[extropy-chat] AI design

Zero Powers zero_powers at hotmail.com
Fri Jun 4 00:16:40 UTC 2004


>From: Eugen Leitl <eugen at leitl.org>
>
>On Thu, Jun 03, 2004 at 09:07:49AM -0700, Zero Powers wrote:
>
> > Hmmm.  I still don't get it.  Even if we are insignificant by comparison 
>I
> > still don't see *why* an AI would "turn us into lunch."
>
>Because you're made from atoms, just as the landscape. You are a part of 
>the
>environment, a raw resource, about to be strip-mined.

Yes, so I've been told.  My only question is why?  Why would the AI want to 
"strip mine" me or turn me "into lunch" or thrust me into the "whirling 
razor blades."  These descriptions of what the all-powerful AI is going to 
do to me (unless an exponentially weaker intelligence like Eli's can trick 
it into being friendly) all sound pretty scary.  But I guess the reason I 
don't feel scared is because so far I haven't heard any convincing 
explanation of why the AI will be motivated to be such a bad neighbor.  I've 
heard:

1.	You're so insignificant the AI will rip you atom from atom before it even 
realizes you cared;

2.	The AI will be programmed to statically seek ultimate "utility" which 
means reconstituting your brain cells to a state of euphoria, and leaving 
you stuck there; and

3.	Just like in _The Matrix_, the AI will use (a) your brain for backup 
storage and/or (b) your atoms for energy.

Those arguments seem laughable to me.  I could go on for a few more 
paragraphs explaining what I see as the ridiculousness of those arguments 
but (1) the explanations should be self-evident and (2) those who don't see 
the counter-arguments inherent in the above reasons would probably not see 
them after my explanations.  So, I guess where I'm at is: does anyone have a 
reason (other than on the order of the above 3) that I should be afraid?  If 
so, I'd be interested to hear.  If not, I guess I'll politely drop out of 
this thread now.

I'm almost tempted to ask how we plan to insure Friendliness in an AI that 
(1) will have God-like intelligence, and (2) will be self-improving and 
doing so at an exponentially accelerating rate.  Somehow this rapidly 
self-improving, God-like super-intelligence will be able to optimize itself 
(hardware *and* software), solve all the world's problems in a single bound 
and propagate itself throughout the entire universe, yet will *never* be 
able to alter those lines of its code that compel it to be Friendly to 
humans?  No, don't even bother to respond.  Believe me, I won't understand.

Take care
Zero

_________________________________________________________________
Stop worrying about overloading your inbox - get MSN Hotmail Extra Storage! 
http://join.msn.click-url.com/go/onm00200362ave/direct/01/




More information about the extropy-chat mailing list