pounds from legitimate musicians.
The scheme unfolds in two stages, resembling a plot from science fiction, yet it’s a reality in today’s digital economy. First, fraudsters mass-produce AI music. Then, they deploy bots to stream these tracks repeatedly, generating royalties.
Essentially, it’s robots listening to robot music. The defense? More robots. The sheer volume of AI music being created is staggering. In 2015, the entire U.S. music industry produced around 57,000 songs. Now, Deezer, a streaming platform, anticipates receiving 21 million AI tracks annually—a conservative estimate, as production scales rapidly.
“It’s a way to totally flood the music streaming services,” says Romain Hennequin, head of research at Deezer, who has developed an algorithm to detect AI music by identifying subtle, inaudible features.
While the tracks themselves aren’t inherently fraudulent, the activity surrounding them is. Automated systems, or bots, are used to repeatedly stream AI-generated songs to accumulate royalties. According to Thibault Roucou, Deezer’s head of royalties, “the vast majority of the listeners of this content is in fact what we call stream manipulation or fraud.” His algorithms, mirroring those used by banks to detect unusual payment activity, suggest that as much as 85% of listens on fully AI music are illegitimate.
This impacts not only Deezer but also artists, due to the royalty distribution model. Streaming services don’t offer a fixed rate per stream; instead, artists are paid from a shared royalty pool based on their proportion of total streams. Consequently, artificially inflated stream counts diminish the pool, reducing earnings for all.
The financial stakes are substantial. “We detect 8 to 9% of the stream as being fraudulent,” says Alexis Lanternier, CEO of Deezer. “If you implement this 8% to the world of music, it’s roughly a few billion dollars, two to three.”
Deezer is actively combating this by identifying and blocking bots, but fraudsters continually adapt their methods. “It’s an ongoing battle,” says Roucou. “I think we will not lose, but we will not win anyway because they will continue to improve, and we will also. And we hope we will be able… to prevent them from taking too much money from other artists.”
Deezer’s Romain Hennequin (right) has developed an algorithm to detect AI tracks
Artists are understandably concerned. “As artists, we get such a small fraction of the money that we actually deserve to get because of the streaming system. And for that to just be getting cut shorter and shorter and shorter through robots… it makes my blood boil,” says folk musician Lila Tristram.
“The music industry needs to get their hands around this a little bit, otherwise it could rapidly get quite out of control,” says Aidan Grant, founder of music production agency Different Sauce.
Currently, the focus is on informing consumers and musicians. Deezer labels fully AI tracks as such, but remains the only major streaming service to do so. Spotify, the world’s largest streaming platform, has refrained from labeling, fearing it might stigmatize artists who utilize AI in their creative process.
Musician Lila Tristram says losing money to bots consuming AI tracks makes her ‘blood boil’
Spotify has joined Deezer in attempting to block the influx of AI music and fraudulent streams, removing 75 million spam tracks last year—many of which were AI-generated. Spotify’s total catalog stands at 100 million tracks.
Despite the hype, AI-generated music hasn’t yet captured a significant audience outside of occasional viral hits and, of course, the fraudsters. For now, music still thrives on genuine connection and promotion. The human element, it seems, still matters… for the moment, at least.
