top of page

Getty drops key copyright claim against Stability AI

  • Voltaire Staff
  • Jun 26
  • 2 min read
Image Source: Unsplash
Image Source: Unsplash

Getty Images has dropped one of the central contentions from its high-profile UK lawsuit against artificial intelligence company Stability AI, abandoning a major copyright infringement claim as closing arguments began Wednesday at Britain's High Court. 


Despite the move, the legal battle continues over remaining allegations involving trademark violations and indirect copyright infringement.


The withdrawal marks a significant shift in Getty's strategy. 


The company, based in Seattle, had originally accused Stability of unlawfully using its copyrighted photos to train the popular AI image generator Stable Diffusion. But after expert testimony and a week of witness examination, Getty acknowledged the difficulty of proving a direct copyright breach under UK law—mainly due to where the AI model was trained.


Although Stability is headquartered in London, the training of its AI systems occurred outside the UK, primarily on cloud computing infrastructure operated by Amazon in the United States. That key technical detail posed a major legal hurdle for Getty.


"It was always anticipated to be challenging to prove that connection to the UK because we know that most of the training happened in the US," said Alex Shandro, an AI legal expert with A&O Shearman who observed the trial, according to Associated Press.


In a copy of its closing arguments, Getty described its choice to drop the core copyright claim as a "pragmatic decision," opting instead to pursue claims of trademark infringement, passing off, and secondary copyright infringement—focused on what happens within the UK when the AI tool is used.


The company continues to argue that Stability violated its trademark by training on watermarked Getty images, which led to some of those watermarks appearing in AI-generated content. 


Getty also maintains that Stability indirectly infringed on its copyright by distributing AI-generated images in the UK, regardless of where the training took place.


The retreat from the core copyright claim follows another setback for rights holders earlier this week, when a US judge ruled that Anthropic, the company behind Claude, did not violate copyright law by training on millions of books—though questions remain over how those books were sourced.


Stability AI welcomed the development. "We are pleased to see Getty’s decision to drop multiple claims after the conclusion of testimony."


Commenti


Stay up-to-date with the latest news in science, technology, and artificial intelligence by subscribing to Voltaire News.

Thank You for Subscribing!

  • Instagram
  • Facebook
  • Twitter

© 2023 by Voltaire News Developed & Designed by Intertoons

bottom of page