When artificial intelligence attacks!

by Erik J. Martin

As the old adage goes, a little knowledge can be a very dangerous thing.

Science fiction writers like Arthur C. Clarke, who made the rebellious robot "Hal" a household world in his book "The Sentinel" (and Stanley Kubrick’s subsequent film “2001: A Space Odyssey”), have popularized the notion that computers with too much autonomous intelligence turn evil, inevitably betraying and resisting their human masters. 

Megalomaniacal computers have run amuck in movies like “The Matrix” and “The Terminator” and on television on “Star Trek,” “Doctor Who” and “The X-Files."

Now there's a new Bruce Willis flick titled "Surrogates," slated to open this weekend, that depicts a future world in which (according to the IMDB.com) "People are living their lives remotely from the safety of their own homes via robotic surrogates -- sexy, physically perfect mechanical representations of themselves." 

Indeed, if science-fiction entertainment has taught us anything, it's that the smarter a machine becomes, the more selfish and, ultimately, evil it grows, as well. So how realistic is this possibility?

"Artificially intelligent software systems don't pose any greater risk of harming humans than standard software systems, which currently control power plants, airplanes, and traffic systems. But we must take care to engineer and deploy all software systems properly," Dr. Marcus A. Maloof, PhD, associate professor, Department of Computer Science, Georgetown University, Washington D.C., told me when I interviewed him a few years back.

In his book “Mind Children: The Future of Robot and Human Intelligence,” Hans Moravec predicted that with the rate that computational power is decreasing in cost and therefore the rate at which computing power will be available, artificial intelligence could reach human equivalence approximately 30 years from now.

Nevertheless, when I asked Dr. Thomas Whalen, a trained experimental psychologist and natural language interface scientist at the Communications Research Centre in Ottawa, Ontario, Canada, he offered these reassuring words: "I wouldn't expect the possibility of arti­ficially intelligent computers turning on humans for centuries. Why is there the assumption that humans and robots would compete for the same resources? One must also consider the issue of whether being smarter is more adaptive in Darwinian terms. There are organisms like insects which are increasing in numbers as fast or faster than human­kind. Insects can coexist with people, can't they? Computer brains are simply not pow­erful enough to turn on human beings."

They certainly are, however, in Tom Maddox's universe. Maddox is a former writer on “The X-Files” TV series and author of the science-fiction novel “Halo,” in which a benevolent com­puter system named Adelph acquires consciousness via interaction with humans. Though Maddox acknowledged that the malevolent-computer theme has grown a bit tired over the years, he thinks it is still a powerful means of expressing one of sci­ence fiction's ccntral themes.

"One of the main things science fiction does is express our anxieties about change," Maddox said. "Anything genuinely different is viewed as negative. And most people have superstitious feelings about computers, which are treated as magic machines on television and in movies. Some sci-fi writ­ers aren't interested in writing about AI because they look at it as too magical. When the artificial intelligence is allowed to turn into a god, it becomes uninterest­ing. In reality, I personally don't see com­puters possessing artificial intelligence turning on humans because they're not competing with us for resources. It doesn't need to steal your land or your girlfriend."

So don’t get too freaked out the next time you watch Hal up to his hijinks in “2001: A Space Odyssey” or cringe when Neo gets roughed up by the Smiths in “The Matrix” trilogy. It’s all in good… wait a minute, my computer is n ot l e TTtingggg me f in iShhhhhhh my sEntnce – HELP! THE PCs ARE TAKING OVER! TELL EVERY ONE Y
 OU KNOW! TELL MY WIFE I LOVE HER AND