Illustration for: Stability AI beats Getty in UK High Court, leaving AI copyright unclear
Policy & Regulation

Stability AI beats Getty in UK High Court, leaving AI copyright unclear

2 min read

When Stability AI walked out of the courtroom with a win against Getty Images, it set off another round of chatter about who really owns the stuff AI spits out. The judges, though, stopped short of answering the big question that’s been pulling tech firms, artists and publishers in opposite directions. The suit was filed back in 2023, making it the first high-profile AI copyright case to land in England’s High Court, and lawyers and developers have already started picking it apart.

Even with the headline-grabbing decision, many people are still not sure whether it actually shifts the legal landscape for machine-generated works. Because the ruling didn’t settle the underlying copyright debate, AI startups and content creators alike seem stuck in a grey zone. Below is a quick look at what the court actually said, and why it could matter:

*UK High Court didn't weigh in on the key copyright issue dividing the tech sector and creative industries. The case, first filed in 2023, is the first major AI copyright claim to reach England's High Court, though the verdict offers little clarity to other AI companies and rightsholders. Getty had o*

UK High Court didn't weigh in on the key copyright issue dividing the tech sector and creative industries. The case, first filed in 2023, is the first major AI copyright claim to reach England's High Court, though the verdict offers little clarity to other AI companies and rightsholders. Getty had originally pursued the core issue of training on copyrighted material but dropped it mid-trial, largely due to weak evidence. Getty, which has a large archive of images and video, sued Stability in 2023 for "unlawfully" scraping millions of images to train its software.

Related Topics: #Stability AI #Getty Images #UK High Court #AI copyright #generative models #machine‑generated works #2023 #scraping

Stability AI left the courtroom with a win, but the judgment says very little about the core question - whether AI models have to get permission to train on copyrighted works. In other words, there’s no clear answer yet. The High Court sidestepped that issue, so both tech firms and creative companies are still in the dark.

Because this is the first major AI copyright claim to reach England’s High Court, the result feels more like a procedural victory than a substantive ruling. Getty Images’ challenge was mostly rejected, yet the legal reasoning stops short of setting a precedent that could guide future disputes. What does that mean for AI training going forward?

It probably means uncertainty will linger over how UK copyright law applies to machine-learning data. Stakeholders may have to wait for more litigation or for Parliament to act before a solid framework appears. Both sides have said they’ll watch any appeal closely.

Until then, the balance between protecting creators and letting AI develop remains unsettled, and the wider impact on the industry is still unclear.

Further Reading

Common Questions Answered

What was the outcome of Stability AI's case against Getty Images in the UK High Court?

Stability AI emerged victorious, as the court dismissed Getty Images' claims, granting Stability AI a procedural win. However, the judgment did not resolve the underlying copyright question about training AI on protected works.

Why did Getty Images drop its core issue about training on copyrighted material during the trial?

Getty Images abandoned the central argument concerning the legality of training AI models on copyrighted works because the evidence supporting its claim was deemed weak. This strategic retreat left the court without a substantive examination of that key issue.

How does the High Court's decision affect future AI companies regarding permission to train on copyrighted works?

The decision provides little guidance for AI developers, as the court deliberately avoided ruling on whether permission is required to train models with copyrighted content. Consequently, both tech firms and creative rightsholders remain uncertain about their legal obligations.

In what way is this case described as a procedural victory rather than a substantive ruling?

The case is labeled a procedural victory because Stability AI won on technical grounds—Getty's claims were dismissed—while the court refrained from addressing the core copyright question. This means the outcome does not set a legal precedent on AI training practices.