h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Psi, just a quick reply as I'm at the computer for a few minutes:

Paragraph 2 - No, my view is precisely that we're not isolated individuals. I think that Musk sees people as isolated individuals which is why he is puzzled by what he thinks is a problem of communication between them. My view is that we don't need to increase our bandwidth because we're not communicating through a linear stream of information. Paragraph 3 - But my point here is that I don't have good grounds for denying the existence of God, it's just an opinion. The rest of it I will attempt to clarify in a longer post.

Hi folks! Interesting discussion going on here. Chris, before you do that long post, just a few things I want to say.

First of all - I utterly do *not* think that minds are nothing but information. Before a mind can have information it needs awareness in itself, the ability to be aware of information. I don't see computers as they are now, as having any awareness. Deep Blue very probably didn't even realize it was playing chess. So what we need before we can have AI's that can be called *minds*, is a sense of awareness in them that will allow them to have awareness and 'sensed' experiences. As I see things we're still far away from that. But things might change once we have quantum computers. Or not! I wouldn't know ... when I say I think a true artificial mind might be possible, I'm not saying that I am 100% convinced that it would be. Maybe it won't. But I am just saying, before we seriously try, we wouldn't know. So *in trying very hard* to achieve it, we might find the answer. But even after serious trials and failures over many decades or even centuries, we might still not know if it was *really* impossible, or if there wasn't just one step in the process that was still momentarily stopping us. But if we *succeed*, then indeed we have learnt something. This is a kind of reverse Popper's falsification principle. We can prove something is possible by doing it; but if we fail to do something that doesn't prove that it is impossible.

But if we still fail after monumental effort we are justified in saying that something might just be really impossible. But AI is such an important thing - at least as I see it - that I consider justifiable the expending of far more effort towards it, than we've expended to it so far. I don't think we've yet worked hard at it at all. We've been working with computers for many decades now, but not all or even much of it has been towards developing a 'real' AI ... instead we've gone for practical things, 'artificial cleverness', and even for 'artificial stupidity', which enables a bot to pass for a rather stupid and one-track-minded human being in the Turing Test. We still lack the basics necessary for *starting* on the project of a true and actually intelligent AI.

My own approach in general is this: we can't settle many of these philosophical issues with just thinking about principles. When I reject a particular kind of philosophical idea, it may be a gut instinct, or I might have things in mind which suggest to me that the principle is wrong. But I can't prove anything and seeing how philosophers debate, it would seem neither can they. The rejection of one principle often is by another principle that can't be proven either. Anyways thus my own idea of seeing if a thing can be proven by something that can be practically done. Not *necessarily* doing an experiment or making something; we can apply thought experiments if they're clearly enough conceived.

For instance: the issue of free will. I can't think of any practical way to prove or disprove free will; people might as well believe they have free will. What would it be like to *not* believe that you yourself have free will? Likely you'd still be going on the same as anyways, thinking you do have some control about what you do, and what you think!

There's also the matter of utility. How would you benefit by not believing that you have free will?

Anyways as for some other ideas ... bandwidth. I would LOVE to have more bandwidth for this discussion. I can only drop by for a few minutes once or twice a week ... if I could communicate a hundred times faster, then I could of course get to so many more of the ideas we're bringing up here. Each of you, also, I think, would benefit from being able to communicate faster. And ... *if* we do get technologically augmented, we might easily gain the ability to communicate at a hundred times the speed that we can manage at the mo... both in sending out info, and receiving it. Would that be beneficial? Yes - if we really wish to learn. If we're going to really think about that 100-page post the previous person has posted in this forum, before hammering out our own 100-page reply. I do think we'd learn faster from each other. We'd clear up misconceptions much more rapidly. But then the goodwill needs to be there - towards the 'opponent' - and also the willingness to learn needs to be there, the willingness to admit a mistake.

I have much more to say but being indeed limited by bandwidth, will stop now and let you folks get back to the discussion!

I have a weird chicken-and-egg question: and okay, I have spent too much of this afternoon reading Aleister Crowley's description of his childhood notions (his family will curl your hair), so this may be the equivalent of drunk-dialing the brain...

What comes first, though: ability to process information and concepts, or self-awareness? I can remember not being self-aware, at least, I think so. Being a sort of camera/actor with no self-concept...

Has anybody seen 'Dark Star'? I believe that's the film. Where the AI-cum-bomb reasons and reasons, then pronounces, 'I think, therefore I am,' followed by 'Let there be light' and the inevitable...

PS to Willem: So you think of free will as a sort of Pascal's Wager? I never thought of that.

Chris,

A quick reply to your reply, I think I have not explained my point very well. As my paragraph 2 implies, I take as a given that neither of us think that we are isolated individuals. My reductio is that despite that fact we actually /are/ communicating via a serial medium to each other in this conversation. Now I can imagine this interface being massively enhanced in terms of bandwidth without it negating my use of language or ideas.

So I don't think we are isolated individuals, yet I can envisage as Musk does, some kind of cortical augmentation to increase output bandwidth. My typing is probably slower than yours!

Therefore, Musk's position does not entail the proposition that we are isolated individuals. That's my argument, sorry for not setting it out more clearly.

On God, I think you might be mistaken in thinking you don't have good grounds for rejecting the proposition that a given stipulated god exists. Consequently, your opinion has a reasonable foundation rather than being arbitrary. Of course, you could still be wrong, and that god might exist. That doesn't make your opinion baseless though. On the other hand there might be certain gods that are so stipulated that we would have no grounds for any stance whatsoever on their existence. But that's a thread in itself.

Dmitri,

Yes, I have seen Dark Star... great film with a truly comic dialogue trying to persuade the bomb not to detonate.

On your chicken-and-egg, the issue I raised is whether or not self awareness simply /is/ a complex way of processing information. This is still a matter of debate between philosophers of course.

Okay, I have another silly question.

Here's some Actual Science being carried out by Actual Scientists. My first response to this New York Times Magazine video was, 'For this they went to graduate school?' But hey, the baby 'scientists' are loving it.

http://www.youtube.com/watch?v=HBW5vdhr_PA

My questions are: Do we buy these results? If so, what does this tell us about human consciousness? Would the ideas represented here be relevant to a discussion of AI?

And could it make Tom, the annoying voice recognition bot from the Poland Spring water company, be more helpful to frustrated customers? Or at least ashamed of himself?

Dmitri,

It seems like interesting research to me, so I didn't have the same response as you did.

In terms of whether we buy the results, I can't tell from the youtube alone, I would have to read the paper. The things to look for in the experimental design would be things like: were the puppets presented for choice counterbalanced between participants in terms of their role in the play, and which hand each was in when presented? Was the researcher presenting the choice blind to which role in the play each puppet had? Obvious things really.

This is relevant to AI in the sense that we might imagine how to instantiate a moral sense. As with other attributes of intelligence, we learn and have predispositions that are acted upon by the environment. Whether this would need to have a counterpart in AI is unclear.

I agree with your assessment completely. That's what I meant by 'Do we buy this?' You'd really have to look at the experimental design. I was worried about whether the babies were getting other non-verbal signals, whether the prompts were carefully controlled, etc...

I found another experiment they did demonstrating that toddlers will take action to keep adults from becoming hostile...

What I'm wondering is, could some of this research be transferred to the learning curve of AI - and how do you guard against bot abuse?

What I mean by that is the way Cleverbot users skew the bot's vocabulary, and those mean teenagers who turned the perky teenager bot into a grumpy racist...you could end up with a bot that had all the bad habits of its creators.

Hi folks! Just quickly want to turn your attention to this bit of AI ... hope you enjoy:

https://www.buzzfeed.com/andyneuenschwander/people-are-losing-it-over-this-computers-hilarious-attempt?utm_term=.hrRDbbx8e&ref=mobile_share#.eiKDJJ3eq

Sorry that is not clickable ... try this:

http://www.buzzfeed.com/andyneuenschwander/people-are-losing-it-over-this-computers-hilarious-attempt?utm_term=.hrRDbbx8e&ref=mobile_share#.eiKDJJ3eq

I want to see a chocolate pickle. Not eat one, just see one...

The last week or two I�ve been thinking about how to explain my views here in a way that makes more sense (to me as well as everyone else) and, as this conversation has considerably sharpened my understanding of what I�m talking about, I�ve decided to go back to basics and review the original question. Dmitri started it by asking �Would making a sentient AI prove that consciousness is purely epiphenomenal? And thus ephemeral? Of course not� I think the universal consciousness is a lot smarter than we baby consciousnesses think. I think there�s a Plan in the works.� What we seem to be looking for, then, is presumably a description of mind and an explanation of how it works, laid out in clear language that can be translated into maths in order for a machine to be programmed to produce recognisable thinking. My view is that life implies awareness (i.e. the daisies in my front garden are aware of light and dark as they open during the day and close at night) but that thinking, self-aware higher consciousness has evolved (and maybe continues to evolve) only recently and so far only humans exhibit it as only humans experience dilemmas about whether it�s right to eat meat, what sort of world their children should grow up in, whether to stand out as an individual or whether to blend in to society and so on. For me, this means that the mind must be studied historically and socially as well as through neuroscience. It would appear from the evidence available that about 100,000 years ago humans started creating symbolic art and ornaments. This may be considered as evidence of the beginning of a theory of mind but what happened in humans that they were able to move to this stage from the basic self-awareness (for example, very close physical attachment to a mother) demonstrated by other primates, elephants, whales and a few other species when, so far, these other species haven�t? Danny Brower suggested that becoming aware that others are equally self-aware, the requirement for understanding the intentions of others, also results in understanding the deaths of other humans and consequently the realisation of one�s own individual mortality. However, this would put an end to reproduction so, for a species to survive with that level of self-awareness requires that it simultaneously evolve a mechanism for denying individual mortality. For me this would possibly provide an explanation for the view that there is an underlying contradiction that drives the human mind. The historical development of the universalisation of the individual has been acknowledged in Western culture (and I�m following Musk here in concentrating on this aspect in order to make the point easier to understand) quite widely since Hegel, generally starting with the Greek parochialism of Socrates and then manifesting itself through the gradual development of Christianity. In the medieval Christian view, humans had a dual earthly and heavenly nature; the earthly, physical realm was the abode of the Devil and anyone such as Faust who attempted to investigate that realm was surely headed for a sticky end. Nature was abhorrent to the medieval mind; gardens, in order to demonstrate the mastery of heaven, had to be made symmetrical with straight lines and perfect curves. This, to me, suggests that individuals were still not fully able to recognise other humans as absolutely self-aware. Instead they recognised God as the self-awareness that defined them. What I think Luther and Descartes achieved was the externalisation of the earthly side of human nature. God became much more distant and less human and the human individual was able to imagine itself as detached from the physical realm. Nature became an object that we could look at in a detached way and manipulate for our benefit and other individuals replaced God as our recognised self-awarenesses but the underlying contradiction of mind also produced solipsism, that is, we can also see other individuals as simply objects to be manipulated in our world. It also allowed the Romantics to see natur

Well, that doesn't seem to have worked very well. I'll try again...

The last week or two I�ve been thinking about how to explain my views here in a way that makes more sense (to me as well as everyone else) and, as this conversation has considerably sharpened my understanding of what I�m talking about, I�ve decided to go back to basics and review the original question. Dmitri started it by asking �Would making a sentient AI prove that consciousness is purely epiphenomenal? And thus ephemeral? Of course not� I think the universal consciousness is a lot smarter than we baby consciousnesses think. I think there�s a Plan in the works.� What we seem to be looking for, then, is presumably a description of mind and an explanation of how it works, laid out in clear language that can be translated into maths in order for a machine to be programmed to produce recognisable thinking. My view is that life implies awareness (i.e. the daisies in my front garden are aware of light and dark as they open during the day and close at night) but that thinking, self-aware higher consciousness has evolved (and maybe continues to evolve) only recently and so far only humans exhibit it as only humans experience dilemmas about whether it�s right to eat meat, what sort of world their children should grow up in, whether to stand out as an individual or whether to blend in to society and so on. For me, this means that the mind must be studied historically and socially as well as through neuroscience. It would appear from the evidence available that about 100,000 years ago humans started creating symbolic art and ornaments. This may be considered as evidence of the beginning of a theory of mind but what happened in humans that they were able to move to this stage from the basic self-awareness (for example, very close physical attachment to a mother) demonstrated by other primates, elephants, whales and a few other species when, so far, these other species haven�t? Danny Brower suggested that becoming aware that others are equally self-aware, the requirement for understanding the intentions of others, also results in understanding the deaths of other humans and consequently the realisation of one�s own individual mortality. However, this would put an end to reproduction so, for a species to survive with that level of self-awareness requires that it simultaneously evolve a mechanism for denying individual mortality. For me this would possibly provide an explanation for the view that there is an underlying contradiction that drives the human mind. The historical development of the universalisation of the individual has been acknowledged in Western culture (and I�m following Musk here in concentrating on this aspect in order to make the point easier to understand) quite widely since Hegel, generally starting with the Greek parochialism of Socrates and then manifesting itself through the gradual development of Christianity. In the medieval Christian view, humans had a dual earthly and heavenly nature; the earthly, physical realm was the abode of the Devil and anyone such as Faust who attempted to investigate that realm was surely headed for a sticky end. Nature was abhorrent to the medieval mind; gardens, in order to demonstrate the mastery of heaven, had to be made symmetrical with straight lines and perfect curves. This, to me, suggests that individuals were still not fully able to recognise other humans as absolutely self-aware. Instead they recognised God as the self-awareness that defined them. What I think Luther and Descartes achieved was the externalisation of the earthly side of human nature. God became much more distant and less human and the human individual was able to imagine itself as detached from the physical realm. Nature became an object that we could look at in a detached way and manipulate for our benefit and other individuals replaced God as our recognised self-awarenesses but the underlying contradiction of mind also produced solipsism, that is, we can also see other individuals as simply objects to be manipulated in our world. It also allowed the Romantics to see natur

And maybe, like the man in Philip Dick's story 'Imposter', I am an android and don't know it. I just hope I don't explode.

Chris,

I'm not sure what's happened here, either to the narrative or the rendering. I hope you can clarify! There are some interesting resonances in what you've said, but a little tangential.

The last week or two I�ve been thinking about how to explain my views here in a way that makes more sense (to me as well as everyone else) and, as this conversation has considerably sharpened my understanding of what I�m talking about, I�ve decided to go back to basics and review the original question. Dmitri started it by asking �Would making a sentient AI prove that consciousness is purely epiphenomenal? And thus ephemeral? Of course not� I think the universal consciousness is a lot smarter than we baby consciousnesses think. I think there�s a Plan in the works.� What we seem to be looking for, then, is presumably a description of mind and an explanation of how it works, laid out in clear language that can be translated into maths in order for a machine to be programmed to produce recognisable thinking. My view is that life implies awareness (i.e. the daisies in my front garden are aware of light and dark as they open during the day and close at night) but that thinking, self-aware higher consciousness has evolved (and maybe continues to evolve) only recently and so far only humans exhibit it as only humans experience dilemmas about whether it�s right to eat meat, what sort of world their children should grow up in, whether to stand out as an individual or whether to blend in to society and so on. For me, this means that the mind must be studied historically and socially as well as through neuroscience. It would appear from the evidence available that about 100,000 years ago humans started creating symbolic art and ornaments. This may be considered as evidence of the beginning of a theory of mind but what happened in humans that they were able to move to this stage from the basic self-awareness (for example, very close physical attachment to a mother) demonstrated by other primates, elephants, whales and a few other species when, so far, these other species haven�t? Danny Brower suggested that becoming aware that others are equally self-aware, the requirement for understanding the intentions of others, also results in understanding the deaths of other humans and consequently the realisation of one�s own individual mortality. However, this would put an end to reproduction so, for a species to survive with that level of self-awareness requires that it simultaneously evolve a mechanism for denying individual mortality. For me this would possibly provide an explanation for the view that there is an underlying contradiction that drives the human mind. The historical development of the universalisation of the individual has been acknowledged in Western culture (and I�m following Musk here in concentrating on this aspect in order to make the point easier to understand) quite widely since Hegel, generally starting with the Greek parochialism of Socrates and then manifesting itself through the gradual development of Christianity. In the medieval Christian view, humans had a dual earthly and heavenly nature; the earthly, physical realm was the abode of the Devil and anyone such as Faust who attempted to investigate that realm was surely headed for a sticky end. Nature was abhorrent to the medieval mind; gardens, in order to demonstrate the mastery of heaven, had to be made symmetrical with straight lines and perfect curves. This, to me, suggests that individuals were still not fully able to recognise other humans as absolutely self-aware. Instead they recognised God as the self-awareness that defined them. What I think Luther and Descartes achieved was the externalisation of the earthly side of human nature. God became much more distant and less human and the human individual was able to imagine itself as detached from the physical realm. Nature became an object that we could look at in a detached way and manipulate for our benefit and other individuals replaced God as our recognised self-awarenesses but the underlying contradiction of mind also produced solipsism, that is, we can also see other individuals as simply objects to be manipulated in our world. It also allowed the Romantics to see natur

Sorry, I give up

I'm seeing it all in Brunel, was that the problem?

The last link goes to Pliny, and comes out blank.