[extropy-chat] Social Implications of Nanotech

[Eugen Leitl]

On Mon, Nov 10, 2003 at 03:46:55PM -0500, Robin Hanson wrote:

> What if aliens show up?  What if time-travelers from the future come back?

We can plausibly show that aliens most likely don't exist (or we wouldn't be here).
Ditto time travelers (they're really indistinguishable from aliens this way). 
As are elves, Ents, unicorns and celestial bodies made of ripe gorgonzola.

However, it will be a tad more difficult to show that certain widespread
local organisms don't think, that we can't figure out how they do it (nor
derive those methods from scratch via brute-force evolutionary algorithms),
using means of atom-scale imaging of vitrified chunks of said animals, and
cheap computational hardware by the metric wagonload.

Meaning: ff you postulate molecular manufacturing, you better show a plausible
mechanism preventing advent of AI via one of the above routes, or both.

> It seems to me to make the most sense to choose some sort of baseline
> scenario, then analyze each substantial change by itself, and only after
> try to combine these scenarios.  So let's assume, for the purpose of

If you look at two subcritical chunks of plutonium, you will not arrive at
the correct conclusion of what a supercritical assembly of them will bring
you, unless you factor in adequate level of theory.

> analyzing nanotech, that automation progresses over the next few decades at
> a similar pace and character as it has over the last few decades.
> 
> You are assuming something about how fast these things work.  What if
> they worked very slowly?

If they work very, very, very slowly they wouldn't be able to fab themselves,
and hence are not there for all practical purposes (see Manhattan project). 
Nevertheless, self-amplification is exponential in
principle. You will notice that humans replicate very, very slowly.
It took them a while to become very visible on this local planet here.
Nevertheless, we're very lucky their self-reproduction rate is adaptive, and
tapering off. Otherwise, we wouldn't be having this conversation, by virtue
of being dead, or not having been born in the first place. 
 
> >Could we have an "open source" economy, where people work on whatever
> >interests them, or create designs that satisfy their own needs, then
> >make the fruits of their labor available to everyone for free?
> 
> The form of the market for designs seems a separate issue.  Logically,

What's the market for gcc? autoconf? GNU/Linux? I notice this particular
market has been eating several Large Companies alive, yet has not yet found
adequate treatment in classical economics circles.

> you could have ordinary factories and open source design, or you could
> have PGMDs and profit-motivated copyrighted designs.

Both have their niches. I'm presuming a more biology-inspired development model,
though.
 
> Sure, if pigs can fly, the sky is going to look different.  But we need
> to do the analysis step by step, and not jump too quickly to big 
> conclusions.

If your premises are bogus, your conclusions are food for the abovementioned
airborne porcines.

I'm not sure whether you're looking for publishable papers, or trying to map
the more radical implications of comparatively simple technologies like
molecular nanofacturing (completely ignoring the AI issue for time being).

If you're realistic, your peers will reject your papers.
If you're conservative, your fellow transhumanists will point and laugh.

I'm not sure there's a sweet spot between those two bonfires.
 
-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144 http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20031110/1d95e97c/attachment.bin>