To Err is Human: Is This Why? And What's to be Learned? 5-18-97

In February 1996, Garry Kasparov, the world chess champion, roundly defeated IBM's Deep Blue chess machine. Almost immediately, the IBM team went to work redesigning hardware, doubling the computer's calculating power, taking it to "chess school" and teaching it tricks. In March of 1996, the Deep Blue project manager made a list of ways in which the computer could be improved. They knew they had to make the computer faster, so they concentrated on making it smarter, trying and discarding as they went. Each team member had a specialty, bringing their particular expertise and forte to the project.

Last Sunday, the Deep Blue IBM computer overwhelmed Garry Kasparov and won the six-game chess match between man and machine. The final score was 3 1/2 points for the computer and 2 1/2 points for Kasparov. Visibly upset at the game's end, Kasparov bolted from the table, shrugging his shoulders. In an interview this past Friday, Kasparov said, "It's very difficult to analyze the results of the match. I know what I did wrong. But I don't know what the computer did wrong or right. It's a mystery."

Although Deep Blue has the awesome calculating ability to examine 200 million moves per second, what irked Kasparaov was that the computer seemed to understand the complexities of chess. "I met something I couldn't understand. It's far beyond anyone's understanding". As Kasparov prepared openings that he figured the computer would not anticipate, it resulted in Kasparov's being forced to play against his own better instincts, which he said in an interview was a mistake.

Judith Polgar, the world's top female chess player, said Kasparov lost his one-way, psychological battle with Deep Blue. After Saturday's draw, Polgar said, "I think he (Kasparov) is feeling a bit nervous." Kasparov himself said, "I'm not afraid to admit that I'm afraid. It definitely goes beyond any chess computer in the world." What happened is that the stress of the match got to Kasparov. He gave his silicon opponent human characteristics, saying he thought he detected signs of human intelligence from Deep Blue during Game 2.

His error proves that computers can only win when humans make mistakes. As Frank Rich said in today's New York Times, "Deep Blue can't write a sonnet. Deep Blue can't hug a baby. Deep Blue has no sense of humor." Computers only do what we make them do. The computer was programmed by a brilliant team who pulled together to do what most thought impossible. When Kasparov allowed fear and intimidation to overcome him, he became unduly stressed and blocked out the most precious gift God gave us as humans, the ability to pay attention to our gut instincts, our intuition, which almost always keeps us on the right track. Is there a lesson in there for us? I think so. Will we pay attention to it? I hope so.