From: on behalf of Ben Goertzel <an
To: Jeffrey Epstein <jeevacation@gmail.com>
Subject: quick summary of situation
Date: Mon, 17 Aug 2009 01:51:56 +0000
Hi,
Thanks for taking the time to talk to me, amidst all your busy-ness...
This is just a quick summary of the practical points we discussed at the end of the call. I
just want to be sure the situation is clear to you.
1)
Working on my own, I can make some progress toward AGI, but not very fast progress. So if
I'm working on my own and don't have a team working with me, I'm more inclined to work on
theory, with the goal of making my AGI design simpler and more robust, so it will be more
effective and easier to implement once I have a team again.
2)
If you were willing to fund say 2 or 3 others besides me, then we could keep making real
engineering progress toward AGI. But this would cost more like $17-20K/month, significantly
more than funding me.
3)
If you don't want to fund some sort of ongoing "build a thinking machine" project, then I
think my best career move is probably to return to academia. Funding my research via running
a small AI consulting company just isn't working anymore. If I return to academia I'll keep
Novamente operating on the side -- and try to sell the virtual pet engine to game companies,
etc. -- but then I won't be relying on it for personal income. But given the academic hiring
cycle, if I want to return to academia in September 2010, I need to apply in the December-
January interval. I.e., there's a long lead time. Maybe I should have applied for academic
jobs for September 2009, but that's water under the bridge. I didn't do so because I thought
Dan Stoicescu would keep funding Novamente, but instead he disappeared and turned anti-AI, as
I described to you ;-(
4)
What would be *ideal* (short of you funding an ongoing AGI project ;-), would be if you were
willing to fund a one-year research fellowship, that I could use at a university near where I
live for the upcoming academic year. Then I would be able to apply for academic positions
from an academic position, which is always nicer. And then, from September 2010 onward I
wouldn't be in such a precarious financial position personally (though I'd still have the
problem of funding my AGI project...).
5)
If that is too much for you, then funding for this fall would certainly be great also. This
would enable me to finish working out and writing up the math I described to you (grounding
my AI design in Marcus Nutter's "Universal AI" mathematics), which I think is of great value
for my AGI project and for the science of AI in general. During the fall I would also apply
for academic jobs, and for other jobs to carry me through the first 8 months of 2010.
6)
By the way, your notion of starting a new financial system (like they tried to do with
PayPal) is certainly very interesting to me. If you want to give me a job masterminding a
new financial system for you, that would be great fun. Not as great as working on AGI but, a
lot better than the kind of boring government AI job I would get around DC, that's for
sure...!
7)
Yeah, my personal situation is a bit desperate ;-p I owe more on 2008 taxes than I have
in savings; and I'm operating without income as of August 1. On the other hand I'm not going
EFTA00773722
to get evicted or anything like that; I have a $23K pension fund I can dig into (though I'll
pay a tax penalty), and my parents would loan me $$ if I asked.
Basically, Dan's $$ ran out in July and I didn't have a backup plan in place to cover the
contingency in which he got flaky and didn't continue his funding. Obviously I have paid a
lot more attention to the math and engineering of AGI than to my personal finances ;-p
I guess that's about it.... I really appreciate your willingness to listen and to help.
And I look forward to talking in more depth about AI science F2F when you get the chance ;)
About why AGI doesn't get more traction: please remember that this is not a problem just for
me!! Academics with interest in AGI are not running out of money personally -- but they're
unable to pursue their AGI projects. They're paid to teach classes, and to supervise
students working on other stuff that the funding agencies like better.
If I didn't have a shared child custody arrangement here, I would probably relocate to China,
where there **is** government funding for AGI !!
What we need is a wealthy visionary to fund a real AGI project. But I sent you a careful
proposal on that last year, and unfortunately it seems not to have crossed your threshold of
fundability, though we never talked about exactly why...
thanks a lot
Ben
EFTA00773723