Getty Images Drags Stability AI to UK Court in High-Stakes Copyright Clash

Two giants collide as Getty Images takes Stability AI to court over alleged copyright violations—because nothing sparks innovation like a good old-fashioned legal brawl.

Who needs creative commons when you've got billable hours? The case could set a precedent for how AI firms 'borrow' content—while the lawyers count their future yacht payments.

Lawyer Newman claims the UK case is legally in ‘uncharted territory’

Rebecca Newman, an uninvolved lawyer at Addleshaw Goddard, said although the case will be critical in setting the boundaries of the monopoly granted by UK copyright in the age of AI, it is legally in uncharted territory. However, she added that a victory for the image company could mean that Stability AI and other developers will face further lawsuits.

Cerys Davies from the Pinsent Masons law firm also said the High Court’s ruling could profoundly affect market practice and the UK’s attractiveness as a jurisdiction for AI development.

However, an associate at intellectual property law firm Brandsmiths, Sukanya Wadhwa, said Getty’s case against Stability AI is not a usual copyright infringement case. The image licensing company’s case also included a representative action on behalf of around 50K photographers and content creators whose works were licensed exclusively to it. 

“The Court will need to first consider how and where the AI training has been done. If it’s outside the U.K., do the U.K. courts have the right authority to decide on copyright infringement?”

– Sukanya Wadhwa, Associate at Brandsmiths

Getty’s CEO, Craig Peters, acknowledged that his company had to file in the U.S. and the UK because it was unclear where this training occurred. However, he stressed that these legal pursuits were “very expensive,” claiming that his company spent millions on each case. 

Stability AI relies on defenses: infinitesimal use, pastiche, caching, and hosting

Stability AI contended that considering the large amounts of non-Getty-owned materials used to train its model, any reproduction of Getty’s works in SD’s output WOULD relate to an “infinitesimally” small part of the expression of that copyrighted work.

The AI firm also argued that such reproduction could not constitute the intellectual creation of any particular author or a substantial part of any original work and, as a result, cannot amount to copyright infringement. It maintained that the outputs were generated from random noise images, and any resemblance to Getty’s works was coincidental.

Stability AI also relies on the statutory “pastiche” fair dealing defense under UK copyright law. It submitted that even if any output from Stable Diffusion resembled Getty’s works, such output should be protected as a pastiche–which was described in the Only Fools case as either imitating the style of another work or being a medley or a combination of several different works. 

The AI firm also asserted that any resemblance to the original work in the outputs was rare, unintentional, and not a substitute for the original work. Such use constituted fair dealing as it did not harm the market for the original work.

Stability AI further sought to rely on the ‘caching’ and ‘hosting’ SAFE harbors provided under the UK’s e-commerce regulations. The AI company argued that it acted as a neutral intermediary for user-generated content and should not be held liable for infringing acts committed by users of its hosted services.

It also maintained that the injunctions sought by Getty would require it to monitor all user activity, which would be impossible and/or disproportionate for Stability AI to comply with. The injunctions would also contradict legal principles prohibiting general monitoring duties for service providers.

Your crypto news deserves attention - KEY Difference Wire puts you on 250+ top sites