From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Tue May 29 2001 - 17:08:04 MDT
Ben Goertzel wrote:
>
> > It is more than mental limitations. And it is more than a limbic system
> > in conflict with the cerebral cortex. It is rationalization. It is a
> > limbic system that can interface with the cerebral cortex and twist our
> > thoughts, and not just our emotions, to adapted purposes. And that isn't
> > really an exclusively limbic thing, either; it's built into our cognitive
> > architecture. From the perspective of a normative reasoning system,
> > humans, who "want" something to be true - not true in the future, but true
> > in the present - and who make up reasons to believe something they "want
> > to believe" - well, they're pretty much insane.
>
> Eli, I don't disagree with your specific statements about human nature, I
> just find it peculiar to apply terms like "insane" or "ill" to what is
> normal for our species. Dogs aren't mentally ill people and we're not
> mentally ill AI's. There are many different kinds of creatures now, and
> will be even more kinds in the future.
"Insane" is probably the wrong term, medically. But in the intuitive
sense of the word, it looks to me like it's quite possible for an entire
species to be insane by its own standards. As an extreme example, if it
turns out that our entire species has some unthinkable blind spot
analogous to anosognosia, we would probably regard this as insane in
general terms - even though we might instantly forget, or rationalize, or
make excuses, on being told (probably by an AI) about the specific defect.
-- -- -- -- --
Eliezer S. Yudkowsky http://singinst.org/
Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:07:50 MST