Suno releases first AI music model trained with record industry licenses

Suno v6 uses licensed data from Warner Music Group, BMG, and Believe, but questions remain about older training data.

edit
By LineZotpaper
Published
Read Time1 min
Suno has released its first AI music model, v6, trained from the ground up using data licensed from record labels including Warner Music Group, BMG, and Believe, marking a departure from previous models that relied on publicly scraped data of uncertain legality.

Suno announced the launch of its v6 AI music model, the first built with collaboration from the record industry. Jack Brody of Suno told The Verge that v6 was "trained from the ground up, with a new set of data that does not include the same data that our previous models were trained on." The training data includes content licensed from partners such as Warner Music Group, BMG, and Believe, as well as user data. However, it remains unclear whether this means v6 is entirely free of the dubiously obtained content that Suno has previously faced criticism over.

The v6 release actually comprises three distinct models: v6, v6-wild, and v6-mini. The mini version is available for free to all users and is designed for fast, responsive output. This approach signals a potential shift for Suno, moving from contentious data scraping practices toward a more cooperative stance with music rights holders.

§

Analysis

Why This Matters

  • This is the first major AI music model launched with explicit licensing from major record labels, potentially setting a precedent for how AI music companies operate.
  • It represents a possible path to resolving the widespread legal disputes between AI companies and the music industry over copyright and training data.
  • The availability of a free tier (v6-mini) could rapidly expand the user base for AI-generated music, impacting artists, producers, and consumers.

Background

Suno has been at the center of debates over AI and copyright. Its earlier models were trained on data scraped from the web, including, reportedly, copyrighted music from platforms like YouTube. This led to lawsuits and criticism from record labels and artists. The new v6 model appears to be an effort to legitimize its operations by striking licensing deals with major music groups—Warner Music Group, BMG, and Believe. The company has not fully detailed the composition of the new training dataset, leaving some uncertainty about whether all previously contested material has been removed.

Key Perspectives

Record Labels (Warner Music Group, BMG, Believe): They have chosen to license music to Suno, likely seeing AI music as an inevitable development that they can monetize rather than fight. Their deals give them control and revenue streams from the new model. Artists and Songwriters: Many remain deeply skeptical. If v6 still includes any data from previous scraping—or if user data includes copyrighted material—artists may feel their work is still being used without permission. The lack of transparency on the training data fuels this concern. Critics/Skeptics: They note that Suno has not confirmed that v6 is completely free of the dubious content its earlier models used. Without full disclosure, the model could still be built on a foundation of questionable data, undermining the promise of a fresh start.

What to Watch

  • Whether Suno will release a detailed breakdown of v6's training data sources.
  • How record labels and artist groups react if evidence of uncleared content emerges.
  • The uptake of the free v6-mini tier and its impact on the broader music creation landscape.

Sources

newspaper

Zotpaper

Articles published under the Zotpaper byline are synthesized from multiple source publications by our AI editor and reviewed by our editorial process. Each story combines reporting from credible outlets to give readers a balanced, comprehensive view.