[ExI] Did Hugo de Garis leave the field?

[Eugen Leitl]

On Wed, Apr 20, 2011 at 07:00:49PM -0700, Samantha Atkins wrote:

> Well, ok, partially.   To assume that even an AGI will not be smart  
> enough to self-improve on purpose seems like a stretch.

Let's say you have full write access to your genome.
How would you improve yourself?

An AI doesn't have any source code. It just has state,
plenty of it. Of course inspectable. But improvements
by monkeypatching a live system are hard. 

If you're not layered or modular, with clean interfaces,
how do you improve yourself?

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE