In 1912, the Spaniard Leonardo Torres y Quevedo built what was probably the first functioning chess automaton, which was then presented at the 1914 World’s Fair in Paris. El autómata ajedrecista is historically the first automaton that could play real parts of a chess game. Strictly speaking, the automaton could not play a complete game of chess. The chess automaton was able to automatically play a white king (initially on a8) and a white rook (initially on b7) against the lone black king located on any square. The algorithm was suboptimal, but could win in less than 50 moves against any defense. Although it sometimes failed to meet the 50-move rule (which states that a game can be drawn if each player has made the last fifty moves in a row without moving a pawn or capturing a piece), it always ended up giving checkmate by following the algorithm by which it was programmed. It used mechanical arms to make its moves and electrical sensors to detect its opponent’s responses. A second, mechanical but not algorithmic, improved El Ajedrecista was built in 1922 by Leonardo Torres Quevedo’s son Gonzalo under his father’s supervision. At the 1951 Paris Cybernetics Congress, the improved machine was presented to a larger audience and explained to Norbert Wiener. Even though the automaton could only play king/rook vs king, El Ajedrecista can be considered the world’s first chess computer, even a special robot that can move its own pieces. It is still functional and can be seen at the Torres Quevedo Museum of Engineering, Institute of Civil Engineering at the Universidad Politécnica de Madrid.
