[Art]

Ray Kurzweil's chatbot is nicely done and pretty on the ball with responses (aren't they all)...
You can also change the speech based on location and the degree of nerdiness....

http://www.kurzweilai.net/ramona4/ramona.html#

[victorshulist]

you> hello

Ramona> Hello, I'm Ramona. What's your name?

you> victor

Ramona> Hi there, Victor.

you> I went to a party given by my company

Ramona> Was it a good experience?

you> It would have been if my wife came with me

Ramona> What do you do not mean you and all of your wife?

Ramona> How much would you pay for it?

you> I am going to New York next week

Ramona> I'm very happy for you.

you> Am I going to New york?

Ramona> I think you've improved a great deal.

you> I own a red car

Ramona> I haven't heard anything like that before.

you> do I own a car?

Ramona> Not that I know of.

you> My wife's name is Christine

Ramona> How long have you been married?

you> 4 years

Ramona> What is your fondest memory?

you> am I married?

Ramona> I think a lot of people are.

-------------

yeah... need's a little work

[victorshulist]

Very good looking woman though. ..and very good speech synthesis.

[Art]

Victor,

When I first tried, I sort of played the devil's advocate and engaged in a simple, leading conversation, which Ramona seemed to follow for the most part.
I tried again with questions similar to yours, testing the programs ability to retain even short term info, which it failed miserably.

It asked my how long I had done a certain task. When I told it 10 years, it then asked what is it like being ten years old? It cannot make any connection to what it is asking compared to the answer it receives.

Clearly, it is in need of some major work (as are most of them).

Thanks!

[squarebear]

From the responses you have posted, this is just another A.L.I.C.E clone. I see nothing of interest here.

[Art]

From the responses you have posted, this is just another A.L.I.C.E clone. I see nothing of interest here.

Well Steve, I suppose the interesting part was that it was from Ray Kurzweil, one of the most inventive minds in our modern world and while it apparently turns out to be an "ordinary" AIML bot, I was initially hoping for something...greater. Sorry if my find was disappointing to you. Some day, someone will get it right...or at least better than what we have today!

[victorshulist]

Agreed.   I don't want to sound too negative.     It was still worth mentioning and creating the thread for it.      But it was an undeniable fact that it does need work

Yes, Ray Kurzweil certainly is a huge name in AI.   I watch all of his Singularity Instiute videos.    I like the way he thinks, a very positive and enthusiastic member of the AI community for sure.    I loved "Transcendent Man".

[squarebear]

Sorry if my find was disappointing to you.

Your find wasn't disappointing at all but the bot was. I was kind of expecting something a little better from such a huge name as Mr Kurzweil.

[Art]

Thank you gentlemen!

I am always looking for something to help spark our interest in this forum and in our hopefully, soon-to-be near future. I have been experimenting with, researching, testing and looking for that all too elusive creation, that chatbot that has potential to react, emote, engage, assist and interact on a much higher plane than the ones we currently have.

There have been and currently are a LOT of bots, unfortunately, only a few are even noteworthy or close to being that one bot that holds our attention and admiration for the potential it holds.

I place the advancement of the conversationalist program somewhat akin to the search for the cure for diabetes (and perhaps other diseases). It's always been touted as being "right around the corner"...soon to be a breakthrough or similar jargon. The end result always seems to be the same. Nada...zip...nothing of substance. Note: I'm just making a similarity by the comparison of one with the other. This is in no way meant to place bots on the same level of priority as that of those diseases that plague mankind but rather to point out a comparable level of disappointment by others in the field. (I happen to have two daughters with diabetes - different types)...so you know.

I keep looking and searching and hoping to find the program that will enhance the community and maybe the entire world if modded to a grander level. The one that Kurzweil often describes as perhaps helping to kickstart the singularity...when computers become self aware. My hope is that it happens before I'm too old to either enjoy it or realize it.

As my dear, departed friend, Knytetrypper used to say, "Onward thru the fog!"

[victorshulist]

This is in no way meant to place bots on the same level of priority as that of those diseases that plague mankind..........

Actually I think of a full fledged 'Turing Test Passing' level AI could assist in the research of disease and cures

[Art]

Yes Victor, imagine what a self aware, learning bot would be able to accomplish on its own! Performing tireless research and experimenting with what-if scenarios without stop, 24/7 and being able to judge, record and post results or findings. Very cool application!

There are a myriad of possibilities for such intelligent entities from research, care giving for the sick, elderly, shut-in, assistant, tutor, Brains behind "Smart homes" of the future, monitor for security, teacher, nanny/babysitter, inventory control, shopping / ordering, planning...etc...the list goes on!

[victorshulist]

That's what I see happening.   First they will be very intelligent assistants, like Gordi Laforge in natural language troubleshooting with the star trek computer on Next Generation.  Then, stage 2 , full blown researchers in their own right.    Such an AI will be the largest breakthrough since the computer and silicon based semiconductors.