Wednesday, 16 September 2026 · Europe
EUR/USD 1.154 EUR/GBP 0.8574 EUR/CHF 0.9449 EUR/PLN 4.347 All rates →
Sign in · Join
EUROPES The European Report
European Edition Wednesday, 16 September 2026
LATEST At least five killed and more than a dozen injured in Russian attacks on Ukraine Le Flash →
Nightlife & Plans

Over 21 million copyrighted songs are circulating among AI developers, watchdog tool launched

Over 21 million copyrighted songs are circulating among AI developers, watchdog tool launched

The Atlantic has exposed the exact copyrighted tracks powering generative AI music platforms

The investigation , conducted by writer Alex Reisner, found that AI developers have access to four major music datasets containing millions of recordings. The work spans music by both chart-topping and independent artists, with Billie Eilish, Taylor Swift, Nirvana, and Bad Bunny among those named in The Atlantic’ s report.

Two of the four datasets contain more than 100,000 recordings each, while the other two are significantly larger, comprising between 9 million and 12 million tracks apiece.

According to The Atlantic , all four datasets have been downloaded thousands of times. The report notes that it is difficult to determine which companies have used the recordings, as details of AI training datasets are rarely made public. Google and Stability AI were, however, identified as having used the Free Music Archive dataset in AI training.

To accompany its investigation, The Atlantic has launched a tool designed to help artists identify whether their music is included in the four datasets. You can access the AI Watchdog tool here , and search by artist name to see if their work is included.

One use of the tool shows that 54 songs by Eric Prydz appear among the four datasets, while as many as 126 songs credited to Honey Dijon showed up. Other artists whose music appears include Björk (411 songs), Moby (213), Fatboy Slim (175), The Chemical Brothers (153), Daft Punk (151), Charlotte de Witte (89), and many more.

Following the tool’s launch, multiple musicians have expressed anger and concern over discovering their work had appeared in the sets. SZA was among them, sharing in an Instagram story, “Jus checked and music AI has trained off 238 of my songs. I’m certain some unreleased. If your a musician and you support this degenerate shit? Your disgusting and there’s NOTHING YOU COULD EVER SAY TO ME TO MAKE THIS OKAY.”

Producer Kenneth Blume, aka Kenny Beats , directly criticised AI music company Suno in a post responding to the report. “I can’t imagine going into work daily knowing you are stealing from countless struggling musicians,” he wrote. “I can’t imagine being proud to earn a paycheck obliterating the work and dreams of artists.”

On Bluesky, DJ Sabrina the Teenage DJ also shared her reaction after discovering 22 of her songs in the datasets. “To everyone who thought my music sounded like AI slop, did you ever think it was because Suno was using a dataset that contained 22 of my songs?”, she wrote. “It’s funny how there were no accusations of my music sounding like AI slop until these datasets started getting used to generate slop.”

Last year, Universal Music settled its own lawsuit against Udio , and a deal was struck for the two to combine on a new platform.

Meanwhile, Suno CEO Mikey Shulman was widely criticised for claiming that “it’s not really enjoyable to make music now” and “the majority of people don’t enjoy the majority of the time they spend making music.”

More from Nightlife & Plans