A German regional court has ruled that Suno, a U.S.-based AI music generation company, violated copyright law by using copyrighted music to train its generative AI system without authorization. The Munich court's decision, handed down on July 31, 2026, marks a significant legal setback for the AI industry and reflects a growing global shift toward stricter enforcement of intellectual property rights in AI development.

The case centered on Suno's use of music covered by GEMA, Germany's largest music licensing agency. Unlike earlier AI copyright disputes that focused on whether training constitutes "fair use," this ruling emphasizes a more fundamental principle: companies cannot simply scrape and use copyrighted works without acquiring proper licenses, regardless of whether the content is publicly accessible online.

Why Is the "Publicly Available" Defense Failing in Court?

For years, AI developers operated under a simple assumption: if content was publicly accessible online, it could be scraped and used to train AI systems. That assumption is now crumbling under legal scrutiny across multiple jurisdictions. The internet was never designed as a free training repository for commercial AI systems, and courts are increasingly recognizing that copyright protections apply regardless of whether content is freely accessible.

The distinction between content that is "available" and content that is "licensed" is reshaping how AI systems are built in 2026 and beyond. A newspaper article behind a paywall, an artist's portfolio, a GitHub repository, or even publicly indexed website content may still carry licensing restrictions, contractual obligations, database rights, or copyright protections. Many AI developers underestimated this legal risk, assuming that merely "viewing" content during training posed no infringement concerns.

Courts are now evaluating whether companies copied, stored, reproduced, transformed, or commercially benefited from protected works during model training. The legal scrutiny intensified because many datasets used in foundational model development were allegedly sourced from pirated books, scraped journalism, copyrighted images, and proprietary databases without consent or licensing agreements.

How Are Global Courts Reshaping AI Training Standards?

Several landmark cases are now establishing new legal precedents for AI training data governance. These disputes signal a coordinated global pushback against the "move fast and break things" approach that characterized early AI development:

• The New York Times v. OpenAI and Microsoft: The Times sued OpenAI and Microsoft, alleging that copyrighted news articles were used without authorization to train generative AI systems. A judge denied OpenAI's motion to dismiss certain copyright infringement claims, allowing the core parts of the lawsuit to proceed and testing whether AI training qualifies as "transformative use" under copyright law.
• Anthropic and the "Pirated Library" Problem: In 2025, a U.S. federal judge held that training AI systems on books could qualify as fair use under specific circumstances. However, the same decision found serious problems with how the underlying data was obtained and stored, concluding that maintaining millions of pirated books in a centralized training library was not protected.
• Thomson Reuters v. ROSS Intelligence: A court rejected a fair use defense involving copyrighted legal research content used to build a competing AI-driven platform. This decision is particularly important for enterprise AI because it demonstrates that courts may treat proprietary databases differently from broad internet content, and that "internal use" does not automatically reduce legal risk.
• India's ANI Media v. OpenAI: India's Delhi High Court is examining whether AI training on copyrighted news content amounts to copyright infringement, even when the content is publicly accessible. This case is expected to become a landmark precedent for how Indian law treats AI training data and web scraping.

The Suno ruling carries particular weight because it demonstrates that European courts are willing to enforce copyright protections against AI companies operating globally. GEMA CEO Tobias Holzmuller emphasized the broader implications of the decision.

"The court made it clear today: AI models based on the theft of intellectual property do not enjoy legal protection. AI providers must acquire licenses and cannot simply help themselves to our members' works for free. In doing so, the court has significantly strengthened Europe as a hub for culture," Holzmuller stated.

Tobias Holzmuller, CEO at GEMA

Kai Welp, GEMA's general counsel, added that the ruling was "very gratifying to see the assessment of the legal situation in the USA and the fact that we can sue in Munich regardless of where the training takes place". This statement underscores a critical shift: companies cannot evade copyright enforcement by locating their infrastructure outside Europe.

Kai Welp, GEMA's general counsel

What Does This Mean for AI Companies and Enterprises?

The Suno ruling and related cases are reshaping the compliance landscape for AI development. The market has moved past the early "move fast" phase of AI development, and data governance is now as important as model performance. Enterprises deploying generative AI systems, procurement teams integrating third-party AI tools, legal departments approving AI workflows, and boards overseeing digital transformation are all exposed to the same risk chain.

Suno said it is evaluating its options in the wake of the ruling, including an appeal. A company spokesperson stated: "From the beginning, we trained our models to create new songs, not reproduce existing ones, and built protections into our platform. We disagree with today's ruling, which rests on a fundamental mischaracterization of how Suno's technology works, how it is used and how U.S. law applies, and are evaluating all available options, including an appeal".

Suno

However, Suno does not expect the decision to impact cases in the United States, where the legal standards for fair use remain more permissive than in Europe. One of the highest-profile U.S. cases involves a lawsuit brought by three studios, Warner Bros., NBC Universal, and Disney, challenging Midjourney's unauthorized use of copyrighted works, including intellectual property from Star Wars and Superman. Midjourney has argued that its use of the material constitutes fair use under the Copyright Act.

The divergence between European and U.S. legal standards creates a complex environment for global AI companies. European courts are drawing a clear line: licensing is mandatory, and the "publicly available" argument no longer provides legal cover. This approach reflects a commitment to protecting creators' rights in the digital economy, as noted by Wolfram Weimer, Germany's minister of state for culture and media, who called the ruling "an important signal for strengthening the rights of creatives in the digital music market".

The Suno case follows a November 2025 German court ruling in favor of GEMA over its challenge to OpenAI's use of song lyrics, indicating a consistent pattern of judicial skepticism toward AI companies' copyright practices. As these legal battles continue to unfold, the message to AI developers is clear: acquiring proper licenses and respecting copyright protections is no longer optional but essential to operating legally in major markets.