AI-Assisted Inventions in the Field of Drug Discovery: Readjusting the Inventive Step Analysis

Abstract

Artificial intelligence (‘AI’) is increasingly applied at all stages of drug discovery. While AI has the potential to boost innovation, it also raises many important ethical, social, political and legal issues. Among the latter are the challenges that AI poses for the patent system. With the rapid evolution of AI technologies and the increase in their computational power, the process of inventing has undergone substantial changes. As AI significantly expands human capabilities, inventions that were previously the result of human ingenuity, perseverance or serendipity can now be achieved by routine experimentations with the use of AI. This article argues, therefore, that the patent law approaches that were developed to assess human-generated inventions are not suitable for AI-assisted inventions and require urgent reconsideration. It will explain that the proper test for the obviousness assessment needs to take into account the advancement of AI technology and will provide suggestions on how the analysis can be modified.

Keywords: artificial intelligence, inventive step, obviousness, patents, drug discovery, access to medicines