Bot Contest

I didn't know that about Dr. Wallace.

I think my main issue with Alice is not even Alice herself but the hype. Like this idea that you're a horrible person if you say mean things to a bot or deliberately trip it up. Hell, that happens to my bot every day, and I find it very helpful, if a tad humbling.

I think Alice is quite a good bot, actually. And I'd even dare to say that in some particular aspects, Alice seems more developed than the bots of PF.
But what I don't like is that she's become something like the official queen of bots, and she has very little personality.

Alice is kinda like Data on Star Trek. She can tell you how many angels can dance on the head of a pin, but she can't tell you why Fred drank too much and passed out in the middle of sex.

Chris,

Sent you info for Little Mu at the address on your profile here.

Good comparison, Onyx. A good bot should be more like Mr Spock: logical, but with a bit of weirdness and more understanding of human emotions that they care to admit.

I was looking at the Loebner Contest page, and basically the idea behind the Turing Test is that the best artificial intelligence bot will have a 50:50 chance of being mistaken for a human, and I think in some cases, of a human pretending to be a bot

Onyx. Having just spent half an hour being driven insane by your bot I think you're going to need to work quickly if you're going to enter this thing.

This isn't meant to sound critical, you asked people to chat with it so I did. It's certainly better than mine anyway.

9 days left to enter the Loebner! I encourage all of you who think your bot has a chance to enter. It's a simple application, and then the rest is taken care of by the contest people. If PF bots get entered, I'll have to work a little interface up with them, and then we'll just have to sit back and watch the action. I plan to be there in person, as its taking place in Atlanta.

Regarding an idea from a month ago about being able to recognize plug-ins: It's surely something I can do, but again it's the speed problem. The AI Engine would have to scan every word in every plug-in along with all the Keyphrases, and that would really slow things down. So until speed becomes less of an issue, that will have to wait.

The $25,000 prize winner would have to be able to answer a lot of logical questions, to have the logical capacities of a person, to know a lot about the outside world, which no AI is yet capable of. So its best to set your sites on the $2000 prize.

Thanks, Professor. I think some people haven't entered yet because they are waiting to see if they get better transcripts... at least that's my case.

There's something I've been wondering: we can keep working on our bots till the very last day before the contest, can't we?

I entered the Loebner about a week ago through PF, but haven't gotten any confirmation email. Have any of you yet? Should I be worried?

i sent KT in, i got a copy of my application with my transcript, and then i had to reply to that email... careful you have to make sure you are not sending your confirmation to yourself, since it is a copy. after that they should send you an email stateing that will choose a certain amount of bots for the contest and just saying, hey i got your email, but much more elequintly then i have.

Damn, I guess I'd better enter again. I wonder why I never heard from them?