 |
|
Digital Soul: Intelligent Machines and Human Values by Thomas M. Georges. Westview. 2003. 288 pages. $26. Check price/buy book.
"I Compute, Therefore I Am": Tomorrow’s Self-Aware Machines
As machines start to simulate human qualities, we need to think seriously about the potential impacts.
Reviewed by Lane JenningsAside from being unable to feel emotions, Commander Data, the android character on Star Trek: The Next Generation, is physically and mentally superior to every other member of the crew aboard the starship Enterprise. So why not mutiny and take over command? This certainly could happen here on Earth if truly intelligent machines—whether or not they resemble humans in form—begin to comprehend how weak and self-deluded human beings are.
Machine mutiny is just one of several possible futures for human/machine relations outlined by Thomas Georges, former research scientist at the National Bureau of Standards, in his book Digital Soul.
Throughout his book Georges draws heavily on the work of other writers. He has gathered ideas from MIT computer guru Marvin Minsky, popular astronomer Carl Sagan, Douglas Hofstadter (author of the mind-stretching best seller Gödel, Escher, Bach), and other deep thinkers from Freud and William James to Alan Turing and B.F. Skinner. Weaving elements from these many sources together, Georges argues that every trait we call "human"—;intelligence, emotions, moral values, and even self-awareness—really result from physical processes that can in theory (and increasingly in practice) be mimicked, duplicated, or surpassed by nonbiological systems—i.e., machines.
In effect, Georges regards humans and all living creatures as subsets of the general class "machine." Some machines, designed by evolution using sequences of genes and modified by experience and chance events, exist in a form made up largely of water and carbon compounds. Others, designed by humans using physical and electronic tools, contain mostly silicon and metal. But once either type attains a sufficient degree of complexity, Georges believes, it deserves to be called "intelligent." Moreover, he would add, if a machine displays what look convincingly like feelings, then it is only reasonable to accept them as such.
"Is there some deep difference," Georges asks, "between emulating emotions and actually having them?" Later he answers his own question: "We attribute emotions to other people based solely on their observed behaviors. It seems reasonable, then, to apply the same criterion to other animals and machines."
In similarly glib fashion, Georges deplores the flawed logic, inconsistencies, counterproductive impacts, and unprovable assumptions that underlie human laws, which are focused overwhelmingly on punishment, not "justice"; organized religions, which generally delude, divide, and discourage free thinking; and moral codes proclaimed, with no convincing proof, to be eternal, absolute, and universal.
Readers willing to accept the author’s general proposition that nothing exists unless it can be physically identified and measured will find plenty of quotable phrases—some original, some borrowed—to use in arguments with "Vitalists" and others who prefer to live with some degree of ongoing uncertainty, in the faith that all we know is less than what may be.
Ultimately, this book is less valuable for the answers it provides than for the questions it dares to pose. Scientists, politicians, clergy, and consumers all need to seriously consider the mounting evidence that networks of computers and automated machines will soon—or have already—become too complex for human minds to fully comprehend and too essential for humans to willingly shut down. The recent power blackout in North America illustrates the problem all too well.
Georges sensibly proposes that, rather than react to our situation with fear—for instance, attempting to restrain research in areas such as artificial intelligence, nanotechnology, or robotics—we should pay more attention to the ways individuals and institutions now use and misuse machines. By accepting that human minds may not be essentially different from machines, we can more likely employ existing technologies beneficially. At the same time, we could work toward a possible future in which machine intelligence is successfully teamed with that of humans to achieve a sustainable partnership instead of perpetuating a master-slave society, where machines serve without understanding—or rule without compassion.
Popular writing on complex issues always risks seeming too superficial. Georges doesn’t escape this trap, but by quoting various distinguished writers as the starting point for his own extended arguments, he gives readers the chance to pursue any topic further. All in all, this book is a good choice to read and discuss with others—face-to-face or online. You might even want to share it with your best machine friend.
About the Reviewer
Lane Jennings is research director of THE FUTURIST and production editor of Future Survey.
[Reviewed in THE FUTURIST, November-December 2003]return to top
COPYRIGHT © 2003 WORLD FUTURE SOCIETY
7910 Woodmont Avenue, Suite 450, Bethesda, Maryland 20814.
Tel. 301-656-8274. E-mail info@wfs.org. Web site http://www.wfs.org.
All rights reserved.