On 4/25/07, Richard Loosemore <[EMAIL PROTECTED]> wrote:

Benjamin Goertzel wrote:
>
> I agree with you, Kurzweil and others that once the brain is thoroughly
> understood, this understanding can be used as the basis for AI designs.

This is a red herring ;-)

I did not say anything about understanding the *brain*:  only the human
cognitive system.  Those are two completely distinct projects.



I think they are closely tied.

I was involved enough with experimental psych, when I worked in a psych
dept. for 2 years, to see how incredibly hard it is to gather useful data
about
the dynamics of human cognition via doing psychology lab experiments.

My bet is that we are going to understand the dynamics of human cognition
thoroughly only when we have really accurate brain scanners...


Waiting
for people to "understand" the brain is pointless:  they will never be
able to do it unless they can understand a good deal about the cognitive
system first, because without the latter they will not know what they
are looking for, or how to interpret what they find.



I think understanding of brain and mind are going to emerge together,
closely coupled...


I disagree that close emulation of human intelligence is the only path
> to AGI, though...

You did not say why, though.

Our previous discussions of the same question always ended with you
declaring that you simply do not believe that it will be a big problem,
as I recall.

If, at any point, you have any new reasons for disbelieving the argument
that we *must* follow the human emulation path, please get back to me
asap, because I can only refine the argument if I get concrete feedback.



At this time I've chosen not to publicly release my AGI design and the
detailed
reasoning underlying it....  This sort of annoys me, in that I would love to
argue with you and others in depth about the details of Novamente, but it
still
seems to me the wisest course for a couple reasons.

-- Ben

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936

Reply via email to