Star Trek

What do you think about AI? Some time ago I was interested in the subject; I've read some texts concerning artificial neural networks and I've studied quantum computers in university. What is unclear is what exactly will make artificial neural network possessing consciousness. I guess no one knows the precise answer, but you are free to tell your opinion.

I definitely fall into the category of not knowing a precise answer.

The question does remind me of previous debates I vaguely recall about whether non-human animals can be said to have "consciousness".  Many claimed that animals were "just" machines, and therefore could not be said to be conscious. 

I think the answer to both questions very much depends on what you define consciousness to be. Absent a precise definition, you are unlikely to get a precise answer.

4_o_20, yeah, I rememeber your mentioning of that movie. Indeed, sounds interesting. Complexity certainly have some role.

Caltrek, I'm in the same category . I've heard question concerning animals too. On our philosophical lectures in university approximately a year ago professor told about Marxism theory - that the work creates consiousness. So, it agrees with materializm - that materia (here it is some random act of work, labour) is first and consiousness is a product of materia. It goes well with Space Odyssey 2001, where in the very beginning of the movie monkey randomly breaks one bone with the help of the other and it starts its thinking process. Well, if you skip The Monolith affection, haha. On exam when I was asked the quesion of what is the prime - materia or consiousness - and why, I tried to follow the lines of Marxism theory, because that's actually what our professor believes to be a supreme theory (and hence that is what you are to tell if you want to get A), but told it in some unclear way, so our professor at first decided that I believe that consiousness is prime and animals possess it, haha. When I went out of the class-room I've told it to my friend, and he said Well, then I guess you actually belive that animals have consiousness.

I actually don't know. I do realize that humans do some work in unconsious way - you can go somewhere while do not realizing it, etc. So, you use consiousness when you realize that you take some decision, that is when you choose among alternatives (or admit that there're no of them). There's a theorem by Conway, that electron being a quantum object has a free will in exactly the same sense as human has it (that is a rough formulation, I don't remember exactly). So, suppose that evolution of some closed (separate, completely isolated from enirinment) system is purely quantum. Then the system is in a superposition of states - like an electron going through two rifts at the same time. Let's call such isolated evolution of the system unconsious. It's reasonable, because such evolution is completely predicted - in quantum-mechnical sense, you use Schroedinger equation and obtain precise trajectory for such superposed state evolution. But if the observer appears, he looks at the system and interacts with it. The system is not closed any more. Moreover, it's not in a superposition of states anumore: when observer looked at it, he collapsed the superposition of states into one of the superposed sates. That is a consiousness process - you break predictable evolution of closed system by probabilistic random collapse. You (or rather the system) - acquires a free will - it "chooses" to collapse in some particular state.

That speculation also goes along lines of Penrose in his Emperors.

Funny, but no quantitative results and such wild speculation makes me sad.

The Master: Caltrek, I'm in the same category . I've heard question concerning animals too. On our philosophical lectures in university approximately a year ago professor told about Marxism theory - that the work creates consiousness. So, it agrees with materializm - that materia (here it is some random act of work, labour) is first and consiousness is a product of materia. It goes well with Space Odyssey 2001, where in the very beginning of the movie monkey randomly breaks one bone with the help of the other and it starts its thinking process. Well, if you skip The Monolith affection, haha. On exam when I was asked the quesion of what is the prime - materia or consiousness - and why, I tried to follow the lines of Marxism theory, because that's actually what our professor believes to be a supreme theory (and hence that is what you are to tell if you want to get A), but told it in some unclear way, so our professor at first decided that I believe that consiousness is prime and animals possess it, haha. When I went out of the class-room I've told it to my friend, and he said Well, then I guess you actually belive that animals have consiousness."

Thank you for this great illustration of my point. Here you had a professor who had rather tautalogical definition of consciousness: "something that is created by work".

caltrek: Yet you have people like me, who tend to believe that non-human animals are "conscious". The problem is that when I say that I have a somewhat different definition of consciousness than did uoyr professor. 

The Master: "There's a theorem by Conway, that electron being a quantum object has a free will in exactly the same sense as human has it"

caltrek: I have a hard time believing that, at least by any definition of consciousness that I would normally use.

The Master: "So, suppose that evolution of some closed (separate, completely isolated from enirinment) system is purely quantum. Then the system is in a superposition of states - like an electron going through two rifts at the same time. Let's call such isolated evolution of the system unconsious. It's reasonable, because such evolution is completely predicted - in quantum-mechnical sense, you use Schroedinger equation and obtain precise trajectory for such superposed state evolution. But if the observer appears, he looks at the system and interacts with it. The system is not closed any more. Moreover, it's not in a superposition of states anumore: when observer looked at it, he collapsed the superposition of states into one of the superposed sates."

caltrek: Sorry, I am not sure I follow all that.

The Master: "You (or rather the system) - acquires a free will - it "chooses" to collapse in some particular state."

caltrek: Ah yes, the old philosophical problem of free will versus necessity dressed up in quantum clothing.

The Master: "Funny, but no quantitative results and such wild speculation makes me sad."

Not quite sure why it makes you sad, but it is a problem of the human condition. How much free will do we really have?

Watson, the recent successful computer contestant on Jeopardy, has greatly excited the imagination of the pastor of a local church that I attend. It was the subject of his sermon this last Sunday.

After the service, my wife and I were discussing the sermon. My wife then blurted out what I thought was a brilliant question: Would a "conscious" computer know the difference between good and evil?

Which takes us back to Genesis in the Bible. Remember? Where Adam and Eve eat from the tree of knowledge and become"like gods, knowing good from evil". To me, that is the intertwining of consciousness and labor. When we come to know, or at least have beliefs, that there is such a thing as good and evil (and not all philosophical systems buy into that notion) then the work starts. For if we "know" good, and we wish to remain "good" then we must continually "work" to do the right thing. We must fulfill our "duty". In today's society, that is an overwhelming challenge, for none of us is perfect. So if we fail to do good, then we have "sinned", and must at least begin to feel that we need to be forgiven for our "sins".

Any reactions from anybody reading this?

I would rather prefer to look at all this problematic questions from mathematical point of view. It's simply and most likely will give you some (maybe modest but) concrete result, which you know for sure is true. Anyway, you may look at Conway theorem here . It's mentioned there that free will is defined by undetermined outcome. This is rather concrete and sensible definition, by the way.

Eveything which is beyond such mathematical description looks for me as purely sociological problem; not the problem of fundamental science. That is sociological problem always depends on concrete conditions created in human society and may be varied without changing of fundamental nature of elementary acts of free will. The later is more interesting for me, because it shows what is possible and what is not on fundamental level. And this later doesn't make me sad .

From the link provided above on Conway's theorem:

"The definition of "free will" used in the proof of this theorem is simply that an outcome is "not determined" by prior conditions, and some philosophers strongly dispute the equivalence of "not determined" with free will. "

Yes, I think I would tend to agree with those philosophers that "not determined" does not equal "free will".

Having said that, I have to admit that I do not understand the mechanism (assuming it even exists) in which "not determined" morphs into "free will" at the macro level.

It may even be possible that human beings have not reached that level yet, that our "free will" is just an illusion. Our decisions may very well be pre-determined by our history. Paradoxically, the very notion of "free will' has become a part of our history, and can affect the out come of our decisions. Whether or not we do have free will, we can act as if we did, and therefore take responsibility for our actions.

Very good. Again, at least we have some concrete definition of free will, so you can mathematically operate with it and prove some facts concerning this definition. You see, it's really hard, always impossible for specific real-world model, to describe what's going on around there with 100 percent accuracy. That's why scientists prefer to buid models to have concrete results. However, some of these models are exactly 100 percent correct, Yes . For example - quantum mechanics. It's mathematically proved that any more fundamental theory must respect quantum mechanics totally. Now I'm gonna tell you what philosphers are trying to do (most of them). They pose an extremelly ambitious aim of working out the very fundamental principle of everything. This is impossible, needless to say. However, that's the reason why they are always dissapointed with the concrete results which scientists obtain. For example, again, after quantum mechanics was constructed, W. Heisenberg and K. Weizsacker had problems with talking to philospohers about validity of their theory. In particular one lady was very dissapointed with limitations which QM in principal sets for measurement ability.

When you pose such ambitious goal and still neglect concrete working models, you are very easy to be delusioned. Philospohy is only good when it serves some real science as aid for conveying of scientific results to everyday language and for everyday purposes. Such translation is almost always flawed and is certainly not useful as scientific method.

Not just philosophers, but theologians as well. This "fundamental principal of everything"  reminds me of the desire to believe in absolute truths. Einstein's theory of relativity led many to complain about ideas of "moral relativism". While many Christians didn't have a problem reconciling their beliefs with this proposition concerning physics, many others felt threatened by this lack of absolutes.

Relativity and quantum mechanics also moved us away from understanding the universe in the way in which Newton saw it. Instead of crisp laws of mathematical power and precision, humanity suddenly found itself confronted with fuzzy calculations of probabilities and uncertain outcomes.

Getting back to the orignal subject of the thread, this may pose a distant problem for humanity.  Just what "rights" should a future artificial intelligence that is credited with "consciouness" enjoy?

Right now, that is an interesting topic to speculate about.  I think by the time it becomes a really existing problem, I will most likely be long dead, so it is not a problem I exactly lay awake at night worrying about.

There is also the feature of how intelligent will such artificial creations become?

They are already beating humans at chess and at Jeopardy. To what extent will we turn over basic decision making responsibilities to such entities?

Will we become like the Borg, creating a scoiety that is not quite fully human anymore?

All of these are great themes for science fiction to play with.

Caltrek, first of all I want to correct you about "fuzziness of quantum theories". I can assure you that quantum theory (at least untill you move to quantum gravity,  but even there there is robust mathematical formulation of superstring theory) is as crisp as Newtonian mechanics; and it is also much more fundamental and incorporates Newton as its classical large-scale llimit. Uncertainty principal is fundamental principle of Nature; not just some drawback of present theories. Every single theory which will ever be derived must 100% respect uncertainty principle.

Well, I think that robot of Data type must have all human rights without exception. I do not see any logical contradiction which it may bring; but I see tons of them for the opposite case.

Haha, I do not worry about it at nights, actually at any other time, too.