The ancient oriental board game Go is a thousand years older than chess, and is the focus of a number of artificial intellignece programmers for a number of reasons: it has an immense number of potential moves and outcomes; it is much harder to assess the outcome of a game from observing the state of the board halfway through; and a computer has never beaten a human master of the game. But they’re getting better all the time – especially a machine known as Crazy Stone that uses ‘Monte Carlo method’ algorithms to learn the most effective moves.