Google Research

A New Paradigm: Chess AI In Game Analysis

Google (2022)

Abstract

In 1997, a chess artificial intelligence, IBM’s DeepBlue, defeated the chess world champion, Garry Kasparov. Since then, chess artificial intelligence has only increased in skill, achieving chess ratings (measures that evaluate player skill) that beat those of top skill by several standard deviations. Seemingly, chess has been “solved.” Yet, the chess community continues to grow in terms of population, interest, and theory. Instead of retreating into existential defeat, chess players embrace chess AI as a critical part of their chess experience. One of the manifestations of this change is game analysis boards. This paper examines the on-going interaction between chess AI and chess players, including the utilization of AI in game analysis, game preparation, and cheating.

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work