Robots to rule? No, at least not anytime soon

By Soren Anderson

 

7/12/2004

 

TACOMA (WASH.) NEWS TRIBUNE

 

Here they come again, mechanical men out to do us humans harm. This time it's in "I, Robot" - the Will Smith blockbuster that opened Friday - based loosely on the late Isaac Asimov's famed collection of sci-fi stories of the same title.

 

Once more, our technology rises up to bite us in the backside - with titanium teeth. Are these movies and others of their ilk - the "Westworlds" of the world and reaching far back to the silent era and 1926's "Metropolis" - showing us our future?

 

In other words: Are we as a race doomed to be tossed onto the scrap heap of history by humanoid thinking machines?

 

It depends on whom you ask.

 

Science-fiction author Robert J. Sawyer, who has written on the subject and once interviewed Asimov, says: "Absolutely."

 

The award-winning Canadian writer calls "The Terminator" "absolutely prophetic" and says that once robots achieve humanlike consciousness - a development he believes is inevitable - human beings will become an endangered species.

 

"The first thing that intelligent computers and robots are going to realize is that the single greatest threat to their survival is us," he said in a telephone interview.

 

Mankind's monkeying with atomic weapons poses an even greater threat to electronic brains than it does to organic ones because a nuclear blast can generate a so-called electromagnetic pulse (EMP), a powerful wave of energy that wipes out a computer's magnetic-based memory. That, he said, a conscious computer cannot allow.

 

Scientists who work in the field of robotics dismiss such fears as feverish fantasies.

 

Mechanical engineer Marcia O'Malley is working with NASA to develop Robonaut, a humanoid robot designed to work with astronauts in space, and she does not foresee a "Terminator"-style "rise of the machines."

 

"I'm not worried," she said with a laugh when asked if her creations and others like them will one day make human beings obsolete. "I'm not worried at all. Something like the 'Terminator,' that can go off and function autonomously . . . it's just not going to happen. There are too many problems that need to be solved."

 

One of the biggest and most basic problems is power. Chetan Kapoor, associate director and chief scientist at the Robotics Research Group at the University of Texas-Austin, said the kinds of industrial robots used in manufacturing plants can weigh anywhere from four to 10 times as much as the objects they're designed to lift. Such massive machines require a lot of power.

 

"A human can lift his own weight," Kapoor points out. "No robot can do that."

 

So there's a big problem with robobrawn. There are even bigger ones when it comes to robobrains.

 

Movies like "I, Robot," "A.I." and "The Terminator" all have as central characters robots who think and behave like human beings. Ian Walker, an expert in robot technology at Clemson University, calls such devices "the Holy Grail for robotics." And like the Grail, they exist at present only in the realm of myth.

 

"The key issue is independent intelligence," Walker said. "Some robots and many other machines like chess-playing machines seem really intelligent, but this is simply the sum of their programming. There is no innate intelligence independent of their programmers."

 

Among experts, "you can find one or two people who will cry from the rooftops that they've achieved machine intelligence, but I think if you interview enough of the community you will find that we are not in general convinced," Walker said.

 

Might such robots someday become a reality?

 

"I really doubt that you will see anything like that in the next 20 years," said Kapoor. What we're almost sure to see, he said, are what he calls "service robots," designed for specific tasks, such as vacuuming buildings or working in operating rooms.

 

Such machines already exist. You can buy a robovacuum cleaner called the Roomba for home use today. And Kapoor is working to develop robotic nurses to assist doctors during surgery. And they will become more sophisticated and pervasive in society in years to come, functioning as companions or helping out in the kitchen. However, our homes will have to be reconfigured to accommodate them.

 

But sci-fi writer Sawyer thinks roboticists miss a crucial point when they focus on how difficult it is for humans to develop robots capable of independent conscious thought. He believes such roboconsciousness will one day be developed by the robots themselves.

 

"There is nothing special or magical or metaphysical about human consciousness," he said. "Most scientists think it's just a function of having enough interconnections in the brain."

 

That's how human consciousness came to be, he said. "When the primate brain reached a certain level of sophistication, consciousness emerged. And we suspect that's exactly what is going to happen as computers get sufficiently complex. At some point, self-awareness is in the cards."