RSS
Twitter
Why Turing Testers Hate The Loebner Prize
from the stupid-rules dept
An interesting discussion about the Loebner Prize for the Turing test where some of the contestants point out why they think the Loebner version of the test sucks. Basically, the main complaint is that in the original Turing Test, the "judges" aren't even supposed to know they're testing to figure out if the person they're talking to is human or a computer. So, they don't try to ask "trick" questions to foul them up. Of course, that's pretty damn difficult to set up as a "test" in real life. Would you just bring in random "judges" off the street, have them chat away on a computer for half an hour or so, and on their way out the door say, "oh, by the way, were you talking to a real person or a computer?".
Reader Comments (rss)
(Flattened / Threaded)
Um....
I thought that's what IM or internet chat rooms were for...?
(reply to this comment) (link to this comment)
Re: Um....
See?
http://www.fury.com/aoliza/
(reply to this comment) (link to this comment)
Test
The Turing test is an entirely stupid and bogus idea so who the hell cares!!
(reply to this comment) (link to this comment)
Loebner Prize and Turing Test
Your statement is wrong. Read Turing's article.
Turing _explicity_ described a game: "The Imitation Game" the object of which was for a judge to decide which of two humans (a man and a woman) was the woman when the man's job was to fool the judge.
The judge knew that there were two people, knew that one was a man and one was a woman, and knew that he (or she) had to decide which is which.
Turing then wrote "We now ask the question, "What will happen when a machine takes the part of A [the man] in this game?"
HGL
(reply to this comment) (link to this comment)
Add Your Comment