The United Kingdom Supreme Court has upheld the decision that artificial intelligence cannot be granted patents, as the law only acknowledges humans and companies as eligible inventors. In a unanimous ruling, the court rejected the application of Stephen Thaler, the creator of the AI system named DABUS, to have his AI recognized as the inventor of new products.
Legislative Constraints on AI Recognition
According to the court's judgment, delivered by Judge David Kitchin, current legislation does not recognize machines as possible creators. Judge Kitchin emphasized that the case before the court involved what are assumed to be original and innovative ideas for devices and methodologies conceived by the AI. However, without legal provisions designating machines as inventors, the plea for Thaler's AI system to hold a patent cannot be entertained.
Stephen Thaler sought to patent two creations attributed to DABUS – a novel food container and an emergency light beacon. Initially, in 2019, the UK Intellectual Property Office (IPO) had dismissed these requests, citing the lack of legal framework for non-human inventors. The IPO further commented that such cases raise important questions about handling intellectual property generated by AI.
Global Perspective on AI Generated Material
This decision in the UK mirrors a similar legal outcome encountered by Thaler in the United States, where the U.S. Patent and Trademark Office also denied his claim to name AI as an inventor. The United States Supreme Court chose not to review the case. The US Patent and Trademark Office (USPTO) rejected the applications on the grounds that DABUS did not qualify as an inventor under the US patent law, which defines an inventor as an “individual”. Thaler appealed the decision to the US Court of Appeals for the Federal Circuit, which affirmed the USPTO's rejection in July 2021.
The ramifications of the rulings are significant as they create precedents for other jurisdictions worldwide. Countries and lawmakers are increasingly faced with the challenge of determining if, and how, creations facilitated by or directly generated by AI systems should be protected under intellectual property laws.
