posted on Sep, 11 2011 @ 02:00 AM
Based on the video, it doesn't look like they actually did anything at all.
Both AI's mentioned cleverbot, which does not actually use bots at all. Instead, it's a chat program, designed to pair up human cleverbot users
with other human cleverbot users so that each user thinks they are talking with a clever AI. Then, every couple minutes or so, you are switched to a
new user to keep the conversation just confusing enough to pass for a scattered computer conversation.
It works the best when you go there thinking that you're talking to a bot. However, when you go there acting like cleverbot, you get some
interesting results. The supposed "bot" often talks like a human talking to a computer, asking profound questions, asking you to compute crazy
equations and such. At that point the illusion comes crashing down. Occasionally, you get other users that have also figured it out and are not so
subtly acting like cleverbot.
You can tell it switches users every couple minutes because cleverbot will all the sudden start to get frustrated because the person on the other side
is confused why you're talking about Texas when they were just talking about love with someone else.
It's pretty funny how many people fall for it. I must admit, they had me fooled for about 20 minutes the first time I went there.
Anyways, it makes me wonder what these grad students were actually doing if they used cleverbot for their source of AI. Hmmm.