In a pivotal case unfolding in the British High Court, Getty Images has decided to withdraw its copyright infringement claims from its lawsuit against Stability AI. This legal confrontation between the Seattle-based company and the AI firm, known for its popular image-creating tool Stable Diffusion, could have far-reaching implications for both the creative and technology sectors.
The conflict originally arose from the common practice among tech companies of utilizing extensive online collections of images and text to train their AI systems. Being among the first, Getty Images initiated legal actions against such practices in both the United States and the United Kingdom earlier this year.
Throughout the trial, Getty presented evidence highlighting the meticulous creative effort involved in the photographs housed in its collection. This ranged from serene Caribbean beachscapes to captivating celebrity shots, such as Donald Glover at an awards ceremony or a candid of Kurt Cobain smoking. These genuine images were contrasted with AI-generated images produced by Stability AI.
However, proving copyright infringement in the U.K. proved challenging, primarily due to a jurisdictional technicality. Despite being headquartered in London, Stability’s AI training took place outside the U.K. on servers operated by Amazon, an American tech giant. Alex Shandro, an AI legal expert, observed that demonstrating the U.K. connection was always expected to be difficult, given that a significant portion of the AI training occurred in the United States.
This move by Getty marks the second recent defeat for efforts within creative industries to legally challenge the operations of the growing generative AI sector. In a separate development, a U.S. federal judge in California ruled that the chatbot company Anthropic did not violate the law when training its product on copyrighted material, although the issue of sourcing those texts from pirated sites remains unresolved.
In its legal complaint in the U.K., Getty accused Stability of exploiting its images in violation of several intellectual property rights, including copyright, trademark, and database rights. However, with recent proceedings and after considering extensive witness and expert testimony, Getty made the strategic decision to drop the copyright allegation, choosing to focus on claims of trademark infringement, passing off, and secondary copyright infringement. The latter involves the reproduction of images still bearing Getty’s watermark by the AI model.
Stability welcomed Getty’s decision with its spokesperson stating the company’s satisfaction at seeing multiple claims withdrawn and expressing gratitude for the court’s handling of the case. The concluding arguments in court are anticipated to wrap up by the week’s end, with the judge’s official decision expected to follow later.
The outcome of the remaining allegations, particularly those addressing trademark issues, could significantly influence the U.K.’s regulatory approach to AI tools developed with U.S.-based data. Nina O’Sullivan, a partner in a leading British law firm, noted the potential impact of the court’s decision on these international regulatory frameworks.