AI, Copyright and the Courts: What the Getty Images Ruling Means

On 4 November 2025, the UK High Court handed down its judgment in the dispute between Getty Images and Stability AI. In short, Getty alleged that Stability trained its popular image‑generator model using millions of images from Getty’s archive without permission. It alleged that some of the AI‑generated outputs even bore Getty or ‘iStock’ watermarks. The full set of claims included copyright infringement, database rights, trademark infringement as well as passing‑off.

In the end, the Court found that while some watermark‑style use of Getty marks by the AI model did amount to infringement under trade‑mark law, the core copyright and database‑right claims largely failed. In fact, Getty dropped its main copyright claims during the trial, recognising the difficulties of proving that training had taken place in the UK. This was a key jurisdictional hurdle. The court held that the AI model did not store or reproduce Getty’s works in a way that created direct copyright infringement under UK law.

The decision provides important pointers about the limits of IP enforcement in the context of generative AI. However it does not give a broad new right for rights‑holders to stop AI training across the board.

What does this mean for UK businesses and brands?

The case also underscores that, even when copyright claims may be challenging to enforce, brand and trade mark risks remain very real. Logos, watermarks and other brand imagery can easily appear in AI-generated content, creating potential infringement and ownership issues if such outputs echo your marks.

Moreover, the dispute reveals a persistent gap in legal certainty. As numerous UK legal commentaries note, the question of whether training on copyrighted material constitutes infringement remains unresolved in many contexts. This ambiguity leaves businesses operating in a grey area.

From a strategic perspective, companies using AI tools to create visual, written or audio content must proactively map their risk. This also includes the sharing of brand assets with third parties, such as vendors or partners. It is crucial to ask questions regarding, who trained the model? On what data? And where will the output be used? Addressing these issues upfront can help mitigate risks.

Your business and how we can help

Your brand is one of your most valuable business assets. If someone uses your name or logo without permission, or if your brand appears incorrectly in AI-generated content, it can lead to confusion and damage your reputation. It can even create legal or commercial risks. Monitoring and control are essential. If you fail to track how your marks appear online or on re-produced material you may lose the ability to manage or enforce your brand effectively. Make sure to set contractual limits on their use. Unchecked use can lead to reputational harm and weaken your legal position in protecting trademarks.

At Birdi & Co, we support growing businesses in navigating a host of these commercial challenges. Specifically, we can help assess your exposure in relation to data protection, IP infringement and use of your commercial assets when partnering with other entities.