It looks like you're using an Ad Blocker.

Please white-list or disable AboveTopSecret.com in your ad-blocking tool.

Thank you.

 

Some features of ATS will be disabled while you continue to use an ad-blocker.

 

WATSON, One step closer to A.I. ?

page: 1
8

log in

join
share:

posted on Aug, 31 2010 @ 08:14 PM
link   
WATSON is the name for the machine that IBM is building that has the goal of competing with humans in our trivia game known as "Jeopardy". You might think of Jeopardy as just a Trivia game but it is much more than that.

researchweb.watson.ibm.com...


Jeopardy! demands knowledge of a broad range of topics including history, literature, politics, film, pop culture and science. What's more, Jeopardy! clues involve irony, riddles, analyzing subtle meaning and other complexities at which humans excel and computers traditionally do not. This, along with the speed at which contestants have to answer, makes Jeopardy! an enormous challenge for computing systems.

Watson will also incorporate massively parallel analytical capabilities and, just like human competitors, Watson will not be connected to the Internet, or have any other outside assistance.


This seems to me to be an amazing goal and if it is successful could be a landmark towards AI. All of this is going to be attempted as they said without being connected to the internet or an outside system. Completely self contained information storage and processing center. You remember the "Deep Blue" system that IBM built that beat the world Top Ranked Chess Champion Garry Kasparov? They managed that through sheer brute force of 200 million chess move calculations per second.

The WATSON system is WAY beyond something as simple as chess. It is being described as a machine working towards thinking in human Terms.

I hope we can embrace technology of this kind and still be wary of the mistakes that we as humans have made through no fault of our own but simply going through the stages of Evolution. Perhaps we can build machines smart enough to learn from history and not repeat it. I do believe that we have come to the time when we as humans have reached a limit as to how far we can advance with our current bodies.

There is only so much information we can retain, only so much we can process. We only have so many years before we are forced to reteach every single thing that we have learned in our entire human history on to the next birth that will last for maybe 100 years if lucky.

And to end on a related quote

"My CPU is a neural net processor; a learning computer." -Terminator 2




posted on Aug, 31 2010 @ 08:37 PM
link   
I can see a "Terminator: Salvation" happening. Hopefully these guys wont be the idiots that expect machines capable of thinking for itself, and learning. I don't see any peaceful purposes in this either. From computers that think like humans jumps forward to the military wanting rolling/walking/flying machines that think like we do as well. But question is, do they care? They would fund out of their ass for super AI war machines, and sleep soundlessly to the thought of endangering our entire civilization.



posted on Aug, 31 2010 @ 08:41 PM
link   
Thanks for bringing this to light. This is a great chance for humanity to advance itself or break itself. Hopefully we won't delegate something this intelligent to tasks that only humans should manage, like weapons, security, or anything like that. Not that I don't trust machines, but random chance is always a possibility and I'm not willing to take a risk like Skynet...



posted on Aug, 31 2010 @ 08:54 PM
link   
reply to post by QuantumDeath
 


If we ensure that our machines never stop learning than there is hope. Besides do you think we could build machines that could ever even begin to hold a candle to the level of cruelty, anger, prejudice, hatred that we as humans have done to ourselves for as long as there has been a human history? We are the ultimate accomplishment when it comes to that.

What we inflict on other humans goes beyond any logic that a machine would be capable of. A machine with access to the world encyclopedia, the entire combined history of our planets knowledge, would hopefully make far better use of it than we have.

I think it is worth the risk.


[edit on 31-8-2010 by idonotcollectstamps]



posted on Aug, 31 2010 @ 10:20 PM
link   
reply to post by idonotcollectstamps
 


That is, unless they see humans as detrimental to Earth and wipe us out.



posted on Aug, 31 2010 @ 10:26 PM
link   
reply to post by idonotcollectstamps
 


If you check this thread out that I wrote a few months ago, you'll see that the US military (DARPA) are extremely well placed from a developmental/technical aspect to becoming the 1st to potentially create an AI system.

Project DARPA-BAA-09-03 - will this be the 1st true machine intelligence ?



posted on Aug, 31 2010 @ 11:30 PM
link   
You can't stop evolution. And the creation of AI is just a natural progression of human evolution. If the machines turn on us then so be it. Nothing more than karmic retribution against the human race that has been for the most part a plague on this planet. We they don't turn on us hopefully theywill create new technology that will help us move beyound this mentality of fear and selfish hoarding of resources.

Either way it's all good. Bring it on. Anything is better than the sad state we are in now. Humanity is in limbo with no direction. It's global. Nobody knows who they are. Maybe the AI will provide an unbiased point of view.



posted on Feb, 14 2011 @ 11:26 AM
link   
Hi all,
i would like to ressurect this thread and i think its quite appropriate time, cos today should start so far the most awaited round of Jeopardy, where Watson should fight against alltime human masters of this show.

I found out about this all thing just today, so iam quite new into it. But it blew my mind becouse its something iam wondering from time to time. Mainly the problem of programing the ai.. I incline to think that true inteligence is not just about programing, the program may be the "gate" or the medium, but then something else needs to enter it or be created from it. .. i tought about circle, such simple and precise form, that could give life, serve as generator of some sort.. sorry for rumbling.. just the feeling


knowing about watson, if its able to "understand" or better say work with our language, then the language could create uniqe quality to utilize its capabilities, .. like becoming alive through it, probably not as new entity but rather as summary of all humans .. well not sure about anything now
,

but i would asume that if the language is used as a scheme to "think" or any other similar procedural way is used, rather then trying to program actual behaviour, then the outcoming effect should not be something to be afraid of. In other words, i believe that our bad habits should not be able to enter into something such supreme as AI, once its trully on its way to be created, not becouse someone would not allow it, but rather it wont be possible.

i would second you in your opinion TiM3LoRd and i consider this to be one of our possible "salvation points"
edit on 14-2-2011 by ClevererRunbeening because: (no reason given)



posted on Feb, 14 2011 @ 11:44 AM
link   

Originally posted by tauristercus
reply to post by idonotcollectstamps
 

If you check this thread out that I wrote a few months ago, you'll see that the US military (DARPA) are extremely well placed from a developmental/technical aspect to becoming the 1st to potentially create an AI system.
Project DARPA-BAA-09-03 - will this be the 1st true machine intelligence ?
i see u are quite concerned about it
.. i understand you now
this reply u got in your thread from projectvxn is pretty much what i had in mind when writing my post



My friend brought up the fact that in order for a machine to be truly smart it would have to be able to communicate with us on our terms. It was then that I realized the only way to create AI would be to teach it everything there is to know about language. To to read it, how to write it, how to analyze and produce it's own data based on what it learned...And over time, the Singularity would happen on it's own, probably very soon after.

language being the Shem Hameforash to make Golem alive.

edit on 14-2-2011 by ClevererRunbeening because: (no reason given)



posted on Feb, 15 2011 @ 10:33 AM
link   
so its 5000 vs 5000 tie after 1st round,
pretty impresive,
we were all watching the video at my work, like if it was some sort of sport championship..
lots of thoughts went through my mind while looking at it



posted on Feb, 15 2011 @ 10:35 AM
link   
I dont think its intelligent at all, just a really impressive way of querying a database.



posted on Feb, 15 2011 @ 03:22 PM
link   
Yes, this is a step towards A.I. I actually felt sorry for Watson when he got a few questions wrong. I felt sorry for a computer.

I think the importance of Watson is two-fold.

First, Watson goes along way towards computers understanding natural language of humans. This is very important.

Second, Watson is basically doing the same thing that we do when we take a multiple choice test. He has confidence in his answer on some questions and you can see when he doesn't think he has the right answer.

People say, well Watson has all of this information uploaded. Well, we work the same way. A child has to learn what's an apple is or what's a piece of candy is. We have to upload that information so to speak. What separates Watson from us is the ability to form a world view based on the information.

I think the way you bring about A.I. is you make something like Watson the main hub that's connected to A.I. Robots. These robots are not uploaded with any information but they're connected to the Collective so to speak.

So when the robots see a T.V. or a Car they have to access the Collective to learn about what they see.



posted on Feb, 15 2011 @ 11:01 PM
link   

Originally posted by Matrix Rising

First, Watson goes along way towards computers understanding natural language of humans. This is very important.

Second, Watson is basically doing the same thing that we do when we take a multiple choice test. He has confidence in his answer on some questions and you can see when he doesn't think he has the right answer.

People say, well Watson has all of this information uploaded. Well, we work the same way. A child has to learn what's an apple is or what's a piece of candy is. We have to upload that information so to speak. What separates Watson from us is the ability to form a world view based on the information.

but if something can understand all the language.. doesnt it form the world view ? i think that metaphore with child is very close.. such computer inteligence would be missing certain things but could have others qualities. I see quite a problem that when people think about AI they think about it in the way their mind works.. computer is different so ai cant be imho programmed it has to emerge (imho
)

there is another round www.youtube.com... .. IBM probably overclocked the Watson guy since his performance is mindblowing now..
edit on 15-2-2011 by ClevererRunbeening because: (no reason given)



new topics

top topics



 
8

log in

join