- within Technology, Coronavirus (COVID-19) and Environment topic(s)
- in United States
Getty has largely lost its case against Stability AI, losing both the claim of secondary copyright infringement via importation of an article which is an infringing copy into the UK and the claim of possessing or dealing with an article which is an infringing copy – with some very limited success on its trade mark infringement claims (Getty Images (US) Inc and others v Stability AI Limited[2025] EWHC 2863 (Ch), 4 November 2025).
Heather Newton, HSF Kramer IP Of Counsel, with a special interest in IP & AI, commented:
"Whilst this is not the complete answer to the AI/IP conundrum that it originally promised to be, the judgment nevertheless helps inform what won't work for a claim in copyright infringement. The court has found that the Stability AI diffusion models do not retain or carry copies of training materials in a traditional copyright infringement sense and therefore cannot be an infringing article. The key infringement (not decided in this case) is the reproduction of the copyright material for the act of training itself, but this did not take place in the UK and was dropped by Getty at trial. It is a difficult outcome for Getty, but helpful for those involved in the AI and copyright industries to focus the complex debate around the use of copyright material to train AI."
We will post on the implications of this case in more detail shortly, but for now here is a brief summary of the High Court's findings.
Summary of the High Court's findings
The case concerned claims brought by Getty against Stability in relation to the unlicensed use of images from Getty's image banks to train its AI. The claims had initially been in relation to the direct infringement of those images in the training of the AI but these claims were dropped at closing of the trial due to the lack of evidence of such training occurring in the UK. Database right infringement claims were also dropped for similar reasons... The UK judgment comes ahead of any decision in the parallel US proceedings (currently proceedings are before the Northern Court of California).
On secondary copyright infringement, the issue concerned whether, under sections 22 and 23 CDPA, Stability AI infringed Getty's copyright by importing an "article" into the UK, namely the Stable Diffusion AI model, "which is and which Stability knew or had reason to believe is an infringing copy of the Copyright works". The court assessed the interpretation of "article" and "infringing copy" and accepted that "article" can include intangible electronic items but ultimately held that to be an infringement, it must be or contain an "infringing copy" and that must be a copy amounting to a reproduction in accordance with section 17 CDPA. As Stable Diffusion does not store or reproduce any Getty copyright works, it cannot be classed as an infringing copy. As such, the secondary infringement claim was dismissed.
" 599. ... Taking the specific facts with which I am concerned, is an AI model which derives or results from a training process involving the exposure of model weights to infringing copies itself an infringing copy?
600. In my judgment, it is not. It is not enough, as it seems to me, that (in Getty Images' words) "the time of making of the copies of the Copyright Works coincides with the making of the Model" (emphasis added). While it is true that the model weights are altered during training by exposure to Copyright Works, by the end of that process the Model itself does not store any of those Copyright Works; the model weights are not themselves an infringing copy and they do not store an infringing copy. They are purely the product of the patterns and features which they have learnt over time during the training process. Getty Images' central submission that "as soon as it is made, the AI model is an infringing copy" is, accordingly, in my judgment, entirely misconceived. Unlike the RAM chip in Sony v Ball which became an infringing copy for a short time, in its final iteration Stable Diffusion does not store or reproduce any Copyright Works and nor has it ever done so. The fact that its development involved the reproduction of Copyright Works (through storing the images locally and in cloud computing resources and then exposing the model weights to those images) is of no relevance. Furthermore, that there is no requirement that an article which is an infringing copy must continue to retain a copy does not assist Getty Images, because it is implicit in the word "continue" that at some point the article has in fact contained an infringing copy. The model weights for each version of Stable Diffusion in their final iteration have never contained or stored an infringing copy."
On trade mark infringement, the court first required proof that UK users actually encountered AI‑generated "watermarks" bearing the Getty Images or iStock marks. Under section 10(1) of the Trade Marks Act, Getty Images succeeded in proving double identity infringement of the iStock watermarks for one image generated by v1.x when accessed via DreamStudio and/or the Developer Platform/API. Under section 10(2), the court found a likelihood of confusion on some v1.x generated iStock watermarks and v2.x generated Getty Images watermarks, however this was fact specific and could not be applied more generally to the AI machine's outputs. The trade mark findings were therefore specific and related to historic infringements. Getty's broader section 10(3) claim failed, as there was no evidence of a change in economic behaviour, and the fact that each watermark and image produced by Stable Diffusion AI would be different meant that there was no sustainable case on tarnishment. The passing off claim was not addressed, as Justice Joanna Smith stated that the claim did not add anything to her findings on the trade mark claims.
For further background on this case, see our blog post Navigating representative actions: takeaways from Getty Images v Stability AI.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
