Spotify Responds After Activist Group Claims 86M Track Scrape

The claim has put new scrutiny on large-scale scraping, copyright enforcement, and how streamed music could be repurposed beyond its original licensing terms.

Activist group Anna’s Archive claims to have collected approximately 86 million audio files, along with metadata for 256 million tracks, which could equate to more than 300 terabytes of data. While the archive does not represent Spotify’s full catalogue, which the company says exceeds 100 million tracks, the group claims it covers 99.6 per cent of all music actually listened to on the platform.

Spotify has been quick to confirm the incident does not constitute a data breach. In a statement, the Stockholm-based company said it had “identified and disabled the nefarious user accounts that engaged in unlawful scraping” and was actively monitoring for further suspicious behaviour.

“An investigation into unauthorised access identified that a third party scraped public metadata and used illicit tactics to circumvent DRM to access some of the platform’s audio files,” Spotify said. The company stressed that no private or non-public user data was accessed, including passwords, payment details or personal information.

We understand the archive appears to prioritise Spotify’s most-played material, helping to explain Anna’s Archive’s claim that it represents nearly all real-world listening behaviour despite falling short of the service’s total track count. Spotify has said it does not believe the music has yet been released publicly, and the archive’s full scope has not been independently verified.

Anna’s Archive has previously been known for indexing and linking to pirated books and academic research and has framed the project as an act of music preservation. In a recent blog post titled "Backing up Spotify", the group argued that streaming platforms are unstable cultural repositories due to licensing changes, regional restrictions, and corporate decision-making. It went on to state that the files would be distributed via torrent networks.

“Of course, Spotify doesn’t have all the music in the world, but it’s a great start,” the group said, describing its mission as preserving “humanity’s musical heritage”.

That justification has already been met with scepticism across the music industry. Critics argue that, unlike academic texts, commercial music remains tightly bound to licensing, royalty and artist-compensation frameworks. There is also growing concern that a dataset of this scale could be repurposed to train generative audio or music models without the consent of rights holders.

“Training on pirated material is sadly common in the AI industry, so this stolen music is almost certain to end up training AI models,” said Ed Newton-Rex, a composer and campaigner for artists’ rights, speaking to The Guardian. “This is why governments must insist AI companies reveal the training data they use.”

Spotify said it has since introduced additional safeguards “for these types of anti-copyright attacks”, as streaming services face increasing pressure from artists, labels and regulators to demonstrate tighter control over how music is accessed and reused.

Continue the discussion