AI, Patent and the threshold of Human Invention- A Legal Analysis

Abstract

The convergence of AI technologies and IP law raises complex questions, most notably about inventorship. The question of the moment is whether AI systems, capable of producing entirely new phenomena and outputs which would qualify as inventions by tradition, can be considered as inventors under current patent law regimes. The dominant legal models, which were carefully designed for a world where only human beings were such a source of innovation, may not be fit-for-purpose when it comes to the changes in the landscape precipitated by AI-facilitated technological innovation. This calls for an in-depth reflection on the legal, ethical and policy issues involved in determining if and how AI-generated inventions should be treated. Under the U.S. patent system, the United States Patent and Trademark Office (USPTO) requires that only a natural person can be an inventor on a patent application, a position that has the potential to disrupt the world's balance between economic incentives and societal costs if inventing-AI is recognized as an inventor.