In a significant development in the ongoing legal battles surrounding AI-generated content, Getty Images has dropped its primary copyright infringement claims against Stability AI in the UK High Court. This move marks a major shift in the closely watched case, which has broad implications for the creative and technology industries. The lawsuit, initiated in early 2023, centers around Stability AI's Stable Diffusion, a popular AI image-making tool, and its use of copyrighted images for training purposes.

Getty's initial lawsuit alleged that Stability AI unlawfully copied and processed millions of copyrighted images, including those owned by Getty Images, to train its Stable Diffusion model without obtaining the necessary licenses. However, as closing arguments began, Getty's legal team announced it would no longer pursue these core copyright infringement claims. Instead, the company will focus on trademark infringement, "passing off," and secondary copyright infringement.

Several factors may have contributed to Getty's decision. Legal experts suggest Getty struggled to establish that the alleged copyright infringement occurred under UK jurisdiction, as much of the AI model's training likely took place in the United States. Additionally, it may have been difficult for Getty to prove that the AI-generated images reproduced a "substantial part" of Getty's copyrighted works. AI legal expert Alex Shandro noted it would be challenging to prove a connection to the U.K. because most of the training happened in the U.S. Ben Maling, a partner at law firm EIP, stated that the training claim was likely dropped due to Getty failing to establish a sufficient connection between the infringing acts and the U.K. jurisdiction for copyright law to apply. He also noted that the output claim was likely dropped because Getty failed to establish that what the models reproduced reflects a substantial part of what was created in the images.

Despite dropping the primary copyright claims, Getty continues to pursue other aspects of the lawsuit. These include allegations that Stability AI infringed Getty's trademark by reproducing its watermark on some AI-generated images, potentially misleading consumers. Getty also argues that even if the AI training occurred outside the UK, using the models within Britain could constitute secondary copyright infringement, as it involves importing infringing articles.

Stability AI has welcomed Getty's decision to drop the copyright claims. A spokesperson stated, "We are pleased to see Getty's decision to drop multiple claims after the conclusion of testimony." The company has consistently denied any wrongdoing and maintains that its activities do not infringe on Getty's intellectual property rights.

The remaining claims in the UK lawsuit, as well as a separate but related lawsuit filed by Getty in the United States, could still have significant implications for generative AI companies. The U.S. case, which includes similar allegations of unauthorized use of over 12 million images, seeks $1.7 billion in damages and is awaiting a ruling on a motion to dismiss.

More broadly, the Getty Images v. Stability AI case highlights the complex legal and ethical challenges surrounding the use of copyrighted material in AI training. The outcome of these cases will likely shape the future of AI law and copyright policy, influencing how AI companies develop and deploy their technologies and how creators protect their intellectual property rights. A ruling in favor of Getty Images could lead to increased licensing costs for AI companies, potentially stifling innovation and consolidating power among larger firms. Conversely, a ruling in favor of Stability AI could accelerate AI development but raise concerns about the devaluation of creative works. The court's ruling will significantly influence the future of AI law and copyright policy, providing guidance on regulating data-driven innovation in the coming years. A conclusion is expected by June 30th, 2025.

The legal battle underscores the need for a balanced framework that respects both the rights of human creators and the opportunities presented by generative AI. As AI technology continues to evolve, it will be crucial to establish clear guidelines and regulations that foster innovation while safeguarding intellectual property rights.