RE: How hard a Singularity?

From: Ben Goertzel (ben@goertzel.org)
Date: Wed Jun 26 2002 - 20:15:07 MDT


Stephen,

>
> I speculate that a future Darpa BAA will borrow heavily from CFAI.
>

I don't doubt this.... However, gov't funding of AGI does not imply gov't
involvement in the hard decisions regarding the Singularity.

For instance, if I had a 3-year DARPA grant and my system achieved
near-human AGI after 1.5 years of funding, then what? I'm gonna call my
program manager and ask him if I should let the thing run or switch it off?

I'd rather turn to a team of Singularity wizards for advice....

I guess I'm suspecting that a CFAI-inspired BAA won't happen till 2010 or
later ... and that in that case it may not happen at all, since real AGI may
happen before that, well before the sluggish wheels of DARPA turn in a
CFAI-ish or boldly AGI-ish direction...

> It now occurs to me that Eliezer (or any other researcher on this
> list), with goal of becoming a Darpa program manager or deputy program
> manager, could begin a career path that would develop the required skills
> and get the attention of Darpa management. I gave a brief presentation to
> the Darpa EELD program manager yesterday and these persons enjoy wielding
> the leverage they have over funded project participants. You do
> not have to
> possess a PhD, but publications and graduate work is desirable - you need
> only to have a well thought out vision, and be able to communicate that
> vision to senior Darpa management - in competition with other program
> managers. There is a good deal of turnover with Darpa PMs so Darpa is
> allays looking.

This is sort of an amusing idea. Specifically what sort of career path
would one follow to secure such a position?
I've got 8 years in academia and 5 years in the software industry -- am I
qualified? ;)

I would suppose that getting significant DARPA funding for one's own work
would be a good first step, or is this not the case?

[yeah right -- Ben a government technocrat ;-D ]

Ben



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:39 MDT