Afirst-of-its-kind trial is underway in the UK High Court between Getty Images and Stability AI, creator of Stable Diffusion. Here is what you need to know:
The Core Dispute
- Getty Images claims that Stability AI trained its AI on 12 million unlicensed copyrighted images, including ones with watermarks, violating copyright, database rights, and even trademark law.
- Stability AI counters that training occurred on US‑based servers and that only a tiny fraction of outputs resemble Getty’s originals.
Key Legal Questions
- Does training AI on copyrighted material count as lawful use?
- Are AI-generated images with Getty watermarks infringing trademarks?
- Do UK database rights apply to scraped metadata?
- Can overseas training trigger UK liability if output is used here?
- Is the current law equipped to oversee generative AI’s needs?
Why It Matters
For rights-holders: A Getty win could force AI creators to license content or compensate creators.
For AI developers: A win for Stability may legitimize training on public data without needing licenses.
For regulators: This case could be a catalyst for new rules or licensing systems for AI training.
VE Legal’s Perspective
This landmark case marks a pivotal moment in the balance between creativity, technology, and legal protection. It highlights the necessity for clear policies, innovative licensing models, and balanced safeguards as AI continues to reshape creative industries.
