Information Technology Reference
In-Depth Information
gent. For example, the speech recognition and answering systems do
not attempt to analyze what is said or determine its meaning.
Rather, they seek to perform useful tasks, based on rules, compar
isons with many examples stored in a database, and the use of sta
tistical analysis. The systems do not attempt to summarize words or
carry on general conversations. They succeed because they focus on
solutions to specific and limited problems, and they utilize mathe
matical models that take advantage of a computer's ability for the
highspeed processing of data. Their methods are often quite differ
ent from anything that a human being might be expected to use.
Although these applications might pass a blackbox Turing Test in
a very limited domain, their developers make no attempt to mimic
human problem solving or approach something that might pass
whitebox testing.
intelligence?
Over the years, futurists have established a long tradition of
rarely foreseeing the future of technology; for the most part, few
people have been able to make reasonable predictions about the
longterm future of computing—that is, the nature of computing be
yond three to five years. Technology typically evolves differently
than futurists predict, or individuals' visions of the future may be
shaped too much by their past experiences.
Also, in looking ahead, we need to remember that research nor
mally proceeds incrementally. Although some people theorize about
a single breakthrough that will revolutionize technology, the nature
of scientific discovery usually is more methodical and evolutionary.
Here is another segment from the talk by Woody Bledsoe at the
1985 AAAI meeting:
First let me express my annoyance with some of our de
tractors who criticise AI researchers for not “jumping to in
finity” in one leap. Somehow to them it is OK to work step
bystep on the dream of obtaining controlled thermonuclear
energy or a cure for cancer or a cure for the common cold,
but no such stepbystep process is allowed for those trying
to (partially) duplicate the intelligent behavior of human be
ings. To these cynics, a Natural Language system which
converses with us in a restricted form of English is somehow
