Leaked files show Suno trained AI music models on scraped songs from YouTube, Deezer, and Genius

A hacking incident has yielded internal Suno data showing the company trained its AI music models on audio and lyrics scraped from major platforms, including YouTube Music, Deezer, and Genius, according to reporting by 404 Media and The Verge. The leaked materials include Suno source code from 2023 and 2024, as well as scraping instructions targeting audio files from YouTube Music, Deezer, Genius, Pond5, Jamendo, Freesound, and the International Music Score Library Project (IMSLP).

Files referenced in the reporting indicate Suno compiled datasets totaling over 2 million YouTube Music clips, hundreds of thousands of hours from Deezer, Genius, IMSLP, Jamendo, and Pond5, and hundreds of hours of Freesound and MuseScore lyrics. Additional code reportedly sought to download roughly one million hours of podcasts via PodcastIndex.

The revelations come amid lawsuits against Suno, including one filed by the Recording Industry Association of America (RIAA), which alleges the company used copyrighted materials to train its models. Suno has publicly argued that training on publicly available music files is protected under fair use, though the RIAA’s filings claim Suno unlawfully circumvented YouTube’s copyright protections by using stream-ripping techniques.

Suno confirmed a security incident in November 2025, asserting that the breach primarily involved outdated source code and that no sensitive personal information was compromised. The company stated it did not notify customers because the exposed data was not deemed sensitive under applicable privacy laws. Customer information accessed included email addresses, phone numbers, and Stripe payment details, though Suno claimed it does not retain full credit card numbers.