Getty images let his primary claims of copyright against stability AI fall on the London Supreme Court on Wednesday, making one of the most viewed legal fights on how AI companies used copyrighted content to train their models.
The move does not fully end the case – Getty still strives for other claims and a separate lawsuit in the US – but it underlines the gray areas around the future of content possession and use in the age of generative AI. The development also only comes a day after an American judge chose the side of anthropic in a similar dispute as to whether the training of AI on books without the author’s permission is contrary to copyright legislation.
Getty has sued Stability AI – The startup behind AI image generator stable diffusion – in January 2023 after he claimed that stability used millions of copyright images to train his AI model without permission.
The Image Database Company also claimed that many of the works generated by stable diffusion were similar to the copyright content used to train it. Some, Getty said, even had his water brands on them.
Both claims were dropped from Wednesday morning.
“The training claim was probably dropped because Getty does not make a sufficient link between the infringing actions and the British jurisdiction for copyright legislation to bite,” said Ben Maling, a partner at the EIP law firm, tech crunch in an e -mail. “In the meantime, the output claim has probably fallen because Getty cannot establish that what the reproduced models reflect a significant part of what has been made in the images (for example by a photographer).”
In Getty’s final arguments, the company’s lawyers said they dropped those claims because of weak evidence and a lack of expert witnesses of stability AI. The company has framed the relocation as a strategic frame, so that both and the court can concentrate on what Getty believes are stronger and more profitable allegations.
What remains in Getty’s trial is a secondary infringement claim and claims for infringement of trademarks. Regarding the secondary infringement claim, Getty argues in essence that the AI models themselves infringes copyright legislation, and that the use of these models in the UK could form infringing articles, even if the training outside the UK would take place.
“Secondary infringement is the person with a widest relevance for Genai companies that train outside the UK, namely via the models themselves, which may be infringing articles that are then imported into the UK,” said Malinging.
A spokesperson for Stability AI told Techcrunch that the startup was “happy to see Getty’s decision to drop several claims after the end of the testimony.”
The spokesperson also noted that stability was convinced that the trademark of Getty and the passage of claims will fail because consumers do not interpret the water brands as a commercial message of stability AI.
The US division of Getty also sued Stability AI in February 2023 due to trademarks and infringement of copyright. In that case, Getty claimed that stability used no less than 12 million copyrighted images to train its AI model without permission. The company requires compensation for 11,383 works at $ 150,000 per infringement, which would be up to a total of $ 1.7 billion.
Separately, Stability AI is also mentioned in another complaint alongside Midjourney and Deviantart after a group of visual artists has sued the three companies due to copyright infringement.
Getty Images has its own generative AI offer that uses AI models that are trained on Getty Istock Stock Photography and VideoBibliotheken. With the tool, users can generate new license boots and illustrations.
#Getty #falling #important #copyright #claims #stability #British #court #case #continues #Techcrunch