Researchers at Tisch University have created an AI chess engine that isn’t trying to absolutely crush its human opponent.
Since IBM’s Deep Blue beat its human opponent in 1997, Artificial Intelligence has been dominating humans when it comes to chess. These engines are so smart no human has won a tournament against them since.
“Chess sits alongside virtuosic musical instrument playing and mathematical achievement as something humans study their whole lives and get really good at. And yet in chess, computers are in every possible sense better than we are at this point,” said team member Jon Kleinberg in a statement. “So chess becomes a place where we can try understanding human skill through the lens of super-intelligent AI.”
These researchers wanted to create an AI that doesn’t destroy the soul of the person it’s playing against and rather puts the enjoyment back in facing off against a computer.
The AI has been trained to model human chess moves rather than aiming to win the game overall. This meant it mimics human behaviour and therefore can be adjusted to different skill levels. The researchers have called the engine Maia and it’s been released on the free online chess server lichess.org, where it was played more than 40,000 times in its first week.
“Maia has algorithmically characterized which mistakes are typical of which levels, and therefore which mistakes people should work on and which mistakes they probably shouldn’t, because they are still too difficult,” said co-author Ashton Anderson, assistant professor at the University of Toronto.