Can Computers Think or Have Consciousness?

can we use the word think when it comes to computers do computers think well you know what Alan Turing said about that but that that was just fraught with trying to understand what the word think meant so I kind of side with Alan Turing but I did I wanted to get back to something it's kind of underneath this question of computers and can they well should they think like us I agree entirely with Roger why I mean when you we I mean I think we can say something about why but when you use a computer to model systems in the physical world you don't expect them to do it or act like those systems in the world and yet there's something that just grabs people when you're modeling us and our thinking that that suddenly they want to know if it's going to be like us or if it's going to take over and I did you opened with talking about a science fiction so I wanted to point out that people's fear and it's really fear that gets reflected in the science fiction that the robots are going to take over and get rid of us what but this is an old fear that goes back at least to the Golem of Prague which was in the 1500s something made of clay right that you know acted like us so by the way this is a deep-seated fear in the Western world it's not there in Japan so there's there's something going on here but that's what Frankenstein was about and are you saying that we shouldn't worry about that that yes so Elon Musk is wrong Stephen Hawking was wrong they're these people who said that you know watch out I mean you know run away eh I could take over it eliminate us I think there are many things to worry about many right now with AI systems and that those worries are distracting us from paying attention to the things that we should be concerned about and that everybody in this room could be taking action to help get more attention paid to so okay I'm gonna come back to those later but Raj you don't worry about AI taking over no no I'm not I'm not worried I there are I'm worried about a lot of things but not particularly that it's an interesting question though but I'm not worried no and what about that question of do computers think is how would you respond to that does it tree think so what constitutes thinking it's a hard question and we but we use words very haphazardly what's this word easily we say that the computer didn't understand me or the computer is thinking now and the computer remembered my phone doesn't remembers we use all these words that we use for humans about our things let me just follow up on the trees for a moment it actually that's a very interesting question because trees have all kinds of ways of responding to natural stimuli to whether it's insects whether it's wind patterns and they respond based on all of this stuff you could say I don't know is that thinking yeah is it conscious so there are all these things that are hard to define and we're pork' poking into them you know life thinking consciousness understanding and these these words are all vague and they can be defined in different ways when you settle on the definition then you can answer those questions but that's the hard part so how to define okay well okay I mean lokay let's just throw out that word conscious so I mean when people talk about consciousness does something have consciousness it usually means is their subjective experience does you know is there of can there be a feeling of pleasure or pain is that would that be possible for computers or computer systems AI sure why not I mean seriously why not I mean that's that's that's the things if you can prove to me that's in its in it's impossible for some reason I would I would go along with that but so I know impossibility is always hard to prove it since certain mathematical context yes um but I'll ask the question of whether an an agent so I'm already presupposing agency and autonomy needs to be alive to be conscious and what's the answer well I think so and I don't think machines are alive yeah so you can but why not go there all the way we can make consciousness why not so there's this simplest proof of that is that Nature has built the thing that its conscious and it's it's our brain so it's certainly possible to configure stuff matter in such a way that it becomes conscious easy proof done you know ostensibly we are here we're conscious why can't we not engineer that that's the hard question and yeah why not I think the answer is we don't know yeah you can't tell by the words that we use actually Turing also said this he predicted not only something about the so-called Turing test but he predicted that 50 years from when he wrote this program which was the he wrote it into 1950 so 2,000 that people would talk about machines thinking and nobody would question them and as Rogers pointed out we all do it but that's attribution to when to went to the Machine and we know there's some really wonderful studies that show that you don't have to be a complicated machine for people to attribute even intentions to you so there psychology experiments where they have like little balls moving around on a screen and triangles and the balls are I quote chasing the triangles now I don't know about you but I have trouble attributing consciousness or intention to it and yet as people trying to explain what's going on we find it very useful to do that attribution yeah and it's very easy to do this you can take a cardboard box and this was actually experiment in New York cardboard box below flag that says help you and the only thing this robot does is what go forward it's what's called tween baht or something in the middie owes of this and people you know had empathy with this thing it's a cardboard box with wheels with a flag this is help I need to go through the South corner and people like and it fell over and people like oh and if it takes that little to get empathy with the Machine think about what's going on now we have voice assistance and we have these a eyes that are you know much more what it was to work powerful and it's easy to achieve attribute these things okay so it sounds like one of the differences between your two perspectives is for you Roger the line between alive and not alive it's not so significant in terms of some of these other things that we talked about whether it's consciousness or maybe even thinking or as for you Barbara that's that's a fundamental dividing line do I have that right at the moment yeah till somebody shows me otherwise and I do want to go back to your question because I think I think that one could ask about kinds of thinking and so we talked earlier about machines being very good at taking large amounts of data and recognizing patterns but they're not very good at determining causes and effects and they have no capability to do what's called reasoning reason counterfactually or hypothetically if I go this direction this will happen and if I go that direction this will happen so which direction should I go

Channel: NourFoundation

Views: 1,620

Rating: 4.818182 out of 5

Keywords: artificial intelligence, ai, consciousness

Id: BCr3b-F0sSQ

Length: 8min 10sec (490 seconds)

Published: Thu Dec 20 2018