"Claude Shannon was known as a strong chess player, and he was the first to not only propose that a computer could play Chess, but to also provide details on how that could be done. In his paper Programming a Computer for Playing Chess published in 1950, he describes how a computer could play chess using the minimax algorithm and a predesigned evaluation function. In so doing, he laid the foundation for virtually all abstract computer game programs used today."
But according to chronology in the Computers and Chess Wikipedia article, in 1948 "Norbert Wiener's book Cybernetics describes how a chess program could be developed using a depth-limited minimax search with an evaluation function."
Even before that, in 1941, "Predating comparable work by at least a decade, Konrad Zuse develops computer chess algorithms in his Plankalkül programming formalism. Because of the circumstances of the Second World War, however, they were not published, and did not come to light, until the 1970s."
 - https://en.wikipedia.org/wiki/Computers_and_chess
If I remember correctly, this relay based machine mates with king and rook versus bare king (significantly harder than king and queen versus bare king)
Another great book that includes information on Shannon and provides context to his work at Bell Labs is Gertner's Idea Factory.
1: https://www.macwelt.de/news/Vint-Cerf-beschwerte-sich-bei-Co... (run your adblocker).
 "A Man in a Hurry: Claude Shannon’s New York Years" [ http://spectrum.ieee.org/geek-life/history/a-man-in-a-hurry-... ]