IS ARTIFICIAL INTELLIGENCE PATENT-DEFEATING? THE IMPACT OF ARTIFICIAL INTELLIGENCE ON THE STATE OF THE ART

Abstract

The impact of artificial intelligence (AI) on the field of intellectual property (IP) law is currently one of the most frequently discussed topics among legal academics. However, one aspect that has received considerably less attention is the impact of AI on the state of the art, specifically in the context of the novelty and sufficiency of disclosure requirements. This impact is potentially wide-ranging and diverse. It is consequently an aspect that merits discussion, given that the technological development of AI could alter today’s prevailing concept of the state of the art and the criteria for patentability. This dissertation discusses whether prior art that is accessible and understandable to an AI agent, but not to a human, should be considered as part of the state of the art for the purposes of assessing novelty. As such, the question is raised of whether this prior art defeats patent. Moreover, an attempt is made to ascertain the extent to which inventive AI agents are capable of sufficient disclosure to meet the policy objective of this legal requirement. Ultimately, it could mean an invention becoming part of the state of the art in exchange for limited monopoly on its use.