From: Dani Eder (firstname.lastname@example.org)
Date: Wed Aug 01 2001 - 08:41:01 MDT
> I think we will have a human-equivalent-intelligence
> AI at least a few years
> before we have a master-of-AI-and-computer-science
> AI that can launch the
> hard takeoff.
> Just as I myself was a human-equivalent intelligence
> well before I became a
> master of AI and computer science myself ;)
But your hardware is fairly fixed in capacity.
An AI can rapidly upgrade by adding more
processors, memory, etc.
The only thing I see that could keep an AI
"stuck" is if the required hardware to run it is
not yet available.
But I will posit a hypothesis:
If your hardware setup is sufficiently powerful to
develop and test an AI, then the technology is
probably available to build a superintelligent AI.
Assume that some fraction of human-level capability
1/n is required for reasonable development and
testing of an AI. Probably n<1000.
Call the fraction of computing power on the planet
the AI project will consume 1/m. Probably m>10,000.
Since m>n, the power to run a super-intelligent AI
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:37 MDT