|Posted: Dec 21, 2007
(Nanowerk Spotlight) One of the more interesting concerns of nanotechnology is 'grey goo.' The term was invented by Eric Drexler to describe one of the dangerous issues that must be faced as nanotechnology capabilities evolve. Here’s how it works. 1. Pretend that nanotechnology truly exists to the point where we can fabricate machines of arbitrary complexity using individual atoms or molecules. 2. Pretend that these machines have sufficient complexity and computational means that they can make copies of themselves using whatever happens to be lying within their reach. 3. Pretend that their fabrication systems are such that they can make a copy of themselves about once an hour. 4. Pretend that one of these machines decides to do nothing except make copies of itself.
THEN we take this
And turn it into this in about 1 week.
It’s a bit worse than the Borg. The idea is that everything gets converted into grey goo: you, me, trees, chickens, and everything. This would not be good news. It would totally ruin my retirement plans.
By comparison, let’s think a bit about my favorite
bacterium, e. coli. This bacterium lives in your stomach and
mine, and is about 10 micrometers in width, and can make a copy of itself in about 20 minutes. If just one e. coli decided
to replicate itself uncontrollably, it could perform the same
feat as our hypothetical nanomachine in about a day and a
Because of this concern, there are those in the
nanotechnology community who have proposed legislation
that would make it illegal to create a nanomachine with the
ability to make a copy of itself. One wonders if we should
advocate similar legislation for bacteria.
I am, for some
reason, reminded of Indiana House Bill #246 submitted in
1897 which would have made the legal value of Π (pi) = 3.2
The interesting issue here is that e. coli long since made
the genetic decision to make as many copies of itself as fast
as possible forever and ever and ever. Just counting the
bacteria that live inside humans, there are about 3.9 x 1023
(39,000,000,000,000,000,000,000) bacteria that have made
that same genetic decision. If we move into the rest of the
world, the number of bacteria that have made similar
decisions is estimated to be 5 x 1030. So how come these
bacteria haven’t converted the planet into sludge? Well,
there are two leading answers.
The first is that bacteria dine on each other, so since
everyone is having lunch at the same time, no single bacteria
has an opportunity to grow at an unrestricted rate without
becoming lunch for another bacteria. The grey goo model
suggests in this case, that since evolution hasn’t created our
runaway nanomachine, there are no natural predators to stop
it, so it will convert the planet into copies of itself in about a
The second one is that, fundamentally, the universe is a
hostile place, and no matter what your intentions, be they
lofty or bacterial, the universe just doesn’t provide you with
all the parts you need to make things on demand. In fact, if
you’re in the mood to make a lot of copies of yourself, the
raw materials in your neighborhood are going to get
consumed rather quickly, which is going to slow down your
rate of replication. Similarly, if you happen to need some
rare substance in your replication process, such as tantalum,
you’re not going to find very much of it floating around
waiting to be consumed.
Thus, while grey goo is certainly an interesting idea and
legislation will be pending to see to it that we don’t make it
outside of Iraq, there are likely to be some significant natural
laws that should bring the threat level down to a manageable
On the other hand, we need not worry too much, for
the nanotechnology community has proposed that along with
grey goo, we should also fabricate blue goo. Blue goo would
be the nanocops of this new age, nanomachines that have the
express purpose of hunting down and destroying grey goo.
The one caveat to all of this is that the military potential
of the various kinds of goo will certainly result in some level of funding over the next few decades. In fact, a larger
version of goo technology has already been funded by
DARPA under the name "smartdust" Perhaps it isn’t such
a bad idea to legislate the value of Π...
By Glenn Fishbine