3 ms·
The Chatbot That's Acing the Largest Turing Test in History
- philipkglass 9y agoIt's possible to lower human expectations to the point that current chatbots can satisfy them. That's a much easier task than raising chatbots up to the point that they can actually hold their own in probing, intelligent conversations without prevarication, as in the original proposal: http://www.loebner.net/Prizef/TuringArticle.html http://www.loebner.net/Prizef/TuringArticle.html Interrogator: In the first line of your sonnet which reads "Shall I compare thee to a summer's day," would not "a spring day" do as well or better? Witness: It wouldn't scan. Interrogator: How about "a winter's day," That would scan all right. Witness: Yes, but nobody wants to be compared to a winter's day. Interrogator: Would you say Mr. Pickwick reminded you of Christmas? Witness: In a way. Interrogator: Yet Christmas is a winter's day, and I do not think Mr. Pickwick would mind the comparison. Witness: I don't think you're serious. By a winter's day one means a typical winter's day, rather than a special one like Christmas. I suppose that cobbling together enough evasions and simulated mood swings to bypass the challenge of a sustained conversation is impressive in its own way. But it's mostly disappointing. Imagine seeing repeated claims of public key encryption weaknesses that all turn out to be variants of social engineering approaches to stealing passphrases. Those wouldn't exactly be incorrect claims but wouldn't be very interesting kinds of correct claims either.