Imagine opening a music app not just to generate a song, but to reshape an authorized performance, isolate its drums, and publish a remix that sends money back to the people who made the original. Suno is betting that this future can replace the copyright battles that defined its first era.
A reset for Suno’s music engine
The AI music company has introduced Suno v6, a three model lineup trained on licensed material from partners including Warner Music Group, BMG and Believe, according to TechCrunch. Suno says the new systems do not use the training data behind its earlier models, which became the subject of accusations and lawsuits over alleged unauthorized use of copyrighted recordings.
The paid Suno v6 is designed to be the dependable, controllable option. Suno v6-wild is a paid experimental model intended to produce stranger and less predictable ideas, while the faster v6-mini is available to all users. The company plans to retire its older models.
The change is more than a software update. Users can revise sections of songs through prompts or lyrics, provide text, images or video as creative references, and separate instruments from a sample before building a new beat. Suno also plans artist authorized remixing through a program being developed with labels.
That could make licensing visible in the creative process. A future user might select a singer’s approved vocal style, alter a chorus, and share the result under terms that automatically divide revenue among the artist, label and platform.
The marketplace still needs rules
The commercial promise remains untested. Suno has settled with Warner Music Group and reached an agreement with BMG, but litigation continues with Sony, Universal Music Group, artist Jason Isbell and users who allege security failures. The announcement also followed Suno’s acknowledgment that earlier models were trained using YouTube videos, making its claimed break with past practices particularly significant.
Several practical questions will determine whether the model works. How much will licensors and artists receive? Will musicians opt into remix rights, and can they withdraw later? How clearly will Suno distinguish music created with older models from work made using licensed systems?
Watermarking, download limits and stronger detection may also be necessary to prevent mass uploads, imitation and streaming fraud. Suno has raised more than $819 million, according to PitchBook data cited by TechCrunch, giving it the resources to experiment. Its larger challenge is proving that consent can become a product feature, not merely a response to litigation.
- Bryan Ledgard from Yorkshire, UK · CC BY 2.0
This article was written with the assistance of an AI system and published automatically.