AI Copyright Clash: UK Court to Rule on Image Scraping

A landmark legal battle is underway in London, pitting Getty Images against Stability AI over the use of copyrighted images in AI model training. The case, heard in London’s High Court since June 9, 2025, could reshape the future of generative AI and intellectual property rights.

Getty vs. Stability AI: The Core of the Dispute

Getty Images accuses Stability AI, the creator of Stable Diffusion, of illegally scraping millions of its copyrighted images to train its AI models. Getty alleges that Stability AI accessed and altered its content via the LAION-5B dataset, a vast collection of image-text pairs used to train AI. They argue that this unauthorized use infringes on their copyright and trademarks.

Getty claims that Stability AI not only copied images but also altered them during the AI training process, sometimes even reproducing Getty’s watermarks in generated images. They further argue that users can prompt the AI with Getty images, resulting in near-identical outputs, exacerbating the infringement.

The lawsuit extends beyond copyright, with Getty also alleging trademark and database rights infringement. They claim Stability AI used the “GETTY IMAGES” and “ISTOCK” trademarks in both training data and generated outputs.

Stability AI’s Defense

Stability AI admits to using some Getty Images for training purposes but contends that this occurred outside the UK. The company denies any infringement, arguing that it did not use Getty’s trademarks in a commercial context and that any similarities in generated images are coincidental, attributing responsibility to users, not the platform itself.

Stability AI challenges the validity of Getty’s database rights, disputing the claim that they unlawfully extracted and reused a significant portion of Getty’s database content for AI training.

A Stability AI spokesperson has described Getty’s lawsuit as a threat to the generative AI industry, arguing that their tools enable artists to create works based on “collective human knowledge,” which aligns with “fair use and freedom of expression.”

Impact on the AI and Creative Industries

This case arrives at a critical juncture, as creative industries grapple with the implications of AI models trained on copyrighted material. Legal experts anticipate that this trial will provide crucial guidance on intellectual property rights in the age of AI.

Iain Connor, an intellectual property partner at Michelmores LLP, emphasized the case’s significance: “This is the most significant AI case to reach the English High Court. The legal community is waiting with bated breath to see how far the judgment will go to rule on the legality or otherwise of the use of AI models.”

He further noted the complexity of the case, highlighting that the judge’s decision could broadly impact the lawfulness of AI’s use of input materials and whether AI outputs infringe on third-party rights. Alternatively, the decision could narrowly address jurisdictional issues, potentially limiting its overall impact.

The outcome of this trial will likely set a precedent for future AI-related intellectual property disputes, shaping the balance between technological innovation and copyright protection.

Deeba Ahmed is a veteran cybersecurity reporter at Hackread.com.