Getty Images has withdrawn its primary copyright claims against Stability AI in the UK lawsuit over Stable Diffusion’s training data. The case involves allegations that millions of Getty images were used without permission to train the generative AI model. While the withdrawal narrows the legal focus, it does not end the overall case. Getty is still pursuing claims of trademark infringement and secondary copyright violation. The final ruling could impact how intellectual property laws apply to AI and the use of visual content in model training.
The Backstory: Getty vs. Generative AI
The legal dispute between Getty Images and Stability AI began amid rising concern over AI training on unlicensed online content. Getty claimed Stability AI scraped millions of copyrighted images without authorization to train its Stable Diffusion model. Some AI outputs even included versions of Getty’s watermark, prompting serious copyright and trademark concerns. The lawsuit became a major legal test for AI trained on internet-sourced datasets. Artists, rights holders, and regulators closely followed the case as it challenged AI’s use of copyrighted content without permission.
Getty Backs Down on Key Copyright Claims
On June 25, 2025, Getty Images dropped its central copyright infringement claims in the UK. Specifically, it abandoned the part of its lawsuit that focused on the training of Stable Diffusion on Getty’s images, which Getty had argued constituted direct copyright infringement.
Why the Retreat
According to legal analysts and court filings:
- Jurisdiction issues: Stability AI has stated that its model training took place on servers that were located outside the UK, possibly in the US, creating jurisdictional ambiguity under UK copyright law.
- Burden of proof: Getty needed to demonstrate that Stable Diffusion’s outputs significantly reproduced its original photographs to support its claims. This is particularly challenging with generative AI, which learns patterns rather than retaining or copying specific images.
- Strategic pivot: Getty can enhance its chances of a positive ruling by focusing on claims that will be easier to prove, such as claims of trademark infringement under UK law.
According to a TechCrunch report, Getty Images has dropped its primary copyright claims but continues to pursue Stability AI on other legal grounds.
What’s Still in Play?
Although Getty has withdrawn its primary copyright claims, several other allegations remain active in the UK lawsuit:
- Trademark infringement
Getty alleges that Stability AI’s model has produced images containing its watermark. This might confuse users and weaken Getty’s brand identity, potentially resulting in a strong claim under UK trademark law. - Secondary copyright infringement
Even if the model was trained outside the UK, Getty argues that distributing or using the AI model within the UK still constitutes infringement. This claim targets how the trained model is accessed and deployed in the region. - Database rights violation
Getty also suggests that UK and EU law protect its curated image database and that anyone who uses this collection without permission infringes on its database rights.
The Getty Images vs. Stability AI lawsuit also reflects broader tensions around jurisdiction and regulation, similar to what we explored in our article on federal versus state laws and Big Tech’s legal strategy.
Why This Case Will Shape the Future of AI and Creativity
The legal dispute between Getty Images vs Stability AI is more than a courtroom battle. It represents a turning point in how societies define ownership, originality, and accountability in the era of generative technology. Getty dropping its main copyright claims shows legal challenges while emphasizing the strain modern AI systems place on traditional frameworks.
With trademark and secondary copyright claims still under review, the court’s final ruling will have implications far beyond this case. It may shape global standards for AI training, intellectual property protection, and the coexistence of creators and developers in technology.
Regardless of the result, legal and ethical experts expect this case to become a pivotal point in the landscape of AI.
