[ExI] Universal Computing Red Herring [WAS AI Motivation revisited]

[Anders Sandberg]

On 2011-07-08 12:43, Stefano Vaj wrote:
> 2011/7/6 Richard Loosemore<rloosemore at susaro.com>:
>> Can I just reiterate that universal computing has absolutely NOTHING to do
>> with any of the issues related to Artificial General Intelligence, or the
>> possibility of a hard takeoff, or the viability of human-like artificial
>> intelligence.
>
> Such an unqualified assumption sounds quite weird. Do you seriously
> expect that AGIs could be implemented on non-universal computing
> devices? After all, human beings themselves do exhibit universal
> computing capabilities... :-/

Humans approximate universal computation, but since we are finite 
systems we are not universal. And neither are our computers. However, we 
are rarely close to the limits of these approximations, at least in the 
computer case.

-- 
Anders Sandberg
Future of Humanity Institute
Oxford University