Legal action being brought by Getty Images against Stability AI over alleged copyright infringement is a “day of reckoning” over the use of photographs to train AI models, the High Court has heard.
The visual content company, based in the United States, is suing Stability AI Limited in London, alleging that it scraped millions of images from Getty’s websites without consent to unlawfully train its deep learning AI model Stable Diffusion.
Getty claims this infringes copyright and that the works created by Stable Diffusion, which can generate synthetic images in response to commands entered by users, are also in breach of copyright laws.
Stability AI is opposing the claim, which it told the court was “an overt threat” to its business.
On the first day of trial on Monday, Lindsay Lane KC, for Getty, told the court the company “recognises that the AI industry overall may be a force for good”, but that did not justify AI companies “riding roughshod over intellectual property rights”.
She said: “It is important to say that this is not a battle between creatives and tech, where a win for Getty Images means the end of AI.
“Getty Images’ position is that the two industries can exist in synergistic harmony.”
She continued: “There is no AI without creative works on which to train. The problem is when AI companies want to use those works without payment.”
She added: “[The case] is about the straightforward enforcement of intellectual property rights.”
Lane told the court Stability “did not care” if it used works that were copyrighted, had watermarks or if they “were not suitable for work or pornographic”.
She said: “This trial is the day of reckoning for that approach.”
In a judgment in May concerning preliminary matters before the trial, Mrs Justice Joanna Smith said that the case was “highly complex and technical” and raised “numerous novel issues for consideration by the court”.
She added that the court had had to “devote unprecedented amounts of court time” to deal with the case.
Hugo Cuddigan KC, for Stability AI, said in written submissions that the case concerns “an issue of real societal significance” related to AI technology and copyright law, adding that Getty “appear to see generative AI as an existential threat”.
He continued that some of Getty’s claims “represent an overt threat to Stability’s whole business, and the wider generative AI industry”.
He said: “By pursuing them, Getty seek to establish that such generative AI image models should not be available to UK users.
“We consider Getty’s output claims to be susceptible to dismissal on a host of independent bases.
“Indeed, they are so weak that their introduction seems likely the result of an intuitive reaction to Stable Diffusion and not a clear-headed analysis of the rights in issue.”
Cuddigan said that Getty’s case “can be dismissed with confidence and clarity”.
The trial is expected to conclude later in June, with a judgment expected in writing at a later date.
Email pged@pressgazette.co.uk to point out mistakes, provide story tips or send in a letter for publication on our “Letters Page” blog
