The United Kingdom’s High Court of Justice issued a mixed ruling on Tuesday in the Getty Images v. Stability AI intellectual property case filed in 2023. The decision mostly favored Stability AI but left key questions unanswered regarding the use of copyrighted material by artificial intelligence.
Getty Images, which owns a vast library of copyrighted online stock images licensed to users for a fee, alleged that Stability AI’s Stable Diffusion model infringed upon its trademark and copyrighted material. The company contended that Stable Diffusion, trained using online material, violated its rights.
Specifically, Getty argued that Stability’s Stable Diffusion AI model infringed on its trademark by reproducing its watermark in certain instances. However, Justice Joanna Smith ruled that these findings were “extremely limited in scope.”
Getty failed to demonstrate that any UK users utilized Stable Diffusion to reproduce the watermark, a requirement under UK law to establish “primary infringement,” the judge ruled.
Furthermore, Justice Smith dismissed the “secondary infringement” allegation. The court found that the AI model does not store or reproduce the images, thus failing to meet the criteria for a violation under the UK’s Copyright, Designs and Patents Act (CDPA) of 1988. Justice Smith wrote:
“Although an ‘article’ may be an intangible object for the purposes of the CDPA, an AI model such as Stable Diffusion, which does not store or reproduce any Copyright Works, and has never done so, is not an ‘infringing copy’ such that there is no infringement under sections 22 and 23 CDPA.”
This ruling allows brands to protect their trademarks from AI reproduction. However, the specific technicalities of this case prevent a broad legal precedent from being established, leaving significant questions about AI training and intellectual property rights open for ongoing debate.
Previous Rulings and Web3 Solutions for Content Creators
In a similar development, a US judge, William Orrick, issued a ruling in October 2023 that dismissed most copyright infringement claims against Midjourney AI, DeviantArt, and Stability AI. Orrick stated that images generated by AI models do not constitute copyright infringement because they do not bear a resemblance to the artists’ original work used for training the models.
The current lack of robust legal protections for content creators and artists has motivated several blockchain and Web3 companies to develop data provenance solutions. These solutions aim to record ownership and verify the sources of information, copyrighted material, and other intellectual property.
Notable among these solutions are non-fungible tokens (NFTs), which can be employed to track original ownership and assign royalty rights for various creative works, including art, essays, books, and musical productions.