Turing Test

by C Mitchell

The Turing test is an assessment of a machine's ability to act like a human. It involves three participants- a human judge, A, and two others, B or C. Both B and C, one of which is a human and one of which is a computer, will try to act as human-like as possible. All participants are in separate spaces and communicate using only written text, so it will remain an intelligence test rather than a test of how well a machine can replicate the human voice. During the test, B and C are supposed to engage in normal conversations with the judge, A. If player A cannot always determine which player is the really human, the computer is said to have passed.

Turing_Test_version_3.png

Although many people have discussed and considered the topic of artificial intelligence, or the question “Can machines think?” throughout history, it was Alan Turing who first really asked the question “Can machines think like humans?”

_TURING.GIF

Alan Turing was a British mathematician and philosopher who lived from 1912 to 1954. He was a member of the Ratio Club, a small club of psychologists, mathematicians, and engineers who frequently discussed the subject of artificial intelligence. Turing in particular had been considering the idea of intelligent machines since at least 1941. In 1950, he published a paper called "Computing Machinery and Intelligence”. In it seems to be the precursor to the Turing test of later years:

"It is not difficult to devise a paper machine which will play a not very bad game of chess. Now get three men as subjects for the experiment. A, B and C. A and C are to be rather poor chess players, B is the operator who works the paper machine. […] Two rooms are used with some arrangement for communicating moves, and a game is played between C and either A or the paper machine. C may find it quite difficult to tell which he is playing."

He asked a new question: "Can machines do what we (as thinking entities) can do?" Turing then proposed to turn a previous party game called the "Imitation Game" into a computer intelligence test. In the original game, a man and a woman each go into a separate room. The other guests then try to figure which of person, both of which are trying to pretend to be the other, is which, but by only communicating with written notes. In his new version of the “Imitation Game”, a computer takes the place of one of the “guests” who were pretending to be each other. “Will the interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman?” he asked in his paper.

In 1980, John Seale wrote a paper called Minds, Brains, and Programs that argued against Turing’s position. He said that some software, one called ELIZA for example, could pass the “test” by just using symbols, that they of course could not understand, in a way that would make the judge believe that they were human. This, he argued, could not be called “thinking” in the way that humans do it.

In 1990, the first Turing Colloquium was held at the University of Sussex, where, many people came to discuss history and future of the Turing test. Also that same year, Dr. Hugh Loebner, with the Cambridge Center for Behavioral Studies, underwrote a contest to actually put into practice the first Turing test.

The year after that in 1991, the first Loebner Prize competition was held. Neither the Grand Prize of $100,000 and a gold medal or the silver medal (see picture below), to be given to the computer that could not be distinguished from the human, have ever been awarded, but a bronze medal and $2,000 has been given out every year since the first competition. The computer programs that “compete” are usually chatterbots, or Artificial Conversational Entities (ACE). The Loebner Prize Competition is still being held to this day.
GoldPrizeAMT.jpg

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License