On Wed, Dec 26, 2012 at 9:14 PM, Logan Streondj <[email protected]> wrote:
>> > Best way of proving AGI in an undeniable fashion, is to have "wild" AGI's 
>> > robots running around self-replicating in the environment. Admitedly most 
>> > wild organisms have brains significantly larger than those of their 
>> > domestic counter-parts, so we may develop domestic AGI's first.
>>
>> No, that is the best way to wipe out humanity.
>
> seriously that's just fear mongering.
> Computers have a different ecological niche then that of biology,
> they are better suited for cold dry environments, like polar deserts.

I'm referring to the threat of self replicating nanotechnology.
http://www.foresight.org/nano/Ecophagy.html

Do you think this is a serious threat?

We can design self-replicators for whatever niche we want. I am
concerned in particular that we can already produce solar cells with
10% to 20% efficiency, but plants currently convert only 0.1% of the
sunlight that reaches the Earth into chemical energy. It would seem
that an accidental release of nanotechnology could easily out-compete
plants, thus removing the bottom of the food chain, even if it is not
their intention to kill us.

>> And no, self
>> replication + control is harder than self replication alone. It
>> requires more intelligence.
>>
> that's plain false, as demonstrated by this image
> http://sensualanimist.files.wordpress.com/2012/05/picture-0041.jpg
> domestication of animals decreases brain size by about 10-30%

That's interesting. I didn't know that, but I can confirm it.
http://link.springer.com/article/10.1007%2FBF02436333?LI=true

And that seems to be bad news for our hopes of controlling AI.


--
-- Matt Mahoney, [email protected]


-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com

Reply via email to