Hack reveals Suno AI training on YouTube Music

A November 2025 breach of Suno exposed source code detailing the AI music company's use of millions of songs scraped from YouTube Music and other platforms. The hacker shared files with 404 Media showing over 2 million clips from YouTube Music alone along with thousands of hours from Deezer, Genius and Pond5. Suno confirmed the incident but said it involved outdated code and no sensitive data.

The breach occurred when a hacker known as ellie.191 used a supply chain attack to access Suno's GitHub and cloud credentials. Files from 2023 and 2024 showed Suno scraped music and lyrics from YouTube Music, Deezer, Genius and stock libraries including Pond5. The data also included customer records from payment processor Stripe.

Suno stated in a response that the incident was contained quickly and involved only outdated source code no longer in use. The company added that no sensitive personal information was compromised and that it does not store full credit card numbers.

Suno has faced ongoing copyright lawsuits from major record labels over its training practices. The company maintains that its use of publicly available music constitutes fair use under copyright law. Warner Music Group previously settled its claims with a licensing agreement.

Verwandte Artikel

Suno, der KI-Musikgenerator, der derzeit mit Klagen von großen Musiklabels konfrontiert ist, wurde im vergangenen Jahr Ziel eines Hackerangriffs. Der Vorfall wurde diese Woche durch durchgesickertes Material bekannt, das Journalisten zugespielt wurde.

Von KI berichtet

Musiker, darunter SZA und der Produzent Kenneth Blume, haben scharfen Protest geübt, nachdem sie ihre Songs in Datensätzen entdeckten, die zum Training von KI-Musikgeneratoren verwendet werden. Die Reaktionen folgten auf die Veröffentlichung eines KI-Erkennungstools durch The Atlantic in der vergangenen Woche.

A study has found that unauthorized AI-generated images and videos using celebrities' likenesses caused losses of up to 4.5 billion yen. Over 43,000 such items appeared in a two-month period last year.

Von KI berichtet

A new report indicates that most companies have released software containing known security flaws. The problem is especially pronounced with AI-created code, which exceeds the speed of manual fixes.

Diese Website verwendet Cookies

Wir verwenden Cookies für Analysen, um unsere Website zu verbessern. Lesen Sie unsere Datenschutzrichtlinie für weitere Informationen.
Ablehnen