Re: Implementation

From: Hans Moravec <hpm.domain.name.hidden>
Date: Tue, 27 Jul 1999 22:50:45 -0400

SLP <SLP.domain.name.hidden>:
> I certainly wouldn't because that's not remotely how brains,
> the only known conscious systems, work. The lookup table route is
> one way to produce a zombie, meaning: a Turing test passer that has
> no internal qualia.

I dare you to demonstrate to me that a HLT isn't exactly as
conscious as any other implementation of the same behavior.

You can't, of course, and I reject your ridiculous pronouncement.


> Consciousness means having a self, having qualia. It has nothing to
> do with behavior. People with the locked-in syndrome (a very tragic
> and usually quickly fatal neurologic condition) are conscious,
> although you wouldn't know it from their behavior.

Then how do you know it? By behavior!
Eye blinks, physiology changes and EEG traces ARE behavior.


> Until we understand the actual neural machinery of brains, and
> generalize its abstract principles to nonbiological systems, we
> won't know how to build conscious robots.

NONSENSE! Robots will be telling us about their inner lives,
with clear sincerity, with or without the benefit of brain
studies. They will be conscious by virtue of their internal
models of self and surroundings.

I'm happy for your neuroscience, but don't presume it to be on
the critical path of robot development. Robots are evolving
in their own unique way.
Received on Tue Jul 27 1999 - 19:53:08 PDT