Computers Take the Turing Test for Artificial Intelligence, But Fall Short

[one...two...tree 2,762]

Five sophisticated computer programs chatted with human examiners this weekend and tried to convince the judges that they were conversing with another human being, vying to be the first to pass the Turing Test for artificial intelligence. Although none of the programs achieved their goal of duping 30 percent of the judges, several came quite close to that threshold and all fooled at least one judge. Said organizer Kevin Warwick: “…although the machines aren’t yet good enough to fool all of the people all of the time, they are certainly at the stage of fooling some of the people some of the time” [Telegraph]. The contest draws on the ideas of British mathematician Alan Turing, who came up with a subjective but simple rule for determining whether machines were capable of thought. Writing in 1950, Turing argued that conversation was proof of intelligence. If a computer talked like a human, then for all practical purposes it thought like a human too [AP]. Since 1991, an annual competition has been conducted in which judges chat simultaneously with computer programs and their “confederate” humans, and try to determine which is which.

One judge, Will Pavia, explained that he was fooled by a program that made use of humor and had a fully developed biography: the chatbot claimed to be a teenage boy named Eugene Goostman who loved the novels of Kurt Vonnegut. Later, Pavia said, I was introduced to the charming Ukranian computer program that had fooled me, the creation of Vladimir Veselov, 39, a bio-scientist from St Petersburg. I saw the vast database it accessed: there was the file on Vonnegut, there a list of plausible responses on the subject of Eminem [The Times].

While none of the programs qualified for the gold or silver prizes (which have never been awarded), a program called Elbot took home the bronze for fooling the most judges; it tricked three out of the 12 examiners. But the program’s inventor, Fred Roberts, said he didn’t believe that conversational ability is a true mark of artificial intelligence, as Turing claimed. “I don’t think it’s anything like thought,” he said of Elbot’s conversational prowess. “If you know a magic trick, you know how it’s done, it’s not magic anymore” [AP].

However, organizer Kevin Warwick says that even if these chatty computers aren’t developing a truly human intelligence, they are rapidly becoming good enough at their specialized functions to fill new roles in society. “This demonstrates how close machines are getting to reaching the milestone of communicating with us in a way in which we are comfortable. That eventual day will herald a new phase in our relationship with machines, bringing closer the time in which robots start to play an active role in our daily lives” [BBC News], he said.

http://blogs.discovermagazine.com/80beats/...but-fall-short/

[Peikko 1,575]

Interesting, I love this stuff

[rocks 494]

interesting.

computers can definitely learn from patterns, that is how the weather is forecast and decisions on loan applications are decided by banks etc. i'm not sure that this makes them comparable to the human brain though. artificial neural networks are only loosely based on the brain. this is how they look.

[mawilson 2,133]

It would be trivial to program a computer to debate like Sarah Palin using AJ's flowchart:

[akdiver 3]

Replying to Computers Take the Turing Test for Artificial Intelligence, But Fall Short

Most students of the public education system in America fail the test too.