Deep Blue Defeats Kasparov: New York, 1997
1997 · ai
On 11 May 1997, in a television studio in Manhattan, the world chess champion Garry Kasparov resigned the sixth and final game of a match after nineteen moves. The score was three and a half to two and a half, and his opponent was a computer. Deep Blue, built by a team at IBM, became the first machine to defeat a reigning world champion in a full match under standard tournament conditions.
Kasparov had beaten an earlier version of the same machine in 1996, and the rematch was played on a stage with cameras, commentary and a global audience. Deep Blue was a cabinet of specialised processors capable of examining something like two hundred million positions every second, guided by an evaluation function that had been shaped with the help of grandmasters. It did not think in any sense a chess player would recognise, and its strength came from the sheer volume of calculation combined with human judgement embedded in its scoring.
The match left a dispute behind it. A move in the second game looked unusually subtle to Kasparov, who suspected human intervention, and IBM declined to release the logs at the time or to play again, dismantling the machine soon afterwards. The logs were published later, and the move is now generally understood to have come from the program.
The result was read as a turning point and it was, though not in the way the headlines suggested. Chess did not die. Programs strong enough to beat any human now run on ordinary telephones, players train with them, and the game is more popular than it has ever been. What ended in 1997 was the assumption that certain kinds of thinking would remain beyond machines, an assumption that has been retreating ever since.