For the first time since the recording industry sued AI music platforms, a court has ruled against one on the merits. On July 31, 2026, the Munich Regional Court held that Suno infringed copyrights represented by GEMA, the German collecting society, rejected Suno's fair use defense, and ordered the company to disclose its revenue and damages. This judgment marks a turning point in the legal landscape for generative AI music, moving the debate from pending settlements and licensing deals into actual courtroom precedent.

How Did We Get Here? The Timeline of AI Music Litigation

The Recording Industry Association of America (RIAA) filed the first major lawsuits against AI music companies in June 2024, suing Suno and Udio on behalf of Universal Music Group, Sony Music Entertainment, and Warner Records. The complaints alleged three categories of conduct: training AI models on millions of copyrighted recordings scraped without license from the internet, operating without licensing agreements like those governing streaming services such as Spotify or Apple Music, and generating outputs that resembled specific popular recordings when prompted with their characteristics.

What followed was a bifurcated legal strategy. Some major labels chose to settle rather than fight. Universal Music Group settled with Udio in October 2025, becoming the first major-label licensing agreement for a generative AI music company. Warner Records settled with both Udio and Suno in November 2025. But UMG and Sony continued active litigation against Suno in Massachusetts, and Sony alone continued suing Udio in New York.

The expected U.S. fair use ruling that many observers anticipated would arrive in summer 2026 never materialized. Instead, on June 30, 2026, the Massachusetts court overseeing the UMG and Sony case against Suno reset its schedule, pushing dispositive motions to April 9, 2027, meaning a U.S. fair use ruling is now a 2027 event at the earliest. That delay left a vacuum that Germany filled first.

What Did the Munich Court Actually Decide?

The Munich Regional Court's July 31 ruling is significant because it represents the first time any court, anywhere, has ruled on the merits of AI music training. The court rejected Suno's fair use argument outright, a defense that has been central to the company's legal strategy in the United States. The judgment is not final; Suno is expected to appeal. But the ruling establishes that at least one major jurisdiction does not view AI music training as a fair use activity protected by copyright law.

The court also ordered Suno to disclose its revenue and damages, a step that could expose the financial scale of the company's operations to the plaintiffs. Suno raised $400 million in 2026 at a $5.4 billion valuation, even as its legal exposure grew. The disclosure order suggests the Munich court is preparing for a damages calculation that could be substantial.

Adding weight to the labels' case, a July 2026 hack exposed Suno's own source code, corroborating the central technical allegation made by the recording industry: that Suno used commercial proxy services to route around YouTube's playback protections while pulling over two million audio clips. That evidence now sits inside active litigation, giving the labels concrete technical documentation of the alleged conduct.

How Does This Affect the U.S. Cases?

The Munich ruling does not directly bind U.S. courts, but it signals a legal direction that could influence how American judges approach fair use in the AI music context. The U.S. cases rest heavily on fair use as Suno's primary defense, after both Suno and Udio effectively conceded in their filings that their training corpora included copyrighted recordings. That concession isolated fair use as the central remaining defense on infringement claims, though it left separate DMCA (Digital Millennium Copyright Act) circumvention claims, which do not accept a fair use defense, as an independent line of exposure.

The DMCA claims carry their own teeth. The amended complaint filed in September 2025 added a specific allegation: that Suno obtained training data by stream-ripping YouTube, circumventing YouTube's rolling cipher DRM (digital rights management). DMCA circumvention carries statutory damages of up to $2,500 per act of circumvention, on top of the existing infringement claims that allow up to $150,000 per infringed work.

What Does This Mean for the Industry Split Between Licensed and Unlicensed AI Music?

The Munich ruling accelerates a trend already visible in settlement patterns: the industry is bifurcating into licensed and unlicensed AI music platforms. UMG's settlement with Udio created a licensed replacement service called Starstruck, which operates on a walled garden model, meaning outputs cannot be downloaded or distributed to external platforms such as Spotify, Apple Music, or YouTube. The licensed path offers compensation to artists and songwriters who opt in, both for the use of their catalog in training and for generation outputs that draw on it.

Suno remains on the unlicensed path, at least in the United States, despite settling with Warner. The Munich ruling suggests that unlicensed path carries real legal risk in jurisdictions outside the U.S., where fair use protections may not apply as broadly. For founders and creators considering which AI music platforms to build on or release with, the legal geography is now fragmented: licensed platforms in some regions, contested or prohibited in others.

Steps to Understand Your Legal Exposure When Using AI Music Tools

• Jurisdiction Matters: The Munich ruling applies in Germany and may influence other European courts, but U.S. fair use law remains unsettled. If you are releasing music internationally, research the legal status of AI music generation in each target market before deploying AI-generated assets commercially.
• Licensing vs. Unlicensed Platforms: Platforms like Starstruck operate under major-label licenses and restrict distribution to external services, while unlicensed platforms like Suno offer broader distribution but carry unresolved legal risk. Choose based on your distribution strategy and risk tolerance.
• Copyright Ownership of Outputs: The U.S. Copyright Office has settled that purely prompt-generated tracks carry no copyright for whoever deploys them, meaning competitors can use the same output. This applies regardless of platform; if you generate a track with a prompt, you cannot claim exclusive copyright to it.
• Consult a Copyright Attorney: Before releasing any AI-generated music commercially, consult a qualified copyright attorney licensed in your jurisdiction. The legal landscape is evolving rapidly, and what is permissible in one region may expose you to liability in another.

What Happens Next?

Suno is expected to appeal the Munich ruling, but the company now faces a two-front legal battle: defending against the Munich judgment while waiting for the U.S. fair use hearing to resume in 2027. The timing of the Munich ruling, arriving before any U.S. court has ruled on fair use, gives the labels momentum and may influence settlement negotiations. UMG's decision to settle with Udio rather than fight, followed by the Munich court's rejection of Suno's fair use defense, suggests that the legal instinct to license rather than litigate may have been sound.

For the broader AI music industry, the Munich ruling establishes that at least one major jurisdiction does not view AI music training as a protected activity. Whether other European courts follow that precedent, and whether U.S. courts reach a different conclusion when they finally rule in 2027, remains to be seen. But the legal ground has shifted. The question is no longer whether courts will scrutinize AI music training; it is how strictly they will do so, and under what legal framework.