From: Ben Goertzel (ben@goertzel.org)
Date: Sun Nov 10 2002 - 18:13:18 MST
Doug Bailey wrote:
> Backing up a bit from this argument, I asked yesterday for a
> description of
> identifiable achievements accomplished by Eli in his capacity as research
> fellow at SIAI. Can anyone, including Eli, provide this information? If
> we are serious about identifying replacement funding for Brian
> and Sabine's
> historical generosity we'll need to explain what utility this position has
> had in the past few years. Anyone?
Doug,
Although this question really should be answered by Eliezer, it may be
interesting for you to hear someone else's point of view. I will give my
brief summary of Eliezer's achievements and then make some more general
comments.
What I have seen is that Eliezer has done a lot of interesting conceptual
research into the nature of intelligence, the possible nature of AGI and
Friendly AGI, and related issues. This is represented in the documents on
his website, including CFAI, and "Deliberative General Intelligence."
In my opinion, although Eliezer's style of presentation is less formal than
is typical in academia, the *content* of his research output over the last
few years has been superior in quality to that of the vast majority of
academic theorists.
If someone wants to fund cognitive-science theory, futurist philosophy and
theoretical AI, they'd be advancing knowledge more by funding Eliezer than
by funding nearly any university researcher in these areas. Furthermore,
funding a university researcher costs triple funding Eliezer (including
university overheads).
Eliezer also designed a programming language called Flare, which is being
implemented (and which I have mixed feelings about).
And, as we all know, Eliezer has done a lot to spread the word about the
Singularity and the importance of AGI and Friendly AGI.
So, there isn't much doubt in my mind that he has contributed (vastly more
than) $35K/year worth of intellectual content to society.
And I am just talking about his publicly released work. I know he has also
done some other work on the software design of an AGI system according to
his Deliberative General Intelligence philosophy, and some coding in the
area of causal inference, but I haven't seen this work so I can't really
comment on it in detail.
Now, all these are reasons why Eliezer "deserves funding." I should add
that I also know a lot of other serious thinkers who deserve funding, and
who aren't getting it.
I might, for instance, contend that I myself deserve R&D funding for my own
work on AGI -- the Novamente project, www.realai.net, which is a serious
effort to construct an AGI based on the in-development Novamente AI Engine
software system. Novamente is currently in a phase of intensive software
engineering by a small, international team consisting of:
a) volunteers, and b) individuals being paid to build commercial
applications based on Novamente.
For the 2001-2002 academic year, I had a research grant from a wealthy
individual named Jeffrey Epstein, which paid me a salary to work on
Novamente AGI. My grant ended in June 2002 -- and then I was in the same
situation Eliezer is in now, basically, except that I have a family to
support whereas he's single. What I wound up doing is finding a way to pay
the bills by working on bioinformatics applications of the Novamente
software. This is wonderful and exciting -- but the fact is that right now
there is *no one* being paid to work on Novamente as an AGI system. We're
pushing ahead with the project anyway....
There are many examples besides myself....
Pei Wang (PhD), creator of the NARS proto-AGI system, is teaching long hours
at Temple University. He would be pretty happy to get some research
funding, I know.
Jason Hutchens, creator of a chat program that won the Turing-test prize in
1998 (I think that was the year), and former CTO of the innovative AI
company a-i.com, has pretty much put his AGI efforts on hold because he
needs to earn a living and couldn't find funding for his AGI work.
Hugo de Garis got a professor job at the University of Utah because he
couldn't get research funding for his work on building specialized
AGI-oriented computer hardware and teaching it. His "CAM-Brain Machine" is
indefinitely mothballed because of financial issues to do with paying the
company that built it for him...
The list goes on and on, really....
My point is, this kind of work is tough to fund, and there are a LOT of
folks who have demonstrated records of achievement in AGI and Singularity
related areas, but can't get funding for their R&D work because it's too far
out of the academic & industry mainstream.
Eliezer is very deserving and so are many others.
We live in a society that is wealthy, but does not value innovative R&D very
much unless it fits into a culturally acceptable domain.
-- Ben G
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT