Superintelligent Computers

Stephen DeAngelis

September 5, 2008

Since the beginning of the industrial age, the specter of machines becoming the masters of men has been a staple of science fiction. The success of Arnold Schwarzenegger’s “Terminator” films underscores the fact that this theme remains popular. A number of well-known and respected innovators have predicted that the day computers become smarter than humans is just around the corner (see, for example, my post entitled Looking towards the Future with Ray Kurzweil). This event horizon is often referred to as a singularity, a point after which things change so much that no credible predictions can be made about the future. One individual who has put a lot of thought about how such a singularity might affect the future is a science fiction writer and engineer named Vernor Vinge [“Technology That Outthinks Us: A Partner or a Master?” by John Tierney, New York Times, 25 August 2008].

“Dr. Vinge is a mathematician and computer scientist in San Diego whose science fiction has won five Hugo Awards and earned good reviews even from engineers analyzing its technical plausibility. He can write space operas with the best of them, but he also suspects that intergalactic sagas could become as obsolete as their human heroes. The problem is a concept described in Dr. Vinge’s seminal essay in 1993, ‘The Coming Technological Singularity,’ which predicted that computers would be so powerful by 2030 that a new form of superintellligence would emerge. Dr. Vinge compared that point in history to the singularity at the edge of a black hole: a boundary beyond which the old rules no longer applied, because post-human intelligence and technology would be as unknowable to us as our civilization is to a goldfish.”

The unknowable always raises a certain amount of apprehension in people. The IT singularity, however, seems to raise more than apprehension — it raises the fear of humans losing control of the future. Technologists, however, look to the future with hope not dread. But hope doesn’t sell books, so science fiction remains filled with doomsday scenarios where machines try to wipe out humans. Vinge is no exception.

“The Singularity is often called ‘the rapture of the nerds,’ but Dr. Vinge doesn’t anticipate immortal bliss. The computer scientist in him may revel in the technological marvels, but the novelist envisions catastrophes and worries about the fate of not-so-marvelous humans like Robert Gu, the protagonist of Dr. Vinge’s latest novel, ‘Rainbows End.’ Robert is an English professor and famous poet who succumbs to Alzheimer’s, languishing in a nursing home until 2025, when the Singularity seems near and technology is working wonders. He recovers most of his mental faculties; his 75-year-old body is rejuvenated; even his wrinkles vanish.”

This literary device — placing someone from the past into a future with which he is completely unfamiliar — permits Vinge to describe the future as he imagines it could unfold as the unwitting victim becomes familiar with his surroundings.

“Thanks to special contact lenses, computers in your clothes and locational sensors scattered everywhere you go, you see a constant stream of text and virtual sights overlaying the real world. As you chat with a distant friend’s quite lifelike image strolling at your side, you can adjust the scenery to your mutual taste — adding, say, medieval turrets to buildings — at the same time you’re each privately communicating with vast networks of humans and computers.”

Vinge made “Robert” an English professor so that readers would accept that he was smart even if he was technically challenged — like many of the “greatest generation” who refuse to buy or use computers.

“To Robert, a misanthrope who’d barely mastered e-mail in his earlier life, this networked world is a multitasking hell. He retreats to one of his old haunts, the Geisel Library, once the intellectual hub of the University of California, San Diego, but now so rarely visited that its paper books are about to be shredded to make room for a highbrow version of a virtual-reality theme park. At the library he finds a few other ‘medical retreads’ still reading books and using ancient machines like laptops. Calling themselves the Elder Cabal, they conspire to save the paper library while they’re trying to figure out what, if anything, their skills are good for anymore.”

This theme also touches a bit on the emotions created by globalization — think about the skilled manufacturing workers who have lost their jobs (either to outsourcing or automation) and who require retraining in order to find another well-paying job. This was a theme touched on in John McCain’s acceptance speech as he accepted the nomination to be his party’s candidate for president. He said:

“Opening new markets and preparing workers to compete in the world economy is essential to our future prosperity. I know some of you have been left behind in the changing economy, and it often sees that your government hasn’t even noticed. Government assistance for the unemployed workers was designed for the economy of the 1950s. That’s going to change on my watch. … We will prepare them for the jobs of day — of today. We will use our community colleges to help train people for new opportunities in their communities.”

For some workers that will be difficult. They will feel entirely inadequate for training and education they need to find jobs in a new field. They will field like the “Elder Cabal” in Vinge’s book. Tierney continues:

“Dr. Vinge, who is 63, can feel the elders’ pain, if only because his books are in that [library]. He took me up to the Elder Cabal’s meeting room in the library and talked about his own concerns about 2025 — like whether anyone will still be reading books, and whether networked knowledge will do to intellectuals what the Industrial Revolution did to the Luddite textile artisans. ‘These people in “Rainbows End” have the attention span of a butterfly,’ he said. ‘They’ll alight on a topic, use it in a particular way and then they’re on to something else. Right now people worry that we don’t have lifetime employment anymore. How extreme could that get? I could imagine a world where everything is piecework and the piece duration is less than a minute.’ It’s an unsettling vision, but Dr. Vinge classifies it as one of the least unpleasant scenarios for the future: intelligence amplification, or I.A., in which humans get steadily smarter by pooling their knowledge with one another and with computers, possibly even wiring the machines directly into their brains.”

Intelligence amplification represents a partnership between humans and computers, which is why Vinge sees it as the “least unpleasant scenario” for the future. Like many others, he worries about a conflict between men and machines.

“The alternative to I.A., he figures, could be the triumph of A.I. as artificial intelligence far surpasses the human variety. If that happens, Dr. Vinge says, the superintelligent machines will not content themselves with working for their human masters, nor will they remain securely confined in laboratories. As he wrote in his 1993 essay: ‘Imagine yourself confined to your house with only limited data access to the outside, to your masters. If those masters thought at a rate — say — one million times slower than you, there is little doubt that over a period of years (your time) you could come up wi
th “helpful advice” that would incidentally set you free.’ To avoid that scenario, Dr. Vinge has been urging his fellow humans to get smarter by collaborating with computers. At the conclusion of ‘Rainbows End,’ even the technophobic protagonist is in sync with his machines, and there are signs that the Singularity has arrived in the form of a superintelligent human-computer network.”

And what if the machines take over? Vinge speculates that “it’s possible that artificial post-humans would use us the way we’ve used oxen and donkeys. But he preferred to hope they would be more like environmentalists who wanted to protect weaker species, even if it was only out of self-interest.” That’s a pretty dark scenario and, if believed, would lead technologists to the point of developing fast, but not smart, computers. Most people believe that the march of technology can’t (and shouldn’t) be stopped. They still believe that technology will help us solve our most troubling challenges, create a better quality of life, and improve the lot of mankind not destroy it. Book sales, however, require a compelling storyline and there is nothing more compelling than fight to the death between men and machines — just look at the gross earnings of “The Matrix” and “Terminator” series. If you want to get a better look at the state of things (man vs. intelligent machines), look for re-run of the Discovery Channel series called NextWorld and the episode titled “Future Intelligence.” It provides a “glimpse at smart technology that will put android helpers in the home, network commuters and entire cities to the Web, and bring us entertainment systems that can virtually make dreams come true. Advances in artificial intelligence are creating machines with near human-like mental agility. Intelligence will be embedded everywhere — even in our clothing, thanks to smaller, more powerful computers.” The program will either scare you or assure about the future.