Re: Freitas on Gray Goo

From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Sun May 14 2000 - 03:03:25 MDT


On Sat, 13 May 2000 hal@finney.org wrote:

> A minor criticism first: the title of the paper seems unnecessarily
> opaque: "Some Limits to Global Ecophagy by Biovorous Nanoreplicators,
> with Public Policy Recommendations." Here we have two unfamiliar terms,
> "ecophagy" and "biovorous". As far as I know these are coined terms
> which you will not find in the dictionary or the technical literature
> (they do seem to have been used in some science fiction computer games).

Hal is correct and Robert is a wordsmith. However those are not
reasons to object to the quantitative aspects of the paper.
Doing so falls into the "Scientific American pitfall" of attacking
the style or author when you cannot attack the technological discussion.
Since Hal attempts to do this (below), I will put the above comments
in the "I didn't like the style" editorializing arena.

> While the meaning of these words can be deduced from the root forms,
> and "ecophagy" is defined at least indirectly in the introduction, it
> leaves me with the impression that jargon is being used unnecessarily
> to add political impact.

It isn't politics. Robert has had this style for ~20+ years of writing
that I've reviewed his papers. It is more along the lines of the
"Alice in Wonderland" -- "A word means exactly what I intend it to
mean, nothing more, nothing less" (paraphrasing on my part).

>
> I've heard of Drexler gloatingly putting up slide after slide of dense
> equations from Nanosystems, to tweak his earlier critics who complained
> that Engines lacked technical depth. I can just imagine the glee with
> which Foresight presentations will quote the mouthful of a title with
> which this paper has been burdened.

Robert actually pushed Foresight to restrain itself from commenting on Joy's
perspective until there were some numbers on which productive discussion
"could" be based. I agree that the title is burdensome, but it serves
as a much greater academic effort than Joy's fear-mongoring.

> Well, that is absurd. It will only take 20 months if the goo expects
> to take over the biosphere without anyone noticing! Don't you think
> someone would notice? Hey, we just lost New York. Naw, couldn't be,
> the global temperature hasn't gone up more than four degrees.
>
> Obviously the goo can't help being detected at some point. The 20 month
> scenario just doesn't make sense.

Agreed. The point of the paper is to define limits. There were
discussions behind the scenes on discussing the relative advancement
rates in different environments, but the assumptions and equations
rapidly become more complex than this paper could have dealt with.

The paper sets some general limits. If potential nano-takeover becomes
apparent at an earlier date (Oh my god, Hal has disappeared from the
face of the planet, he must have been consumed by biovorous nanobots...)
then that is fine. There is no discussion of the things nanobots
must avoid consuming to keep humans from suspecting that something
is amiss. The general analysis is one in which nanotech seeks
to remain as unobserved for as long as possible. There is ample
room for subsequent studies to take the paths of "natural" biovorus
replicators that do not care whether they are observed and "synthetic"
biovorus replicators designed to be a stealthy as "current" technology
allows.

> A more plausible scenario is described where the gray goo grows somewhat
> faster and the temperature is allowed to rise more: "For example, taking
> t = 100 sec, TEarth = 300K, and Ediss ~ 100 MJ/kg, the transition to
> the ETPL regime occurs when total global nanomass reaches ~5 1010 kg,
> or only 0.001% of total global biomass, and the last ~17 population
> doublings remain to be completed over a time span of ~2 tlast = 210^7
> sec (~7 months)." 7 months still sounds like quite a bit of warning,
> although not as generous as 20 months.

The question here becomes whether or not people take the threat seriously.
If enough ionizing/radiation-beam weapons have been stockpiled on the basis
of the potential threat the difference between 7 and 20 months isn't
significant. If the threat is not taken seriously, then 20 months
doesn't matter much to the eventual result. If the defensive measures
have been prepared, deployment should take days to weeks. In those
situations the 7-20 months make little difference.
>
> However even with this scenario, it is sensitive to the efficiency of
> the goo. These two scenarios are assuming dissipation of 100 MJ/kg
> for the conversion. This is justified with "Drexler [4] estimates
> that the typical energy dissipation caused by chemical transformations
> involving carbon-rich materials will be Ediss = (q Dbio) ~ 100 MJ/kg of
> final product using readily-envisioned irreversible methods in systems
> where low energy dissipation is not a primary design objective." Well,
> we're assuming the goo is attacking using a stealth method where low
> energy dissipation is an important design objective. Hence Drexler's
> estimate is not very relevant.

I argued very strongly in pre-reviews that there is a significant breakpoint
between the energy dissipative and reversible (low energy) methods.
To grow faster than current biomass is likely to require higher
energy dissipation because you have to invoke transport methods
for molecules that require a significant production of heat.
Most bio-transport currently is by diffusion which is slow but
generates little heat. To reduce energy consumption significantly
you have to pay the price of slower (reversible) operations.

I very much doubt that early nanotech can operate in the energy
efficient regime and when it does I expect it will be slow.

I know of no evidence that suggests that you can operate in an
energy efficient regime and be "fast".

> The paper also suggests that the 100 MJ/kg estimate is appropriate
> because highly dissipative designs are easier to produce and, given the
> difficulty of the problem, these are the kinds of gray goo threats most
> likely to be faced during the early and intermediate years of nanotech.
> Even if true, the paper has not previously stated that its estimate was
> only meant to apply to immature nanotech systems.

The paper underwent an evolution and would require significant modifications
to divide it between the "immature" and "mature" nanotech threats. There
is an assumption, perhaps not well discussed, that the defense capabilities
are roughly comparble to the offensive capabilities. So the questions are
whether immature nanotech can trump itself or whether mature nanotech
can trump itself. These should be explored further in individual detail.

>
> Even terrestrial vegetation is quoted as being able to do better than
> this figure, dissipating 38 MJ/kg. The paper is assuming that nanotech
> will be less capable than biology. That is not consistent with the usual
> conservative design assumptions for nanotech. Drexler and Merkle think
> that 0.1 MJ/kg is possible in theory.

Initially (immature) nanotech *will* be less efficient than biology.
Biology has had billions of years to try and make things as efficient
as possible. That will not happen in nanotech overnight.

> If we assume 10 MJ/kg, modestly better than the biosphere, the 7 month
> scenario shrinks by a factor of 10 to about 3 weeks, much less time for
> defense. And if we ever did reach Drexler's optimum design, things would
> be over in minutes (as little as 76 seconds from single nanite to total
> conversion of world biomass, according to one of the more extravagant
> extrapolations in the paper).

Yes, the *BIG* question is *IF* you can reach Drexler's optimum design.
If you cannot point out a concrete way of acheiving that, then you
have to stick with numbers that are "reasonably" achievable.
If you look at *highly* optimized systems such as ATP production
in the mitochondria, you will realize that "optimum design" is
a very nice theoretical concept that requires concrete paths to be
something more than a pipe dream.

> Another problem is that the paper's emphasis is on detection. In several
> places it seems to assume that this is the hard part of the problem and
> that actual defense is relatively easy, which is certainly counter to
> the conventional wisdom:

The question is *where* is the conventional wisdom? Do you have citations?

I'll cite a nanodefense based on the energy requirements for a highly
disruptive beam of neutrons vis-a-vis the energy costs of constructing
a highly organized nanobot. Disruption is cheap, construction (esp.
at the atomic scale is expensive). Nanobots may be able to withstand
hurricanes but they are going to drop like flies in lightening strikes.

> It's ironic that the standard argument for gray goo advantage, that it can
> be destructive while the defenders must preserve information, is turned
> on its head here.

Precisely -- the only situation in which grey goo is a threat is if
it is *replicative*. A single bacteria in Kenya that has developed
the ability to kill any human is *not* going to be a threat to me
*unless* it can replicate itself and get close enough to come into
contact with me. A single grey goo nanobot is worthless because
the biosphere grows faster than it can destroy. So in order for
the badbots to be a threat, not only must the breakdown and consume
the good stuff, they must do so while making sure they do not suffer
an fatal damage to their own operating systems.

> Now it is the gray goo which is struggling to survive
> and replicate, while the defenders with their numerical superiority can
> lay waste if necessary in order to destroy an infestation.

Why is this surprising? It is what occurs *every* day with the current
designs of our own immune systems. There are volumes written on
the use of anti-oxidant supplements to defend against oxidative
damage from hyperactive immune system cells. Our immune systems
wage war and "collateral" damage is a necessary evil.

> This is not an early or intermediate level of nanotech development.
> It would be among the most sophisticated nanotech applications imaginable.
> By the time such a global system could be designed, developed and put
> into play, gray goo could have wiped out the world ten times over.
> There seems to be a fundamental mismatch between the sophistication of
> the goodbots, who run an active immune system that checks every cell
> on the planet, and the badbots, who can't manage to operate even as
> efficiently as green plants.

That is precisely the point -- *if* the badbots *do* operate more
efficiently than green plants they will be detectable. Anything converting
biomatter into nanomatter is going to alter its spectral signature -- how
do you do that "stealthfully"? The "active immune system" starts with
simple monitoring of heat radiation and molecular gas abundances
(using *current* technology). It grows in complexity as nanotechnology
becomes more sophisticated. You have to make a concrete argument that
badbots can hide their activities from goodbots at the same level
of technological sophistication.

>
> In summary, Robert Freitas does a good job in the quantitative analysis
> of the limitations faced by gray goo. But the conclusion that the gray
> goo problem is readily dealt with is much less convincing. The paper
> would be better if it just focused on the numbers, and left the strategic
> analysis to another day.

A number of people (not myself) pushed for the inclusion of more
strategic analysis. The inclusion may not have been the best
result (perhaps due to a failure to address things in sufficient
detail in a paper that was already getting long).

>
> It also should be much more objective about the seriousness of the gray
> goo threat. Foresight seems to have made a political decision to downplay
> gray goo in the last several years, and this paper unfortunately seems
> to be consistent with that political position.

I doubt Foresight has made a decision to "downplay" the threat. More likely
internal discussions (that I have not been privy to, but can imagine)
have made it seem clear that the defences may provide effective countermeasures
against attackers. I actually was fairly shocked in my first read of
the paper that 20 months was much too fast and we are all doomed.
But then after considering the monitoring options and possible responses
by defender-bots, I realized that 20 months gave us a wide margin
of safety.

There is ample precedence for this in biology as we live in a very diverse
bio-world and have yet to see a monoculture takeover occur

> Much more work needs
> to be done before we have a clear picture of the true scope of the gray
> goo threat. Robert Freitas has made an important contribution, but we
> are not yet in position to settle the matter.

No question. Clearly the paper says, if we relax our vigilance, we may
encounter problems. Less clear is the fact that if we do not prepare
ourselves adequately from a defensive perspective, we may be overwhelmed.

Hals comments however are useful. They engage some of the really
difficult problems regarding *what* are good and feasible defenses
and how rapidly will more efficient nano-ecologies develop.

Robert



This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:11:12 MDT