What Is "Grey Goo"?
Grey Goo! Nooooo!
Imagine that you're driving across a bridge, up high so you can see everything for a few miles. You're stuck in traffic, and at first, you don't realize why all the cars are stopped. As you turn your head to the left, you soon see exactly why: a dark cloud is slowly advancing toward you, and it seems to be consuming everything in its path! Even worse: the cloud seems to be growing as it travels in your general direction, gobbling up the very earth itself, other cars in its path, roads, and even buildings. The "grey goo" scenario has come to life!
Quick side note: this is a follow up to an earlier "Nanotech for Dummies" style article, so if you're unfamiliar with some of the more basic concepts, you may want to check that out before proceeding too much further.
Why Sci-Fi Gets It Wrong
In the above scenario, we have seen a "nanobot swarm" come to life and quickly spiraling out of control, and then destroying everything in its path, and ultimately (perhaps) even the earth itself. How much of this is fantasy, and how much could soon be reality? With popular culture starting to hop onto the sci-fi conceptual bandwagon, it's now more important than ever to lay out a reasonable explanation of why "grey goo" isn't a credible threat (and why nanotechnology is well worth developing). Special thanks to Mike Haydell for helping structure the argument and for giving me the impetus to write this article.
Nanoscale - the Size Limitation
First, it's important to understand just how small the nanoscale really is (again, reference the previous article on nanotechnology for an idea of how to put the scale in human terms). Here's Mike: "Robots sophisticated enough to do the onboard computing required for molecular manipulation, generate the power needed (manipulators, sensors, propulsion), and to fly around the environment freely will be far too large to work with other robots to make grey goo a threat. They certainly won't be nano scale. Sure they could do one atom at a time, but we're losing more atoms from our body every second then a single robot like that could do in a lifetime (these robots don't get the scaling benefits Drexler talks about, they have to move very slowly, so maybe they can pick up and move one atom a second, at best). Also, the fingers might not be too fat, but 10^24 grey-goo-bots can't all work in the same area to build or destroy macro scale objects in the time needed. Plus they would get destroyed by cosmic rays, brownian motion, wind, and who knows what else. Plus the other robot they're building would be subject to the same forces. The grey-goo-bots would take one step forward and nature will bring them two steps back."
Atomic Scale Manipulation
While it's true that Newtonian physics still works on the microscale and even on into the nanoscale, when it comes to atomic manipulation, certain inevitable realities immediately come to bear. Mike again:
"In July of 2014, the University of Basal in Switzerland used an Atomic Force Microscope to manufacture a 5.6 nm wide Swiss cross out of bromine atoms. They did this at room temperature, in air. This means that atomic manipulation is possible without a cryo-vacuum, but this work was only done on a sodium chloride surface and through a process known as "lateral manipulation," which is done entirely at the surface of a material. It was not making or breaking covalent bonds (a prerequisite for robot construction) and it can not be scaled to 3D (also a necessity if you want to build a grey-goo-bot). Furthermore, if you could build in 3D, gas phase molecular motion will then begin to affect your precision.
The average kinetic energy of a particle, for every degree of freedom, increases as 1/2*Kb*T, where Kb is Boltzmann's constant. That means as the temperature increases, so does the kinetic energy, and thus the particle's speed. At room temperature, many molecules in the gas phase travel "on the order" of 100 m/s. "Grey-goo-bots," and their work piece, would be subject to getting slammed constantly by these molecules, which will upset the precision of construction. In order for one grey-goo-bot to build another grey-goo-bot the way portrayed in Michael Crichton's "Prey," "GI Joe: The Rise of Cobra," or even "Transcendence" it would have to be done in vacuum, at cryo-temperatures. At which point it is completely separated from the environment and unable to cause a chain reaction, the primary driving force for the grey-goo scare. Moreover, the grey-goo-bot is constrained to working with only one atom at a time, which is prohibitive when dealing with the number of atoms required to make a single grey-goo-bot with the capabilities expected as such."
It is worth noting that once molecular manufacturing is a reality, a relatively small group should be able to manufacture nanotechnology products well away from public scrutiny, but "grey goo" is hardly as urgent or credible of a threat as cybercrime (immediate) or superior weapons of terror (not so distant future). Having said that, nanotech will certainly help police all of these things, if we are careful about how we mitigate the dangers (read on).
There Are Ways to Mitigate the Actual Dangers
Given that most objects that will be manipulated into crawling, flying, and replicating machines will be at or above the nanoscale (and not atomic in scale, or even remotely close), there are definitely going to be numerous people working to deal with the threat of "runaway" nanotech. Nanobots would need to be self replicating in order to be effective on any massive scale, and as such, nano-factories will have to be contained, protected, and have an external computer and power source. An external power source would limit the potential for "swarms" to escape into the real world. Further, the alarm bells have been raised by Hollywood and the popular press. It's highly unlikely that any sort of large scale nanotech design will slip past media scrutiny.
In addition (and perhaps most importantly), nature has its own built-in self regulation mechanism. Here's Mike:
"Simple: Bacteria will eat them. Microbes are nanobots with a billion years of refinement. They are very good at eating things made of graphene and nanotubes, and working with the limited energy and resources available in their environment. This is why nanobots work best in a controlled environment - they're very useful, but couldn't survive in the wild." Special shout to daelyte on reddit for the original idea of bacteria as nature's defense mechanism, too.
If extremely small scale, self-replicating robots indeed prove to be feasible, cosmic rays from space, simple terrestrial wind, organic bacteria, and other naturally occurring evolutionary defense mechanisms on planet earth (including Brownian motion!) will be fully capable of tearing apart "nanobots" at the atomic scale all on their own.
The bottom line: the amazing things we can do with nanotechnology - from hidden sources of energy, to very small devices in our bloodstream capable of policing our immune system, to Drexler's "factory in a box" idea - will far outstrip the potential risks, particularly with nature's built in mitigation system, and with a tremendous amount of care and concern. The media tends to play on our darkest (and most unlikely) fears becoming a certain reality, and nanotech is no exception.
Here's Mike to wrap things up:
"Predicting technological possibilities far into the future is very difficult. Anything consistent within the laws of physics will inevitably be invented. Furthermore, it's important to realize that ALL technology is a double edged sword. Atomically precise manufacturing will certainly bring with it new dangers when realized, but this article shows that grey-goo is unlikely to be among them. Responsible use of nanotechnology should be at the forefront of everyone's mind, but it is important to realize that just because dangers exist doesn't mean those projects shouldn't be supported. On the contrary, better funding and support will increase safety and responsible use. The sooner nanotechnology is as ubiquitous as cell phones, the safer it will be."
Soruce: Shigeki Kawai, Adam S. Foster, Filippo Federici Canova, Hiroshi Onodera, Shin-ichi Kitamura, and Ernst Meyer. Atom manipulation on an insulating surface at room temperature. Nature Communications, 2014 DOI: 10.1038/ncomms5403
This content is accurate and true to the best of the author’s knowledge and is not meant to substitute for formal and individualized advice from a qualified professional.
© 2015 Andrew Smith