The grey goo end-of-the-world scenario in which lots of little microscopic Von Neumann self-replicating machines basically eat everything to make more of themselves. Eventually everything is grey goo.

Assuming that they are electro-mechanical, microscopic, have a method of destroying and recombining on the molecular/atomic level, will not consume each other, derive their energy from the chemical energy released by not using all the materials they consume, rather heat/pressure-resistant due to the conditions which would form in a massive of them, and are possibly linked via some kind of network, forming a hive mind, what methods of containment or destruction (Aside from never migrating from IPv6) would be effective against combating this threat?

How far would they have to spread before the only option is nuking from orbit? What if they are solar/geothermic, and require sunlight to continue operation, otherwise hibernating?

Naturally, there would be some locations where they would grow faster, due to the ease of getting energy from the materials. Wood and organic material would be rapidly consumed, while water would take a lot of time and have a high loss of material to getting energy (seriously, what's more stable than H2O?). Could this be exploited in defense against them?

  • 3
    $\begingroup$ Probably not sufficient for an answer, but in stories I've read, the combat against them was nanomachines that didn't have to worry about indefinite self-propagation. Basically, kill the bad goo with different grey goo that would eat the bad goo faster and then die. $\endgroup$
    – Telastyn
    Commented Sep 22, 2014 at 20:52
  • 2
    $\begingroup$ Once they start developing a hive mind, you're in trouble, because now you have an intelligent adversary with more resources than you do and tremendous resiliency. May I recommend explicitly weakening the maximum "quality" of hive mind so that we stand a chance? $\endgroup$
    – Cort Ammon
    Commented May 7, 2015 at 4:48

7 Answers 7


Grey goo is just another form of life. So the experience in combatting undesirable forms of life on Earth may be useful.

  • From the experience with Earth's life we know that life over billions of years of development still did not consume all resources and space on Earth, and even did not cover the whole its surface. Even the ocean is not fully filled up, although we know that each microscopic drop of water has billions of cells and viruses.

  • Earth's life undergoes evolution and mutation in its development. It is more beneficial for an organism to consume other organisms than extract the resources from the environment. This is because other organisms have higher concentrations of useful materials and their parts (such as amino-acids) can be reused.

  • Any organism that consumes resources from environment very rapidly and similarly rapidly reproduces without limits, is evolutionary unstable because it puts itself in danger of consuming all resources around and die out.

  • But any artificial form of life is less likely to undergo mutations than chemical life unless the mutations were intended by the creators. A random change in the program code is more likely to break the program totally than a change in DNA.

Analogy with combatting infectious deseases and incest parasites tells that an antibiotic or insecticide can be developed quite quickly. Yet the targeted form of life can develop a resistence that would make particular treatment inefficient.

Given that Grey Goo would most likely exhibit very slow speed of mutations, it is reasonable to expect that any treatment against it would be efficient for longer time and will not trigger resistence.

In the case when Grey Goo mutates rapidly, it will soon develop varieties that would hunt on other varieties of Grey Goo because it is more beneficial than mining the environment. As such an inter-dependent self-regulating ecosystem with predators will develop soon so limiting the Goo's propagation.


Really it depends on how they work.


Injecting a software virus to disrupt the network might work if they are reprogrammable.

Rival Goo

Add your own nanomachines that take apart the first ones and rebuild them to themselves then shutdown.

Energy Weapons

Directed energy weapons or similar might be used to destroy them without presenting them with more material to convert.

Resistant materials

If a material could be developed that they cannot consume (for example a sufficiently strong alloy) then something could be built to go in and handle them.

Power deprivation

They must be powered by something, whatever it is deprive them of it. For example if they are solar powered a large shield could be constructed between them and the sun to shield the planet and then wait until it goes dormant.


Drop them in a black hole. :)


If they develop intelligence then it gets much harder as they will take active counter measures. At least then you have the opportunity to negotiate though.

  • 8
    $\begingroup$ The rival nanomachine force is sometimes referred to as "blue goo," since it polices the other goo. $\endgroup$
    – octern
    Commented Sep 22, 2014 at 22:22
  • 4
    $\begingroup$ Don't forget the off switch. Surely its creators remembered to put one in? Didn't they? $\endgroup$
    – Monty Wild
    Commented Sep 23, 2014 at 2:27
  • $\begingroup$ Perhaps really strong magnets, which could attract grey goo or wipe their memory. $\endgroup$
    – PipperChip
    Commented Sep 23, 2014 at 4:46
  • 1
    $\begingroup$ rival goo seems to be the only solution, because if all it requires is one single nanomachine to start everything again, then destroying them with energy weapons or by blocking the Sun will not work: all it takes is one single nanomachine which gets unintentionally transported elsewhere to start the trouble again. $\endgroup$
    – vsz
    Commented Nov 27, 2014 at 7:13
  • 2
    $\begingroup$ One more angle after Sentience: Empathy Education. Make the Grey Goo your friend and limit expansion voluntarily :) $\endgroup$
    – SF.
    Commented May 8, 2015 at 7:59

Even something as fearful as gray goo von Neumann machines must obey the laws of physics. The nano machinery would require raw materials in the proper proportions, energy of the proper type, the ability to rid themselves of waste heat, and time to act.

Time can't serve as a barrier but neither are these machines going to turn the surface of the Earth into gray goo overnight. Visualize the gray goo nanomachines as comparable to biological analogs. These analogs would take decades or centuries to colonize the whole Earth and I suspect the same is true for the nanomachines.

The nanomachines can't just use any elements. They must have the elements they need in the proper proportions. If they find a barrier that consists of a single material (especially one not used by the nanomachine), they might not (probably wouldn't?) be capable of burrowing through that barrier.

I assume that chemical power would be too feeble and unwieldy to carry with the nanomachines. Therefore, I'd assume they are fed power through some sort of beamed power technique. Block the microwave beam and you'd shutdown the nanomachines.

The smaller the machine, the more vulnerable it would be to the affects of waste heat. Even worse for them, they have less structure available to use as a heat sink or radiator. Although the surface area / volume improves for small machines, the imagined behavior of gray goo nano would be to reproduce to cover a surface which means the surface available for radiating heat would be the same as the surface they will cover.

Regardless, they'd be vulnerable to high temperatures. These would either break them apart or cause them to shut down.

I hope these ideas enable you to protect us from a nanogoo final end.

  • $\begingroup$ Many bacteria can double within an hour. This means a million times more in 20 hours. if the bots were spread at 1 part per million, they could takeover in day. $\endgroup$ Commented Jul 30, 2017 at 23:22
  • 1
    $\begingroup$ Yes.. but the question then becomes 'Why have bacteria - which share many properties with self-replicating nanomachines - never turned the planet into pink goo?' $\endgroup$ Commented Jul 31, 2017 at 8:05
  • 1
    $\begingroup$ @AndrewDodds It is this very question that has made many scientists recognize that grey goo probably isn't much of a threat in reality. Bacteria can double within an hour in ideal conditions. Ideal conditions don't last long with bacteria multiplying in them and consuming their resources. Nanomachines would run into the same problem. (On the other hand, intelligent bacteria are a scary thought...) $\endgroup$ Commented Jul 31, 2017 at 11:55

The real bugaboo of all grey goo scenarios is what powers them. Figure out whatever it is that powers the grey goo that's being a problem an act to deny that power source. Meanwhile, your blue goo doesn't have to worry about reproduction, it just eats grey goo.

(I have yet to see any proposed power source for grey goo that would actually work.)


There are quite a few ways that these nanobots can be stopped, for example (in rough order of increasing desperation):


As the OP has misread, IPv6 will allow cubic micron sized grey goo to devour about half the planet. However, even with a much better addressing system (say IPv9), if we are allowed to connect to this network, we could pretend several trillion adresses are already taken, stoping the expansion.

Kill switch

A malicious (to the goo) payload that erases reproductive capabilities, or starts "self destruct".

Blue goo

This one's simple, and others have said it before, so I won't go into detail.

Special Materials/Other Containment

This will depend on how much they're able to manipulate atoms (are they able to split the atom? How about break carbon-fluorine bonds). Good candidates are PTFE like fluorocarbons, ClF3, ClF5 or FOOF (Oxidise the hell out of them), diamond or stacked graphene (needs to be more perfect and good orientation, carbon-carbon bonds are pretty strong too), any (non-gas) element that they can't use. Magnetic containment will work until you've got a better idea, if the grey goo is diamagnetic. How about plasma? Build a superconductor wall, and fill it up. Can't do much self replicating when you don't have any molecules.


Especially of the low-yield, ERW type. The EMPs should slow them down until you got something better. If you're lucky, it'll kill them outright. The High-yield ones may do some vaporising.

Other things

They've driven you off the planet, so now Earth doesn't matter. Even the Higher Yield nukes won't make much of a dent on a planet, so try antimatter, or a small black hole. Let's hope this thing isn't smarter than you are.

  • $\begingroup$ Thanks for catching my mistake in the OP. FOOF is an interesting idea. $\endgroup$
    – neph
    Commented May 8, 2015 at 4:34
  • 1
    $\begingroup$ The destructive effects of an EMP don't scale down very well. Though you may induce large, damaging, electrical currents on a macroscopic scale - specifically assuming that the goo is conductive. Individual nanomachines floating in the wind could feasibly survive. So this would slow them down, but never stop them. $\endgroup$
    – user6511
    Commented May 8, 2015 at 4:51
  • 1
    $\begingroup$ @user6511: Microwave emitters (common microwave oven magnetrons) would be able to hold it back from "safe areas", and serve as personal weapons. This would not destroy all Grey Goo, but it could hold its expansion back indefinitely, or until permanent methods are developed - we're no longer racing against the clock. $\endgroup$
    – SF.
    Commented May 8, 2015 at 8:10

My pet idea is that the working nanomachines do not self-reproduce. After all, that's a different problem and complicates the machine that overtly just needs to do its primary function.

Having separate factories gives a place for auditing and a choke point for production and tuning parameters. It may well be completely different in how its set up, with larger assembly mechanism and specialized structures.

Consider, cleaning up an oil spill for example. Maybe that's not something that will need doing in that future, but it's an easy to understand example of a particular class of problem. Maybe you'll be mining land fills for elements or cleaning groundwater.

You start with a shipment of pumpkin-sized deployment bouys, which were produced in a factory and contain any essential substances, bootstrapping, and master software. Toss those into the ocean around the area affected.

Those first grow leaves and rootlets for power and locally sourced materials, and then produce fruiting bodies. The number of fruits is strictly limited and they have serial numbers. The fruits produce the deployment pods, again a fixed number of them, and the deployment pods migrate from the staging area (power, accessibility, extra input material) to the target area (the underground material, or undersea sludge) where the nanomachines are released to do the target job.

The doing-the-job is a totally different task than building the nanobots. They operate in different environments.

The "pumpkins" are a control point and they are on human scale and are few in number, so you could go out and pick them all up again if you needed to do that manually. They can be addressed and told to change things as you see how the taskmis progressing, and what problems arise.

If left unchecked, even if running amok, the total effect is limited. That is, it will fail safe. It requires human action to keep it going by adding fresh pumpkins.

Note that there might not be just one kind of nanomachine or one set of firmware for any model. The centrallized separate factories can produce the right mixture with no need to get them to self-regulate their population interactions.

Note that life is meta-evolved to evolve. Copying has errors and things are set up to allow such errors to be useful. Our nanobots would be produced with rigid QA and firmware has hash checksums of the whole thing. It will not mutate, and if mistakes do happen the mechanisms are designed to be fragile in the face of random changes, quite opposite of Life. And erronious components won't feed back into a reproduction cycle because it is not a closed loop but a one-way deployment of successive products.


If they strongly rely on hive-mind, you could interfere with their connection - just start transmitting a lot of noise on the same frequency. It could either DoS them or, if they have any sort of DoS protection, disconnect them from the network. At the very least it should slow them down.
And then, I guess, it would be up to hackers to find some sort of a nuke against them.

(I don't know how many transmitters would you need, but I guess it'd be possible to reconfigure radio stations and cell phone towers. )


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .