[alicebot-aiethics] FW: [alicebot-general] ZOMBIES

[Christopher Fahey [askROM]]

Richard Wrote:

> Turing disposed of the "Lovelace objection" in his 1950
> paper. 

I just re-read the paper - for like the 5th time. Turing's "Computing
Machinery and Intelligence" is worth reading every few months or so just
to see how cogent, thorough, and prescient it is. I can't punch a hole
in it. And everything I've read where someone tries to punch a hole in
it seems to use crazy circular- and overly-complex logic... or worse,
relies on intangible, even mystical ideas of what consciousness is.

The most widely-accepted counterarguments appear to be of the "Argument
from consciousness" and of the "Lovelace" variety (which Turing himself
seemed to think was the counterargument most worthy of his attention):
Machines can't possibly think, it says, because machines cannot
originate ideas. Original ideas are a by-product of consciousness, which
a machine cannot have (how's that for circular logic?).

What I really appreciated about Turing's paper is his forthrightness
about where he stands personally. He believes that computers will win
his challenge, and that they will, in fact, think. He writes: 
     "The popular view that scientists proceed inexorably 
     from well-established fact to well-established fact, 
     never being influenced by any improved conjecture,
     is quite mistaken."

What strikes me about his naysayers is this: I really can't tell where
they stand on the nature of consciousness. They don't tell us where they
fall on a core aspect of the definition of consciousness. It seems that
you can only be in one of two camps: 

1) Consciousness is a real phenomonon, inexplicable by the laws of
physics (and very likely unique to humans).
2) Consciousness is not a real thing, but rather a perceived effect of
an enormously complex mechanism. It is not necessarily restricted to
human minds, or even to biological minds.

Camp #1 strikes me as a magical or spiritual belief. It's fine if that's
how you see things, but you should at least say so so that people like
me can know where you're coming from. Me, I'm with #2.

I'm very interested in the concept of ZOMBIES: A "Zombie" (in the
AI-naysayer argument) is a creature/being whose outward behavior has all
the trappings of consciousness - it walks, it talks, etc. But it's inner
self is an empty place where it doesn't even know what it is doing or
why it is doing it. It has no understanding of its own actions. The
ZOMBIE argument says, essentially:
     "Okay, so you've made a robot that passes the Turing 
     Test (or even what is called the "Total Turing Test": a 
     robot that is physically and behaviorally indistinguishable 
     from a person but who is nonetheless artificial, 
     engineered and built by humans.). But it does not have 
     any awareness of what it does, so it is not really
     conscious the way we humans are.

My counterargument to this is twofold:

1) As Turing himself notes, there is no way of knowing the answer to
this even with regards to other human beings - what goes on in someone
else's mind is a (forever?) unsolvable mystery.

2) Who is to say that real live humans even have a consciousness where
we know what we are doing and why we are doing it, or that such an inner
experience is all that much different from that of an ape or a pig? My
mind is constantly doing things that I didn't necessarily ask it to do -
I don't ask to fall in love or get angry, I just become in love or
angry. We are beginning to understand the enormous degree to which an
individual human's behavior is governed by often-volatile chemical
activity and fragile psychological mechanisms - it seems inevitable that
neurological research will eventually figure out the exact nature of
friendship, decision making, contentedness, love, jealousy, thrills,
fear, and all of the other things that humans think are somehow
intangible aspects of human consciousness. Perhaps what we think of as
self-awareness is simply an illusion, and that in fact we are 100%
driven by explainable electrochemical forces, i.e., a Turing
state-machine.

Anyway, my real question is: Shouldn't AI theorists be required to
choose which of the above two definitions of consciousness they
subscribe to? I think many AI-naysayers subscribe to #1 (Searle, for
example) but refuse to admit it since it is, at its core, utterly
unscientific.

-Cf





_______________________________________________
alicebot-general mailing list alicebot-general@list.alicebot.org
http://list.alicebot.org/mailman/listinfo.cgi/alicebot-general