Getty Images CEO Craig Peters says Getty is spending “millions and millions” on its Stability AI lawsuit in the UK and the US, accusing it of copying 12M images
Peters told CNBC in an interview that both Stability AI — the U.K.-based startup best known for its text … Bluesky: @neilturkewitz Bluesky: Neil Turkewitz / @neilturkewitz : “Part of the reason Getty Images is pursuing legal action against Stability AI & not other firms is because such legal pursuits are “extraordinarily expensive. Even for a company like Getty Images, we can't pursue all the infringements that happen in one week. www.cnbc.com/2025/05/28/g...
Context & Ripple Effects
Getty’s legal campaign began with UK plans and a US complaint alleging that Stability AI used more than 12 million images for model training. The reported spend shows how the dispute has become a sustained enforcement effort rather than a one-off claim.
The UK case subsequently reached the High Court in a trial over alleged image scraping, alongside the earlier US training-data complaint. Getty’s willingness to fund parallel cases makes the economics of enforcing image rights central to the story.
First-order effects
- Getty absorbs substantial legal costs to pursue its claims in both jurisdictions, while Stability AI must continue defending allegations over the provenance of Stable Diffusion’s training data.
- The expense constrains Getty’s ability to litigate every alleged misuse, concentrating enforcement on cases it considers strategically important.
Second-order effects
- Other image-rights holders may face the same enforcement constraint: even potentially consequential claims require enough scale and resources to justify cross-border litigation.
- AI image-model developers face added pressure to document training-data sources and assess legal exposure, particularly where commercial image libraries are involved.
Third-order effects
- If costly, selective litigation remains the main route to resolving training-data disputes, enforcement may favor large rights owners and leave many smaller creators without a practical remedy.
- The cases are part of a broader test of whether AI-content commercialization will depend on negotiated licensing or be shaped case by case through courts.
The trend: Generative-AI image markets are moving toward a more contested licensing environment in which the cost of proving and defending training-data rights matters as much as the underlying claims.