Suno, the AI music startup, announced Thursday a series of changes to combat misuse of AI-generated songs, including audio watermarking and a new download policy. The measures target schemes where users mass-produce AI tracks, upload them to streaming services, and use bots to collect fraudulent royalties.
The scale of the problem is significant. Earlier this year, a North Carolina man pleaded guilty to using hundreds of thousands of AI-generated songs and billions of fake streams to collect more than $8 million in royalties.
Co-founder and CEO Mikey Shulman wrote in a blog post that Suno will introduce audio watermarking and fingerprinting technology to make it easier for music platforms to identify and trace songs created with the tool. The company will also limit how many songs users can download, a move aimed at preventing mass distribution. Suno updated its community guidelines to explicitly prohibit scams, spam, fake engagement, presenting deceptive audio as authentic, recreations of existing songs, and the use of copyrighted material or a real person's voice or likeness without permission.
Balancing fraud prevention and creative legitimacy
Suno is trying to avoid condemning all AI music while cracking down on abuse. "These tools are designed to be durable and resistant to tampering, without affecting the listening experience," wrote Shulman. "They are also not intended to pass judgment on whether a song is good, meaningful, or sufficiently human."
He added that it should be up to artists and platforms to decide whether to disclose that a song was made using AI. In a somewhat ironic move, Suno recently announced plans to let users make vinyl copies of their AI-generated songs.
Legal pressure mounts
The new safeguards arrive as Suno faces legal battles on several fronts. Universal Music Group and Sony Music Entertainment are suing the company, accusing it of using copyrighted recordings without permission to train its AI models. Warner Music Group, which was initially part of the lawsuit, settled with Suno last year and reached an agreement allowing participating artists and songwriters to opt into approving the use of their music and likeness in AI-generated songs.
Just last week, a German court ruled that Suno violated the copyrights of musicians represented by GEMA, the country's music-licensing agency. Suno disputed the ruling and is considering an appeal.
The company is also facing a proposed class-action lawsuit in Massachusetts over a data breach that allegedly exposed the personal information of more than 55 million users. 404 Media reported that the breach revealed Suno scraped YouTube, Deezer, Genius, and other online sources to train its models.
Suno did not immediately respond to a request for comment. Meanwhile, Spotify appears to have gone all-in on AI: the streaming platform and Universal recently announced a tool that will let users create AI-generated covers and remixes using music from participating artists.
Why this matters for legal professionals
For lawyers and legal teams, the Suno case touches on several active legal questions: whether training AI on copyrighted material violates copyright law, who is liable when AI-generated content is used for fraud, and how watermarking and disclosure requirements will be enforced. The German ruling and the U.S. lawsuits could set precedents that shape how AI music companies operate. Legal professionals tracking these cases will need to watch how courts treat both the training-data claims and the fraud-prevention measures. Those who want to build familiarity with these issues can follow AI for Legal coverage or explore an AI Learning Path for Paralegals.
Your membership also unlocks:
