AI music is flooding streaming services, but who wants it?

This is it Step backa weekly newsletter covering one essential story from the world of technology. To learn more about how AI is changing music and the music industry, follow Terence O’Brien. Step back It arrives in our subscribers’ inboxes at 8 a.m. ET. Subscribe to Step back here.

Generative AI began being used in pop music almost as a gimmick. There was a feeling of experimentation in 2018 I am AI By Taryn South, 2019 Proto By Holly Herndon, albums created with a lot of help from artificial intelligence. Others got in on the action as well, exploring the outer limits of tools like Google’s Magenta and even training their own models. But things changed quickly with the launch of Suno in December 2023 and Udio in April 2024.

Both Suno and Udio allow users to quickly create entire music tracks with a simple text message. AI-generated music is no longer the preserve of fringe technical experts and experimenters, but is now accessible to anyone with an internet connection. This led to the flow of machine-made music to streaming platforms.

In September 2025, Deezer said that 28 percent of uploaded music was entirely generated by artificial intelligence. By the end of the year, that number had grown to more than 50,000 clips per day, representing 34 percent of uploads. Both users and artists have expressed their frustration, demanding that streaming platforms do something to combat the growing problem that is diluting playlists and taking millions in revenue away from legitimate artists. Udio did not respond to a request for comment.

Things have only gotten worse at Deezer, where daily uploads of AI-generated content have increased to 75,000, and threaten to overtake actual human-made music. Spotify removed more than 75 million unwanted audio tracks in just 12 months.

Deezer was the first major streaming platform to implement a system that detects and categorizes content generated by artificial intelligence. The service also prevents its algorithm from recommending it and has demonetized 85 percent of streams. In a recent press release, Deezer CEO Alexis Lanternier said, “AI-generated music is now far from a fringe phenomenon and as daily deliveries continue to increase, we hope the entire music ecosystem will join us in taking action to help protect artist rights and promote transparency for fans.”

Qobuz was the next to implement a detection system. It also published an AI charter, promising that it would never use AI in its editorial or regulatory content. While the company stopped short of banning AI-generated content, it edged in discontent, saying: “Coboz’s heart is and will continue to be human.”

Apple soon followed. Although his rating system has an obvious flaw, it is based on self-reporting. Apple Music requires labels and creators to voluntarily add transparency tags to their metadata. When asked how the requirements would be enforced, or what penalties, if any, there were for failing to classify AI-generated content, Apple declined to comment and referred me to an industry newsletter issued in early March that it says is “up to content providers to determine what qualifies as AI content.”

Spotify has also opted for a voluntary system. It recently launched AI Certifications, which identifies pathways created using generative AI. It is working with the DDEX standards group to create an industry standard for AI content classification. It goes beyond just blanket tagging, allowing artists to identify whether AI has been used to create lyrics, vocals or backing music. Initial glimpses of this effort began emerging in mid-April, with DistroKid as its first partner.

While DDEX counts most of the industry heavyweights as members – including Amazon, Google, Meta, Apple, Songtradr (home of Bandcamp), Pandora, BMI, UMG, Sony Music Entertainment, and Warner Music Group – not everyone is necessarily on board with the Spotify standard just yet.

Spotify has been criticized for its handling of artificial intelligence and so-called ghost artists. But recently, it has been at pains to talk about its transparency efforts and its growing offensive against spam and impersonation. The company also recently launched a Verified by Spotify badge that is supposed to ensure there’s a human behind an artist’s profile. Sam Duboff, global head of marketing and policy at Spotify for Artists, said: Edge It is experimenting with third-party detection tools, but still makes “a significant amount of incorrect assessments.”

Google also requires classification of AI-generated content, whether it’s on YouTube or YouTube Music. Although the company wouldn’t publicly detail how its systems combat AI spam, it said it “relies on… well-established systems that have been very successful at combating spam and click fraud, and reducing the spread of repetitive, low-quality content.” It also says that failure to disclose information could result in penalties, including removal of content or suspension from the YouTube Partner Program.

In poll after poll, public opinion toward AI music has been largely unfavorable. A study conducted by Deezer and Ipsos showed that 51% of respondents believe that AI will “lead to the creation of more low-quality, generic music.”

An opinion poll was conducted Hollywood Reporter The Frost School of Music found that 66% of people never listen to AI-generated music. 52% said they would not even want to listen to music by their favorite artist if they knew it was made with the help of artificial intelligence.

Researchers from Singapore also found a significant negative bias towards AI-generated content. The authors of the research claim that this is because emotion plays a central role in how we interact with music. “Because of its lack of expressive intent, AI-generated music may be viewed as less able to convey true emotions or foster meaningful connections with listeners,” they say.

However, only Bandcamp has completely banned AI-produced music. Of course, the agreement states that “music and sound generated in whole or in part by artificial intelligence is not permitted,” but enforcing this policy is easier said than done. Bandcamp does not proactively scan uploads to pick up AI music. Instead, it relies on manual reports from users to report suspicious content.

The flood of AI music shows no signs of abating. The number of AI tracks uploaded has steadily increased over the past year, and according to Deezer’s research director, Manuel Mosallam, “deliveries are likely to continue to increase.”

If there’s a silver lining, it’s that while the number of generative AI uploads has increased by about 40 percent, there doesn’t seem to be a noticeable increase in streams. “Consumption after defrauding doesn’t gain much traction and is still highly concentrated on a few viral paths,” Musallam says.

AI-generated music represents at least 1% of streams on Deezer as of April, compared to about 0.5% in early November. But at that time it was a percentage Fraudulent AI music streams have increased dramatically from “up to 70 percent” to 85 percent. This suggests that people are searching for AI music less, and the novelty may have worn off.

said Jack Malone, YouTube’s policy communications manager Edge The company is “engaged in the active development of new industry standards for AI disclosures in music credits,” though it stopped short of saying it is collaborating with Apple or Spotify specifically. Google has been heavily involved in creating C2PA for content authentication, but it has been criticized for inconsistent implementation, potential for abuse, and creating a false sense of security.

It doesn’t appear that Google or Spotify are ready to start eliminating or excluding AI-generated music from their recommendations engine. “Over time, we believe that the use of AI in music will increasingly be a spectrum, rather than a binary. Tracks will no longer be ‘categorically AI’ or ‘not at all AI’ with no break in between,” says Duboff.

Creations like Velvet Sundown, Breaking Rust, and Solomon Ray may be anomalies at the end of the day. They generated more interest in being artificial intelligence than in the quality of the music. Fully AI generated music will still pose a threat to working musicians, session artists, library music composers and the like. But they may find it difficult to find a foothold on the charts.

However, artists are embracing AI more frequently, even if it is largely behind the scenes. It’s made its way into songwriting sessions in Nashville and replaced samples for hip-hop producers, and Diplo says creatives need to adapt. (Or “Just like giving up and becoming an Uber driver until everyone gets Waymo.”) “We’re always hearing from top artists, songwriters, and producers who are incorporating AI technology into their creative processes,” says Duboff.

Companies are reluctant to sanction the use of AI in part because they expect it to become a standard tool in the industry. Even at the launch of Verified by Spotify, the company left the door open to AI work, saying: “The concept of artist authenticity is complex and evolving rapidly.”

But when Suno users crank out an entire AI clip on Spotify every two weeks, demand for dramatic steps is likely to grow. The Deezer/Ipsos study found that 45 percent of people would like to filter all AI-generated music from their streaming music library. It’s a solution that neither Deezer nor any other streaming service has committed to. It will face steep hurdles of its own, including establishing an industry-wide standard for labeling that will be implemented consistently, and robust and reliable AI-powered detection tools.

If someone wants to listen to Zannia Monet, no one should stand in their way. If you could flip a switch and instantly hide all the AI ​​music on Spotify, I bet a lot of people would do it.