đ Share this article Advocacy Collective Asserts to Have Extracted Countless of Music Files from the Streaming Giant An activist organization has claimed to have harvested millions of songs from the popular streaming service and is getting ready to release them online. Possible Boost for Artificial Intelligence Training Observers said the apparent leak could boost artificial intelligence firms looking for material to train their models. A group known as Annaâs Archive reported it had scraped 86 million audio tracks from Spotify and 256 million rows of associated data such as performer and album names. Spotify, which carries more than 100 million songs, stated that the breach did not represent its entire catalog. âWe have identified and deactivated the nefarious user accounts that were involved in illegal scraping,â stated a Spotify spokesperson. The Stockholm-based service, which has more than 700 million subscribers worldwide, further explained that an investigation into unauthorized access found a external actor scraped public metadata and employed illicit tactics to bypass DRM [digital rights management]. Archive's Stated Goal and Worries The streaming service does not think the audio acquired by Annaâs Archive has been made public yet. Annaâs Archive, which is known for offering links to pirated books, wrote in a blog it wanted to create a âpreservation archiveâ for music. The group asserted the audio files represented 99.6% of all music streamed to by its listeners and would be shared via âpeer-to-peer networksâ, a method of exchanging large digital files online. âOf course the service doesnât have all the music in the world, but itâs a great beginning,â commented Annaâs Archive, which describes its mission as âpreserving humanityâs knowledge and cultureâ. Model Development and Intellectual Property Debate A composer and campaigner for protecting artistsâ rights, remarked the breached music would probably be used for training AI models. âEducating models on illegally sourced material is sadly common in the tech sector, so this stolen music is almost certain to end up training AI models,â he stated. The Annaâs Archive site makes references to LibGen, a vast online repository of copyrighted books that has allegedly been used by the parent company of Facebook to train its systems. The co-founder of an artificial intelligence company posted on the professional network that individuals of the public could in theory âbuild their own personal free equivalent of the serviceâ. He said it could also enable tech companies to âeducate models on contemporary music at a large volumeâ. Intellectual property law has become a battleground between artists, writers and content creators on one side and AI companies on the other. Government Proposals and Responses In the UK, artists and writers have protested against a ministerial suggestion to let AI companies use protected work without consent unless the rights holder signals. Almost every participant to a government call for evidence on the idea has supported artistsâ concerns. The secretary of state for science, research and development informed parliament this month there was âno clear consensusâ on the matter, noting that ministers would âtake the time to get this correctâ. The government has pledged to make policy proposals on AI and copyright by March 2025.