Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

John Searle: Consciousness in Artificial Intelligence [video]

youtube.com

Post reply on HN
Print view View original

51–60 of 62 posts

Re: John Searle: Consciousness in Artificial Intelligence [video]

#51

Post by ecopoesis »

chubot wrote: post #3
bhickey wrote: post #2

There isn't much new here. Skip ahead to the first audience question from Ray Kurzweil ( http://www.youtube.com/watch?v=rHKwIYsPXLg&t=38m51s ). Kurzweil, in summary, asks "You say that a machine manipulating symbols can't have consciousness. Why is this different than consciousness arising neurons manipulating neurotransmitter concentrations?" Searle gives a non-answer: "My dog has consciousness because I can look at…

Yeah honestly I don't get what he is really contributing (and I'm sort of an AI skeptic). In 2000 in undergrad, I recall checking out some of his books from the library because people said he was important, and I learned about the "Chinese Room" argument [1] in class. How is it even an argument? It doesn't illuminate anything, and it's not even clever. It seems like the most facile wrong-headed stab at refutation, by…

I've always thought that the Chinese Room proved just the opposite of what Searle thinks it does.

I think of it this way:

I have two rooms: one has a person who doesn't speak Chinese in it, but they have reference and books that allow them to translate incoming papers into Chinese perfectly.

The second room just someone who speaks Chinese, and can translate anything coming in perfectly.

Searle says that AIs are like person in room one: they don't know Chinese.

I would argue that is the wrong way to look at things. A better comparison is that an AI is like the system of room 1, which does know Chinese, and from observation is indistinguishable from the system of room 2. What's going on inside (a human with Chinese reference books vs a human who knows Chinese) doesn't matter, it's just internal processing.

If it walks like a duck and quacks like a duck, then it's a duck.

If a machine claims to be conscious, and I can't tell it apart from another conscious being, who am I to say it isn't conscious?

Re: John Searle: Consciousness in Artificial Intelligence [video]

#52

Post by orblivion »

adrianN wrote: post #19
orblivion wrote: post #18

So, supposing there's any chance that it has consciousness, is there any sort of movement doing all it can to put the brakes on AI research? If it's true, it's literally the precursor to the worst realistic (or hypothetical, really) outcome I can fathom, which has been discussed before on HN (simulated hell, etc). I'm not sure why more people aren't concerned about it. Or is it just that there's "no way to stop progr…

We are sufficiently far away from creating machines that humans would consider to have consciousness that it's not really a problem so far. Eventually we'll probably have to think about robot rights, but I guess we still have a few decades until they're sufficiently advanced. But judging from how we treat, eg. great apes, who are so very similar to us, I wouldn't want to be a robot capable of suffering.

I'd think that if there are people forward thinking enough to consider the consequences to humans (Elon Musk, Singularity Institute), there should be people forward thinking enough to consider the consequences to the AIs.

Re: John Searle: Consciousness in Artificial Intelligence [video]

#53

Post by mtrimpe »

amoruso wrote: post #46

Searle makes two assertions: 1) Syntax without semantics is not understanding. 2) Simulation is not duplication. Claim 1 is a criticism of old-style Symbolic AI that was in fashion when he first formulated his argument. This is obviously right, but we're already moving past this. For example, word2vec or the recent progress in generating image descriptions with neural nets. The semantic associations are not nearly as…

I think Claim 1 is actually more about determinism; that if by knowing all the inputs you can reliably get the same outputs what you have isn't consciousness.

Neural nets are somewhat starting to escape that dynamic but there still isn't a neural net that reliable pulls in a continuous stream of randomness to generate meaningful behaviour like our consciousness does.

Now to be honest; I'm not entirely sure if John Searle would agree that that is consciousness when we do get there but I do agree with him that deterministic consciousness is essentially a contradictio in terminis.

Re: John Searle: Consciousness in Artificial Intelligence [video]

#54

Post by leafee »

DonaldFisk wrote: post #34

I think Searle's mostly correct and Kurzweil's completely wrong on this. It took me a long time to understand Searle's argument, because Searle conflates consciousness and intelligence and this confuses matters. Understanding Chinese is a difficult problem requiring intelligence, but I don't think it requires consciousness. It is important to distinguish between "understanding Chinese" and "knowing what it's like to…

> conflates consciousness and intelligence and this confuses matters

I think this is an excellent point. I like your example with colors, which shows that there is a difference between seeing (i.e. experiencing) colors and producing symbols which give the impression that an entity can see colors.

I don't follow any argument that proposed that computers can be conscious but other machines (e.g. car engines) cannot. In the end, symbols don't really exist in physical reality - all that exists is physical 'stuff' - atoms, electrons, photons etc. interacting with each other. So how can we say that one ball of stuff is conscious but another is not? And why isn't all of the stuff together also conscious? Why not just admit we don't know yet?

Consciousness may be hard to define, but lets take something simpler - experience, or even more specifically - pain. I can feel pain. While I can't be 100% sure, I believe other humans feel pain as well. However I don't believe my laptop has the capacity to feel pain, irrespective of how many times and in how many languages it can say 'I feel pain'.

Perhaps the ability to experience is the defining characteristic of consciousness?

Re: John Searle: Consciousness in Artificial Intelligence [video]

#55

Post by leafee »

redwood wrote: post #45
DonaldFisk wrote: post #34

I think Searle's mostly correct and Kurzweil's completely wrong on this. It took me a long time to understand Searle's argument, because Searle conflates consciousness and intelligence and this confuses matters. Understanding Chinese is a difficult problem requiring intelligence, but I don't think it requires consciousness. It is important to distinguish between "understanding Chinese" and "knowing what it's like to…

I disagree completely. After time the color filter will start to associate various concepts and feelings add images with various colors. This association is what starts making the colors themselves have meaning even if they can't see the colors the same way that you and I can. There's no way to prove that we all see colors the same way anyway. But that doesn't mean that we don't believe that were conscious. I think I…

> The experience of experience and color is only important because experience and color has a web of other associated experiences that those colors remind us of

So what about those original experiences? How are they important at all if there is nothing to associate them with?

Re: John Searle: Consciousness in Artificial Intelligence [video]

#56

Post by chubot »

tobiasSoftware wrote: post #48
chubot wrote: post #3

Earlier quoted context omitted.

Yeah honestly I don't get what he is really contributing (and I'm sort of an AI skeptic). In 2000 in undergrad, I recall checking out some of his books from the library because people said he was important, and I learned about the "Chinese Room" argument [1] in class. How is it even an argument? It doesn't illuminate anything, and it's not even clever. It seems like the most facile wrong-headed stab at refutation, by…

As a believer in Weak AI the Chinese Room argument really gave me more understanding of my position. His argument is based on the concept that interpretation of symbols is not the same as the understanding that we do. As an example, say that a person learns 1 + 1 = 2. Because that person understands the concept, he can then go apply it to other situations, and figure out that 1 + 2 = 3. Whereas because the Chinese ro…

Yeah I see that, but the problem is that we don't know how humans are conscious, i.e. where meaning rises. If you believe that brains are just atoms, then meaning arises from "dumb physics" somewhere.

Another way to think of it is: a fetus, or a sperm, or ova is not conscious. Some might argue that a newborn isn't really conscious. Somewhere along the line it becomes conscious. How does that happen? Where is the line? We have no idea.

You can't assert that meaning can't arise from "dumb symbol manipulation" without understanding how meaning arises in the former case. We simply don't know enough to make any sort of judgement. The Chinese room argument is trying to make something out of nothing. We don't know.

Re: John Searle: Consciousness in Artificial Intelligence [video]

#57

Post by ttctciyf »

amoruso wrote: post #46

Searle makes two assertions: 1) Syntax without semantics is not understanding. 2) Simulation is not duplication. Claim 1 is a criticism of old-style Symbolic AI that was in fashion when he first formulated his argument. This is obviously right, but we're already moving past this. For example, word2vec or the recent progress in generating image descriptions with neural nets. The semantic associations are not nearly as…

Regards claim 2, Searle repeats the phrase "specific causal properties of the brain" quite a few times without spelling out just what he's referring to, but from other remarks he makes it seems clear he means actual electrochemical interactions, rather than generic information processing capabilities. I think his view is that consciousness (most likely) doesn't arise out of "information processing", which he would probably class as "observer-relative", but out of some as yet not understood chemistry/physics which takes place in actual physical brains.

So the question, to Searle, is not "about what kinds of information processing systems would have subjective conscious experiences", but "what kinds of electrochemical interactions would cause conscious experiences".

The intuition/assumption of his questioners seems to be that whatever electrochemical interactions are relevant for consciousness, they are relevant only in virtue of their being a physical implementation of some computational features, but plainly he does not share this assumption and favours the possibility that the electrochemical interactions are relevant because they physically (I think he'd have to say) produce subjective experience - and that any computational features we attribute to them are most likely orthogonal to this. Hence his example of the uselessness of feeding an actual physical pizza to a computer simulation of digestion. His point is that the biochemistry (he assumes) required for consciousness isn't present in a computer any more than that required for digestion is.

Another example might be: you wouldn't expect a compass needle to be affected by a computer simulating electron spin in an assemblage of atoms exhibiting ferromagnetism any more than it would be by a simulation of a non-ferromagnetic assemblage.

To someone making the assumption that computation is fundamental for explanations of consciousness, these examples seem to entirely miss the point, because it's not the physical properties of the implementation (the actual goings on in the CPU and whatnot) that matter, but the information processing features of the model that are the relevant causal properties (for them.)

But to Searle, I think, these people are just failing to grok his position, because they don't seem to even understand that he's saying the physical goings on are primary. You can almost hear the mental "WHOOSH!" as he sees his argument pass over their heads. In an observer-relative way, of course.

As you imply, until someone can show at least a working theory of how either information processing or biochemistry can cause subjective experience the jury will be out and the arguments can continue. I won't be surprised if it takes a long time.

(Edited to add the magnetic example and subsequent 2 paragraphs.)

Re: John Searle: Consciousness in Artificial Intelligence [video]

#58

Post by qbrass »

rjsw wrote: post #28
dnautics wrote: post #23

Earlier quoted context omitted.

I was never satisfied with the chinese room thought experiment. Let's momentarily replace the thing in the chinese room with a human, to parse Searle's notion of "understanding". Searle would argue that a human trained in to emit meaningful chinese characters would still lack understanding. But I think this is backwards and speaks to your identification of Searle begging the question: the only way a human could emit…

I have always felt that the human in the room would start to recognize patterns and develop an "understanding". Their "understanding" may have no basis in reality but I don't see that it is any less valid to them. If Searle is right then we should be able to perform a MRI on a blind person while they are talking to someone and spot the point where their brain switches into "symbol manipulation mode" when the conversa…

The guy in the room can memorize all of the rules and the ledgers and give you the same responses the room did, and if you asked him in his native language if he knew Chinese, he'd honestly tell you no.

He could have an entire conversation with you in Chinese, and only know that what you said merits the response he gave. He doesn't if he's telling you directions to the bathroom, or how to perform brain surgery.

Re: John Searle: Consciousness in Artificial Intelligence [video]

#59

Post by simonh »

DonaldFisk wrote: post #36
simonh wrote: post #27

Earlier quoted context omitted.

It's just obfuscation. In reality the 'room' would have to be the size of a planet and if one person was manipulating the symbols it might take the whole thing the life span of the universe to think 'Hello World'. But by phrasing it as a 'room' with one person in it he makes it look inadequate to the task, and therefore the task impossible.

Neither the size of the room nor the speed of the computation is important to Searle's argument. You could replace the person in the room with the population of India (except for those who understand Chinese), and pretend to the Chinese speaker that the communication is by surface mail. Or use a bank of supercomputers if Indians aren't fast enough.

Fair enough. In which case Sear's argument is that even fantastically complex, sophisticated information processing systems with billions of moving parts and vast information storage retrieval resources operating over long periods of time cannot be intelligent. If that's what his position boils down to, what does casting it as a single person in a room add to the argument? As Kurzweil asked, how is that different from neurons manipulating neurotransmitter chemistry? Searl doesn't seem to have an answer to that.

Re: John Searle: Consciousness in Artificial Intelligence [video]

#60

Post by DonaldFisk »

simonh wrote: post #59
DonaldFisk wrote: post #36

Earlier quoted context omitted.

Neither the size of the room nor the speed of the computation is important to Searle's argument. You could replace the person in the room with the population of India (except for those who understand Chinese), and pretend to the Chinese speaker that the communication is by surface mail. Or use a bank of supercomputers if Indians aren't fast enough.

Fair enough. In which case Sear's argument is that even fantastically complex, sophisticated information processing systems with billions of moving parts and vast information storage retrieval resources operating over long periods of time cannot be intelligent. If that's what his position boils down to, what does casting it as a single person in a room add to the argument? As Kurzweil asked, how is that different fro…

No, his position, as I understand it, is that it cannot be conscious. It certainly can be intelligent.

Searle does try to explain why there's a difference. Although the person in the Chinese Room might be conscious of other things, he has no consciousness of understanding the Chinese text he's manipulating, and will readily verify this, and nothing else in the room is conscious. Chinese speakers are conscious of understanding Chinese.

Post reply on HN