Ever-Smarter Machines Raise Some Old Questions

John Henry told his captain

A man ain't nothin' but a man,

And before I'd let your steam drill beat me down

I'd die with this hammer in my hand.

I'd die with my hammer in my hand.

EVER since John Henry, the legendary black railroad worker, bet he could keep up with a newly invented steam-hammer, Americans have viewed technology with mixed feelings.

They cheer new inventions that make life easier; but they worry that machines are slowly taking over. Last Thursday, a machine chalked up another small victory in technology's march. A computer program won the world checkers championship.

There's still a chance that a human will regain the checker crown. The program, Chinook, bested the world's second-best player after the acknowledged No. 1 player, Marion Tinsley, withdrew because of illness. But many technology observers say that it's only a matter of time before humans build a machine that, arguably, could ``think'' in a limited way.

``Many people will find it disconcerting,'' says Daniel Kevles, professor and head of the Science, Ethics, and Public Policy program at the California Institute of Technology. But ``sooner or later human beings will be able to construct some kind of approximation of at least a primitive brain, where the machine will have some kind of decision-making powers and be conceptually vital.''

Already, machines ``have relieved the burdens on people's backs,'' says Melvin Kranzberg, professor emeritus at the Georgia Institute of Technology. ``Now, they're relieving the burdens on people's minds.''

In one sense, the predicted coming of an intelligent machine is nothing new. During the Industrial Revolution, and especially in the 19th century, machines came to dominate everything from transportation to farming. But in another sense, the phenomenon is quite new.

``The difference is: It's not body-oriented this time,'' says Mel Seesholtz, a faculty member in the Science, Technology, and Society program at Penn State University's Ogontz campus. ``It's mind-oriented. The mind is the last great taboo.''

As a matter of fact, building a thinking machine has turned out to be far more complicated than artificial-intelligence experts first thought. Computers are still quite dumb compared with the cognitive abilities of, say, a two-year-old. Nevertheless, the appearance of checker- and chess-playing computers that beat highly rated players suggests that machines will one day ``out-think'' people, at least in certain limited ways.

``There can't be anything supernatural about the brain,'' Professor Kevles says.

Machines that out-think people - long a staple of science fiction - intrigue some people and scare others. When Professor Seesholtz toured the world in the late 1980s demonstrating virtual-reality technology, he found a definite split in his audiences. ``Literally half of them were terrified and the other half couldn't wait,'' he says. ``And there really was no in-between.''

Technology has had this divisive effect historically. ``It has enormously increased human productivity and saved man from a lifetime of work,'' Professor Kranzberg says. Until the Industrial Revolution, ``people worked from dawn to dusk six days a week just in the hope that they had enough to eat until harvest came.'' But technology also threw people out of work, created grimy industrial cities, and appalling conditions for the urban working class. In the 19th century, some workers revolted and smashed the new machines.

Technology historians and culture-observers are split on whether a limited ``thinking'' machine, like world-beating checker or chess software, will stir an unusually strong backlash. Kevles doesn't think it will.

``Just because it can solve a chess program doesn't mean it's going to interfere in our lives in any way,'' he says. ``There are so many things still that we can do better than machines, and most of the things that we rely upon in our lives that give us a sense of dignity and self-esteem, machines by and large have not been able to express.''

Seesholtz is more worried: ``When machines become that threatening - and they will be seen as threatening - Americans will do what they always do. They'll pass laws against it. And it will go underground. Once it goes underground, we lose control of it.'' He sees the coming of a drug-culture type of underground that will use mind-altering machines instead of chemicals to let humans get ``high.''

There are no easy answers. John Henry won his bet but lost his life driving a steel spike into solid rock as fast as the automatic steam hammer.

``Don't forget the machines are made by us,'' Kranzberg says. ``Machines are made by human beings, human imagination, human skill. We made up the game of chess to begin with. We're always posing new questions for ourselves.''