Chesstories

Machine vs. Man
By Steven Levy

Newsweek Magazine
July 21, 2003

Checkmate

We are sharing our world with another species, one that gets smarter and more independent every year

Garry Kasparov's head is bowed, buried in his hands. Is he in despair, or just stealing a minute of rest in his relentless quest to regain the world championship, promote chess and represent humanity in the epic conflict between man and machine? HE PROFESSES the latter. But no one could blame the greatest grandmaster in history if he did succumb to bleakness. His own experiences indicate the end of the line for human mastery of the chessboard.

In the sport of brains, silicon rules. Still, Kasparov is preparing to throw himself into the breach once more. In November he will play his third computer opponent in a highly touted match. The first, of course, was IBM's Deep Blue, which in 1997 beat him in a battle that he insists to this day was unfairly stacked against him. Then, earlier this year, he fought to an unsatisfying draw against Deep Junior, programmed by two Israelis. Next up will be X3d Fritz, a world-class program modified to play in the third dimension, where his 3-D glasses will create the illusion that a virtual chessboard is floating between Kasparov and the screen. Kasparov believes that it's still possible to conceive of a human's winning a series of games against a top chess program - but the window is closing.In a few years, he says, even a single victory in a long series of games would be "the triumph of human genius".

Meanwhile, the Deep matches have already yielded one truth in the evolving tension between humans and machines. Our very humanity puts us at a profound competitive disadvantage. We got a whiff of this in the Deep Blue match. Kasparov was so rattled at IBM's tactics - essentially, the computer team played to win at all costs when Kasparov had been expecting a gentleman's game - that he spectacularly blew the last game and thus the match. But this past January's Deep Junior contest revealed the problem more clearly. Kasparov won the first game and cruised to a draw in the second. However, in game three, after starting strong he made a glaring mistake. And it suddenly became obvious that when computers and humans compete, it's really not the same game at all. Kasparov was devastated in a way that an unfeeling machine never would be. Worse, having yielded the advantage, he had no hope as he would have against a human that his well-programmed opponent might make its own mistake and let him back in the game. The realization paralyzed even the great Kasparov, and it haunted him for the rest of the match.