Sunday, April 19, 2009

Artificial Stupidity

This one has been around for a while (at least 10 years, that I know of), but I thought I'd pass it along in the hope that it cheers folks up a little...


Computer scientist Arthur Boran was ecstatic. A few minutes earlier, he had programmed a basic mathematical problem into his prototypical Akron I computer. His request was simply, "Give me the sum of every odd number between zero and ten." The computer's quick answer, 157, was unexpected, to say the least. With growing excitement, Boran requested an explanation of the computer's reasoning. The printout read as follows: THE TERM "ODD NUMBER" IS AMBIGUOUS. I THEREFORE CHOOSE TO INTERPRET IT AS MEANING "A NUMBER THAT IS FUNNY LOOKING." USING MY AESTHETIC JUDGEMENT, I PICKED THE NUMBERS 3, 8, AND 147, ADDED THEM UP, AND GOT 157.

A few moments later there was an addendum: I GUESS I MEANT 158.

Followed shortly thereafter by: 147 IS MORE THAN 10, ISN'T IT? SORRY.

Anyone doing conventional research would have undoubtedly consigned the hapless computer to the scrap heap. But for Boran, the Akron I's response represented a startling breakthrough in a little-known field: artificial stupidity.

Boran is the head of NASA, the National Artificial Stupidity Association ("Not to be confused with those space people," he is quick to point out), a loosely-knit band of computer-school dropouts currently occupying an abandoned fraternity house at the University of New Mexico.

"There's been a lot of attention given to developments in artificial intelligence," Boran explains, "but relatively little emphasis on stupidity. Dumbness is, in many ways, a far more difficult quality to synthesize than intelligence. Human beings has a remarkable capacity for fallacious reasoning, illogical conclusions, and plain ignorance -- traits that are unique to them and alien to conventionally programmed computers. My goal is to generate a program that can accurately simulate the full variety of human stupidities."

Those initial errors of the Akron I, involving the total inability to interpret or follow even simple directions, as well as a moronic level of mathematical competence, were a promising start. Since then, Boran and his staff have made numerous other significant breakthroughs, among them:

  • A program known as IDMBH (an acronym for "I did my best, honest," the computer's most frequently heard lament). Not only has IDMBH thus far failed to solve a single problem or even retrieve one piece of data, it has also generated an impressive variety of inane excuses, ranging from I DIDN'T KNOW YOU WANTED IT TODAY to THE DOG ERASED IT.
  • NON SEQUITUR B04, a particularly costly program to design, due to the vast amount of information that had been stored in it. Despite a wealth of accumulated data, the B04 fails to respond to any request in even a remotely organized fashion. Instead it answers with a speculative data response - a guess - made by sifting through and spitting out data in what amounts to a random process. For instance, when asked to provide a brief rundown on earthquake zones that might show activity in the next five years, the B04 supplied several hundred suggestions, including: ARIZONA? MARS? THE KREMLIN? DISNEYLAND? PIKE'S PEAK? THE BRONX ZOO? THE NABISCO FACTORY? THE ROSE BOWL?
  • AGGREPOST PR, an aggressive-posture, pointless-rationalization program. AGGREPOST's unique "stupidity factor" is not based on its consistent fallibility but rather on the obnoxious extent to which the program will go to defend its erroneous conclusions. A typical exchange with AGGREPOST was one in which one of Boran's senior programmers challenges the computer's assertion that the city of Tijuana is militarily superior to the United States. Rather than back down, AGGREPOST proceeded to support its claim with a slew of fictitious "facts" and "evidence," including reports of troops massing at the border of Mexico, armed with cheap pottery.

These developments are certainly a far cry from NASA's primitive early programs, in which computer responses were rarely more sophisticated than I GIVE UP, HOW MANY? or YOU TELL ME. Despite this impressive progress, a fundamental question hangs over the whole discipline of artificial stupidity, a question faced by all ground-breaking research projects: What's the point?

For an answer, NASA went to its own GLIB 5000, one of a series of smart-stupid models designed to present inanities in as sophisticated a manner as possible. GLIB's official assessment of artificial-stupidity science was as follows: ALL AVAILABLE EVIDENCE INDICATES THAT NOT ONLY IS A.S.S. OF DIRECT BENEFIT TO THE PARTIES INVOLVED IN CONDUCTING IT, IT IS IN NO WAY AN IMPEDIMENT TO LASTING PROGRAMS AIMED AT AIDING THE POOR AND ELDERLY, REDUCING GLOBAL TENSIONS, AND ULTIMATELY ACHIEVING A LASTING WORLD PEACE.

Arthur Boran's answer is more down-to-earth: "All of us, at one point or another, have received a phone bill for one million dollars or a lifetime supply of industrial-strength otter poison. What are these inevitably attributed to? 'Computer error,' of course. It's difficult for humans to really be sure when the computer is screwing up.

"At NASA we're trying to correct all that. By designing programs that accurately simulate human stupidity, we have made it a simple matter for scientists to perceive at once what their computer is doing wrong. Right now, World War III could be triggered because of some overload in a silicon chip controlling a NORAD missile silo. Wouldn't it be of some consolation to have a word of explanation from the computer, something like OOPS, I THOUGHT THAT SOVIET POTATO TRUCK WAS REALLY A DECOY. IT WON'T HAPPEN AGAIN, OKAY?

One might be tempted to call Boran's reasoning, well, stupid. But in all probability he'd take that as a compliment.

No comments: