From: Ben Goertzel (ben@webmind.com)
Date: Tue Jun 12 2001 - 16:40:35 MDT
Hi Peter,
It's all quite sensible, and of course to me rather elementary
I do take issue with one peripheral point. You say
"Working with biological cell/ axon/ receptor structures essentially
requires advanced biological nanotech" but I'm not sure. I think that
extreme human intelligence enhancement may be possible via genetic
engineering. If we can unravel the metabolic pathways by which the genome
yields the proteome, which yields brain cells, neural networks, etc., then
we may well be able to engineer smarter brains. It's the same principle as
gene therapies for curing cancer (which admittedly are only in an early
stage now).
One possible course of development is then:
1) AI's analyze genetic & cellular data, thus helping figure out how to make
humans smarter
2) the genetically enhanced humans, working together with the AI's that
helped create them, help create yet smarter AI's, humans, and AI-human
bio-silicon-computing hybrids
I personally tend to think that AI's are going to get pretty smart before
genetic engineering hugely improves human intelligence, so I largely agree
with you. But I do consider it likely that when our seed AI gets really
clever, it decides the best way to further enhance itself involves
*biocomputing* as well as silicon computing. DNA, proteins, brain cells and
the like are pretty powerful devices, which may be deployed and modified
creatively by an AI system that is smarter than humans, but not yet powerful
enough to automagically restructure all the molecules in the solar system.
This I consider part of my "moderately paced Singularity" scenario, as
opposed to the hyperspeed scenario that is apparently favored by Eli, in
which the path from human-level intelligence to sysopmind-level-intelligence
occurs during a single episode of the Jetsons...
Ben
> -----Original Message-----
> From: owner-sl4@sysopmind.com [mailto:owner-sl4@sysopmind.com]On Behalf
> Of Peter Voss
> Sent: Tuesday, June 12, 2001 5:22 PM
> To: sl4@sysopmind.com
> Subject: Extro5 talk - feedback requested
>
>
> I'd appreciate comments/ suggestions on these notes for the panel
> discussion
> at Extro5 "Convergent or Divergent Super-Intelligence: Can we keep up with
> AIs by integrating with technology?"
>
>
> Advanced Intelligence: SI, IA, and the Global Brain
> http://www.optimal.org/peter/si_versus_ia.htm
>
> Why Machines will become Hyper-Intelligent before Humans do
> http://www.optimal.org/peter/hyperintelligence.htm
>
> Thanks
>
> Peter Voss
>
> www.optimal.org - Any and all feedback welcome: peter@optimal.org
>
>
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:36 MDT