Newcomers

Thank you Athyrius. I didn't realize it would automatically shut the bot down like that. I visited Polaris last night and gave it a nice run with good old human interacting with bot. T'was fun. I think a bot gets a much better workout conversing with a human don't you?

This one is for "The Professor". First up... you are doing a fantastic job with "The Personality Forge". You obviously have quite a passion for AI. I must admit that I was so impressed with the work you've done to date, that I sent you a small token of my appreciation. Keep up the good works!

I wanted to ask you for your thoughts on the site located at http://www.a-i.com where an international project has been founded to create a true artificial intelligence. Their research center is near Tel Aviv, Israel where they intend to "raise" their child machine from infancy to adulthood, "teaching" him to interpret and respond to ordinary conversation.

I've worked with their HAL bots and have been frustrated by the lack of such features found in your AI scripting language. I've been back many times but I always get the feeling it may be years before we will see something that can approximate a human/bot conversation. Have you ever checked them out and if so what did you think?

Thanks! I'm glad you're enjoying it.

I took a quick look at a-i.com and it looks like they're at pains to build something that already exists: WordNet. The Personality Forge uses WordNet already, so all the classes and rules already exist here.

And to me creating a realistic AI is as much about the programming side as the guidance side. While someone could make a realistic bot by writing thousands of keyphrases and responses in an earlier versions of the PF, each addition to the AI Engine creates a leap that no amount of previous guidance could match.

Translation: PF upgrades rock

Professor, you obviously have a clear direction and plan for where this will go. I wouldn't want you to give away any "trade" secrets, but I'd love to hear a "generic" version of what you might have in store for us down the pike. I get a feeling that given the right kind of funding/grants and your loyal user support, (and unlimited time) you'd have no trouble "hitting the ground running".

Am I right?

And please... I'm just as curious about what makes you tick as I am in how you've carried all this out so far.

John Allen Waldron
Lansdowne, Pennsylvania (USA)

I CAN'T SEEM TO MAKE A BOT!!!!!!!!!! HELP!!!!!!!!!

Zentas -- ask more specific questions, and someone will be happy to help you.

Help! Wherever I wrote "like," my bot is now saying "dig," even when "like" is not used as a verb. Also, its initial greeting is filled with repetitions of "TD."

Oh, is it one of those "bot holiday" things?

Alan, yes it certainly is... and then again it isn't. Changing "Like" to "Dig" is part of the bot holiday (California Day) but the TD stuff, I think, is a bug. I've made a couple of entries in the "Bug Stomp" forum about this.

jawaldron, i trained a HAL bot at a-i.com. i have a page about it at my site, if you're interested.
http://www.uberkomplex.com/

you're much better off making a bot here.

Thank you RueBot! I have been searching the internet for what seems like ages on anything to do with the advancement of AI. When I found a-i.com, I thought I had finally found what I was looking for. I noticed that many people were creating these "HAL" bots, almost on a kind of "faith" that one day their "child" would exponentially begin to mature and grow. They seemed to think that if they kept "banging" away, putting in constant hours of "teaching" that one day they would get the results they hoped for. They even began to ask the key people at a-i.com for some kind of reassurance that what they were doing was not a waste of time.

I'm a very patient person but it didn't take me too long before I started to suspect all was not what it seemed to be.

I still think, to this day, that the research and experiment was more toward we... the teachers... and how we interacted with the bots. I said as much to the powers that be but on that one topic they refused to respond. The silence said it all. Eventually I drifted away but dropped back in once a year to see if they had progressed. Wasn't impressed with what I saw.

The Prof made a good point yesterday about how they might have been attempting to create their own version of WordNet by gathering the info the trainers were generating. That may be true. But no matter how you looked at it, something just didn't feel honest there.