Wednesday, June 10, 2009

6c Can my computer think?

As promised and on Samuel's request we'll address the problem of "Other Minds" today. As such it is already a serious philosophical issue, but in relation to artificial intelligence, it even becomes a more serious problem.

In the former lecture I presented three arguments. Let me repaet them:

1. Thought is some kind of computation.
2. Digital computers can perform all possible computations.
3. Digital computers can think.

1. Thought is some kind of conscious experience.
2. Machines can't have conscious experiences.
3. Machines can't think.

1. Thoughts are specific biological brain processes.
2. Artificial computers can't have biological brain processes.
3. Artificial computers can't think.

As you see, all depends on the definition of thought. Only reasoning A. will confront us with the problem of other minds in relation to machines.

I suppose we can agree at least on one thing: a thought is a mental state, a mental state we are aware of. So eventually we should have to admit that computers can have mental states.

A thing often seen in Science Fiction movies and books. A nice fellow in this category is Data from the Enterpriseor in HALL from Space Odyssee 2001.

We have an other matter to deal with: intelligence. Could we at least define it as a quality level of thinking, where unintelligent and intelligent behavior differ in the quality of how for instance a problem is solved?

So, a mind in the computer of the future? An other mind, not my mind. Of course in our daily life we are certain that other human beings have their own minds, like I have mind.

But we assume a lot of things and it works well, till we approach all these assumptions philosophically. Then we have to admit that the alleged certainty is questionable.

In this situation what it is all about is:
how do we justify that certainty we have of knowing that other people (and in the far future maybe humanoid robots) have their own minds ?

In introducing "methodic doubt" into philosophy, Descartes created the backdrop against which solipsism subsequently developed and was made to seem, if not plausible, at least irrefutable.

Solipsism is the theory that the only thing we are certain of is the content of our own mind and that our knowledge of our mental states is private. Noone but me can experience for instance my headache.

But it works also the other way around. Neither can I experience the mental states of another person.

What then of my knowledge of the minds of others? On Locke's view there can be only one answer: since what I know directly is the existence and contents of my own mind,

it follows that my knowledge of the minds of others, if I am to be said to possess such knowledge at all, has to be indirect and analogical, an inference from my own case. This is the so-called "argument from analogy" for other minds.

I wont bother you with all arguments against this argument from analogy, but one is interesting. If you apply scientific standards and you observe here,

that we make a generalisation based on only one observation, the observation of my own mind, you can imagine that this argument from analogy isnt a strong one epistemologically.

Assuming that the argument from analogy is unacceptable, the most obvious alternative is to adopt some form of that variety of behaviorism according to which all psychological expressions can be fully understood in terms of behavior.

We have seen that before..... a computer shows the same behavior as a human, so we may call it intelligent. But does a translation of all psychological expressions in terms of behavior do the job?

That means does it offer the epistemological justification to claim with certainty, that my computer can think? Of course, we'll come up with the next criticism.

It is implausible to give a behavioristic account of some first-person psychological statements. When, for example, I say that I have a terrible pain, I do not say this on the basis of observation of my own behavior and the circumstances in which I am placed.

How far did we get today? Will there be a moment in the far future that I ascribe a conscious mind to a computer, or say at least that my computer thinks?

If you look at the epistemological problems of "Other Minds" I am not so sure. We are not even capable to come up with a univocal solution for the "Other Minds" issue. And that problem wasnt raised because of artificial intelligence, but by the observed intelligence in my fellowmen.

The Discussion

herman Bergson: So much on AI
Gemma Cleanslate: it does make one wondr about the future tho
herman Bergson: In what way Gemma?
Gemma Cleanslate: what science will come up with in the area of computers
Gemma Cleanslate: and how much they will progress
Gemma Cleanslate: toward an AI
Gemma Cleanslate: it is all math
Ze Novikov: electro/biologic human computer interface
herman Bergson: Well...the AI community came up with high expectations....
Gemma Cleanslate: yes ZE
Paula Dix: ive been talking with my friends here about that, and we came with an idea
herman Bergson: But they sized down the big dreams
Ze Novikov: computer will augment our physical processes
Paula Dix: if we find some alien with ships and technology we will be certain they can think, even if probably in ways different from ours
Paula Dix: in the same sense, when we have computers that have a level of processing "thinking" similar to ours, we will accept that they think
Ze Novikov: we have it in a crude form with artifical limbs
Paula Dix: even if we have no idea what thinking is
herman Bergson: There definitely will be a computer/central nervous system interaction in the future
Ze Novikov: yes
herman Bergson: But that is something different from AI
Paula Dix: lol thats the idea of my science fiction lover friend, the computers will be us :)))
Zen Arado: what about self awareness?
Zen Arado: wouldnt that be a criterion?
herman Bergson: Yes indeed Zen....there you run into fundamental philosophical problems again
herman Bergson: Hume even didnt discover a Self in himself
Zen Arado: yes there isnt one to be aware of I guess :)
herman Bergson: So self awareness is already questioned by....what are you aware of
Cailleach Shan: Are we assuming that 'computers that can think' are a bad thing?
Paula Dix: like that Denett text about mind and gravity center
Paula Dix: (if anyone want this text, i have it here)
herman Bergson: Well Cailleach.....computers that can would create serious ethical problems for instance..
herman Bergson: Is a thinking computer a new species?
Paula Dix: i believe they think already and that this is good :))))
Alarice Beaumont: i think they do that right now
herman Bergson: Does ethics apply to it or may I turn it off at will for instance..or is that murder?
Zen Arado: we are afraid they wouldnt share our moral sense
Alarice Beaumont: but only in the meanings of the person which program them
Gemma Cleanslate: or throw it out the window even worse
Ze Novikov: yes
Zen Arado: that was the prob with Hal :)
Paula Dix: i guess its not unethical to turn them off, since they are designed to do so, it wont damage them like turning us off
Cailleach Shan: That was my next question Herman. Can computers turn themselves back on?
herman Bergson: Ever heard of the Laws of Robotics, formulated by Isaac Asimov?
Zen Arado: yes
Paula Dix: oh yes im reading Cave of Steel series :)))
Cailleach Shan: Nope
herman Bergson: Cool
Zen Arado: old sci fi
Zen Arado: :)
herman Bergson: Yes it is Zen...but so is the bible and people still read it :-)
Cailleach Shan: lol Bible Sci fi
herman Bergson: I mean the 'old' feature...
herman Bergson: not the SF
Zen Arado: sure ...but reminds me of my youth
Zen Arado: :)
Paula Dix: first book is dated, but ok if you read it like a steampunk thing, but second is good
herman Bergson: too
Zen Arado: wasnt criticising it he came up with good ideas I think
Cailleach Shan: Re. the 'ethics' question. I think we have made up ethics ourselves, so eventually we would make up a new set to incorporate thinking computers.
herman Bergson: Well he made the robot in fact a servant of men
herman Bergson: not an autonomous being
Gemma Cleanslate: and they are supposed to be
Zen Arado: yes they mustnt harm a human
Zen Arado: 1st law?
herman Bergson: For instance...
herman Bergson: Yes
Paula Dix: ah, yes, then there is the bicentenary man book/movie, also by asimov, discussing when they turn into men
herman Bergson: So what when an intelligent computer is regarded as an autonomous being?
Zen Arado: when they have emotions ?
herman Bergson: yes....even more complicated
Cailleach Shan: We have too many autonomous beings on the planet already.
Paula Dix: its like to discuss animal rights
herman Bergson smiles
Zen Arado: its the lack of emotion that scares us maybe
Paula Dix: a chicken can think?
Paula Dix: they have emotions...
Ze Novikov: chickens?
Cailleach Shan: What is scary about lack of emotion?
Paula Dix: lol any animals
Paula Dix: just got some down on the scale
Alarice Beaumont: one might decide wrong
Zen Arado: lack of compassion
Alarice Beaumont: no emotion.. no feelings .. no ethics
herman Bergson: Well.....I think, that we'll never see autonomous computers as independent beings
Zen Arado: you couldnt make up rules to substitute for lack of compassion
Zen Arado: I dont think
herman Bergson: I even dont believe in a thinking computer
Paula Dix: herman so you think thinking is a biological only thing?
herman Bergson: So I would answer the question Can my computer think? with NO
Gemma Cleanslate: hmmmm
herman Bergson: is a biological thing primarily
Paula Dix: ok
Zen Arado: trouble is we dont really know what thinking is you said earlier Herman
Paula Dix: yes, maybe computers are other way of thinking
herman Bergson: Exactly Zen.....we even cant find the right answers on our own philosophical questions
Paula Dix: and i believe we need an ethics toward them, same as toward other animals
Gemma Cleanslate: ow wow
Paula Dix: just in case :))
herman Bergson: I just turn it off at will Paula ^_^
Alarice Beaumont: lol
Ze Novikov: lol
Paula Dix: oh, but that is ok for them
Gemma Cleanslate: i think i will leave mine on from now on lo
Zen Arado: you cant turn off the internet though :)
Alarice Beaumont: lol
Paula Dix: they will even ask to be turned off from time to time :))))))))
Gemma Cleanslate: just in case
Zen Arado: people are worried about that
Alarice Beaumont: ah.. but you can quit the connection!
Gemma Cleanslate: well if it asks ok
Paula Dix: lol Gemma
herman Bergson: Yes Gemma and next morning it will say I think, so I am :-)
Gemma Cleanslate: lol
Ze Novikov: lol
Gemma Cleanslate: yes
Paula Dix: My friend Kore, who was here some times, is a neuroscientist and he is working with AI systems
Alarice Beaumont: does he say that it's possible to make machines think?!
Zen Arado: yes Paula ?
Paula Dix: he has this problem, if he can or not turn the AI that has emotions off
herman Bergson: Yes there are AI systems....
Cailleach Shan: Now there's an interesting thought... if we are 'turned off' from existence are we still there somewhere waiting to be turned back on.
Paula Dix: he says they can think and are alive
Gemma Cleanslate: ah
Gemma Cleanslate: too bad he is not here now
Paula Dix: or at least these ones he is making :))
herman Bergson: If we were computers with a powerswitch, Cailleach
Alarice Beaumont: isn't that called coma?
Paula Dix: i asked him to come, but he couldnt... :(
Gemma Cleanslate: too bad
Ze Novikov: the pod people
Zen Arado: keeps coming back to the personal identity issue doesnt it?
Gemma Cleanslate: yes it does
Gemma Cleanslate: always
Paula Dix: he is developing AIs that can learn, and thats a big step toward awareness i guess
Zen Arado: is there an 'I' in the computer
herman Bergson: The AI systems are only applicable in a limited world
Zen Arado: or in us even
Cailleach Shan: It's a very good way of looking at our fears around 'control'.
herman Bergson: Oh..Personal Identy...... an other headache chapter in philosophy
Paula Dix: lol yes Caill and all these people that believe computers can think also say they will be better than us in some time
Paula Dix: well maybe not all :)))
herman Bergson: Well...We dont need to be afraid of thinking computers in our lifetime
Paula Dix: meanwhile, police and students and teachers confrontation on the main university here... :((( i wonder if they can think
Cailleach Shan: mmmmm.... not too sure about that Herman.... look how quickly technology advances.
Gemma Cleanslate: that is true too
Samuel Okelly: :)
Gemma Cleanslate: so so fast in computer world
Paula Dix: yes, most people lost jobs to computers already
Paula Dix: many not most
Gemma Cleanslate: have to go now
Gemma Cleanslate: bye!
Paula Dix: bye!
Cailleach Shan: cu Gem.
herman Bergson: Bye GEmma :-)
Zen Arado: bye Gemma
Alarice Beaumont: bye Gemma :-))
Samuel Okelly: tc gem
herman Bergson: I think we may conclude our session on this question and move on to the next question
Ze Novikov: ty herman
Ze Novikov: :))
Zen Arado: is/was interesting Herman
herman Bergson: Interesting...... Believing and reasonable..:-)
Ze Novikov: bb everyone until next week :))
herman Bergson: thank you
Zen Arado: bye Ze
herman Bergson: and thank you all for your participation
Cailleach Shan: Nice one Herman..... me and my computer thank you.
Paula Dix: yes very nice discussion :))))
Samuel Okelly: apologies for being so late herman :(
herman Bergson: Give my reagrds to your computer, Caileach...:-)
Cailleach Shan: lol ta.
Paula Dix: our computer is turning itself off at random... we need to call the doctor!!
herman Bergson: Things happen, Samule :-)
Alarice Beaumont: lol
Cailleach Shan: Bye all.
Samuel Okelly: ill look forward to catchinh up with the transcript
Samuel Okelly:
Samuel Okelly: †
Samuel Okelly: † (( take care everyone )) †
Samuel Okelly: †
Samuel Okelly:
Zen Arado: I have to go too
Alarice Beaumont: oh.. need to go
Zen Arado: bye
herman Bergson: Ok Samuel....will be in the blog soon
Alarice Beaumont: i will call you for a game of chess Herman ;-)
Alarice Beaumont: everyone have a nice nite / day :-))
Paula Dix: bye
herman Bergson: Ok...:-)
Alarice Beaumont: bye
Paula Dix: back to building...
herman Bergson: Bye Alarice
herman Bergson: Ok Paula
herman Bergson: Happy building:-)

No comments:

Post a Comment