One possible way is an MRI, assuming the nano-locomotives have metallic properties.

The process wouldn't be pretty (bruising), depending on the amounts and locations of them, but would be effective.

At least that's the way it happened in Michael Crichton's book Prey. :}

I concede that it will probably eventually be possible to construct nano-scale machines which will be all but impossible to stop from killing a single person, or even a group of people in a local area. What I do not think will be possible, however, is scenarios such as the "grey goo", or any derivative thereof.

Such as it is, I don't think we have to worry about the world being completely consumed by tiny robots — I think there will likely be some form of protected enclosure that prevents the extinction of the human race if something like this ever did manage to spread across the globe. I have to think this is the case, simply because if I were to take the route Kadamose seems to be driving down then I would not care to protect myself or humanity.

That being said, I believe that we are all constructs of nature, and that we are not special amongst the rest of the natural universe in any profound way. This does not lead me to the conclusion that we should all just stop caring and die. Rather, I believe we are capable of some complex and interesting behavior and am very curious to see where it all leads. Alas, this has me veering towards a philosophical essay, and I'm sure you will be happy that I do not do that

It is possible that it all leads to a dead end, pun intended. It is also possible that we are the very harbingers of that end. However, if one does not think, does not use intuition and imagination, then it is not only possible, but it is probable.

There will always be people like Kadamose, people without any sense of hope or reason. This is unavoidable. And, as a yin to that yang, there will likewise always be people who are perhaps too optimistic and hopeful, like me.

Give me some time to think and I will likely reply to your previous post, Chem. At least, in a manner that is on topic and not wandering and aimless such as this one has been.

> Pre-programming will work in the way you have
> described it. However, the problem being debated
> here is one of preventing malicious intent.

We are talking about two different methods then. In my scenario the bots will be programmed to do a task, will enter the body for a brief time, and then self-destruct and leave the body. Your scenario assumes that bots live in the body permanently and respond to instructions when those are given. This seems a lot more dangerous to me. If you want them to work only at a predetermined time and place, then having them hang around like inert debris for the rest of the time is not only unnecessary but a danger to the patient's health.

> Therefore, if you can program in advance to do a
> malicious act you should also be able to program
> a way to counter that act.

Well… Yes. But how will you convince each terrorist to put backdoors into his programs? You see, those who program malicios acts are not the same people who design the bots in the first place, and to attempt to limit the bots' capabilities to prevent any conceivable danger is beyond the powers of my imagination, since the same type of actions that are required to do cure the patient may be used to kill him if performed in a different location.

> The interface itself wouldn't be complicated.

Could you give us a general idea of what you are thinking?

> The old saying in sports applies: The best offense is a good defense.

And an even older saying in martial arts is "offense is the best defense". This also happens to be the opinion of the U.S. government. I have not yet seen any proposal on how to fight, or even detect, the type of nanoplague that I keep on describing (see the other thread of this article).

> I think it will be easier to stop nanomachines
> than it will be to keep them going

You might as well say that about bullets. We still can only stop those with very heavy body armor, and only in a very small area of the body.

> I still believe there will be a way to differentiate
> and target human created machines such as these

If you design them to be differentiable. But what if such a machine were to be designed specifically in order to evade detection? It really is not that hard because unlike with macroscopic objects, there is no way to look inside a nanobot. You can feel its surface and try to decide whether it is malicious, which seems rather futile to me when the surface can be made to approximate any chunk of biological matter simply by creating binding sites for cellular junk on the skin surface. The only way to be really sure is to decompile each and every object you are not absolutely sure is hostile, and that means every grain of dust, every flake of skin, every dust mite, and every thread of clothing. Think of the energy requirement for keeping such an ultraclean space!

> On top of that, virii are very hard-pressed to survive outside of a living host for very long.

Only because of the way their jackets are constructed. A nanomachine is not constrained to protein materials; it can build a buckyball-type shell, which would be nearly indestructible under all external conditions.

> the whole point is that when they are outside of
> the host then you can perform all sorts of
> things on them that you wouldn't want to had they still been inside.

Again, think of the energy requirements to decompile them all. Furthermore, it is the ones that are inside that pose immediate danger; the ones outside are merely potential infectors.

They will try, and in some cases they will succeed. And the only way to ensure safety of the human race is to keep the Good ahead of the Bad… The old saying in sports applies: The best offense is a good defense. I think it will be easier to stop nanomachines than it will be to keep them going, so I don't see a problem with having defense that outweighs offense.

You're not serious with this statement, are you? If you are, then it means you are the one who is delusional…in this world, there are no good guys and everyone is only out for themselves. Therefore, it's simply best to give the technology to those who will utilize it to its full potential…and if it destroys the human race in the process, it just proves that extinction by our own hand was inevitable since the beginning of our pitiful existence.

I am currently ashamed to be human, and I'm surprised everyone else isn't ashamed, as well. We are an extremely vile race, with nothing but selfish and evil intentions…so perhaps it's best that our existence comes to a swift end.

"By programming them in advance for a specific task. This may have to be done in two stages; first run to find the problem, second run to fix it. With storage capacity of 10^21 bits/L, I don't think you will have any problem with space for programs or results."

Pre-programming will work in the way you have described it. However, the problem being debated here is one of preventing malicious intent. Therefore, if you can program in advance to do a malicious act you should also be able to program a way to counter that act.

"There have been some doubts as to the safety of sitting in a variable magnetic field for long periods of time."

Who said anything of long periods of time? I am talking about magentic immersion on the scale of an MRI, etc. Nothing extreme should be required.

"Really, now. If you are trying to power a microsocopic machine with a macroscopic force, it would be unwise to make the interface this complicated. It is difficult to tell if the power is getting through as it is."

The interface itself wouldn't be complicated. The machinery would actually probably be quite easily implemented, but only on a scale that would work in a square meter of area, for example. As far as being able to tell if the power is getting through as it is… this is why we have something called experimentation on cadavers etc. It shouldn't be too difficult to make sure it works before you try it on living breathing person.

"What technology? While this is a good idea, you should keep in mind that somebody has to build them first. And in that case the information concerning their design can be stolen and modified by other people to run on ATP (adenosine triphosphate) instead. So while you are correct in trying to make a specific design difficult to misuse, you must also take care to make it difficult to reverse engineer and the designs impossible to steal, because they will try; oh yes, they will try. "

They will try, and in some cases they will succeed. And the only way to ensure safety of the human race is to keep the Good ahead of the Bad… The old saying in sports applies: The best offense is a good defense. I think it will be easier to stop nanomachines than it will be to keep them going, so I don't see a problem with having defense that outweighs offense.

"What makes you think they will be inorganic? They will have to be constructed of biocompatible materials, and may have to accumulate a protein jacket to avoid being eaten by the immune system anyway. Its internals are either going to be similar to biological machinery, or harder substances like diamond. In the former case they would be pretty hard to differentiate from the body, and in the second case you would have to work hard to make them destructable."

Certainly portions of the machines will be constructed of diamond and protein.. and most of them will be largely organic. I still believe there will be a way to differentiate and target human created machines such as these, unless, as I have said, they are simply virii which have been engineered.. and in this case there may still be a way.. On top of that, virii are very hard-pressed to survive outside of a living host for very long.. and even if they can, the whole point is that when they are outside of the host then you can perform all sorts of things on them that you wouldn't want to had they still been inside.

> Tell me, if they're shielded from all outside
> interference and signals, how exactly do you
> communicate with them?

By programming them in advance for a specific task. This may have to be done in two stages; first run to find the problem, second run to fix it. With storage capacity of 10^21 bits/L, I don't think you will have any problem with space for programs or results.

> Electromagnetic oscillatory propulsion may be
> unlikely, but as far as I can see it would be
> one of the most safe ways to do it.

There have been some doubts as to the safety of sitting in a variable magnetic field for long periods of time.

> Because it would be hard to duplicate, let alone
> even achieve, without extremely sensitive and
> specific equipment and an operating environment
> engineered within strict boundaries.

Really, now. If you are trying to power a microsocopic machine with a macroscopic force, it would be unwise to make the interface this complicated. It is difficult to tell if the power is getting through as it is.

> It would make that technology extremely
> difficult (if even possible) to misuse on any
> sort of large scale.

What technology? While this is a good idea, you should keep in mind that somebody has to build them first. And in that case the information concerning their design can be stolen and modified by other people to run on ATP (adenosine triphosphate) instead. So while you are correct in trying to make a specific design difficult to misuse, you must also take care to make it difficult to reverse engineer and the designs impossible to steal, because they will try; oh yes, they will try.

> Anything inorganic inside an organic system
> immediately differentiates from the environment

What makes you think they will be inorganic? They will have to be constructed of biocompatible materials, and may have to accumulate a protein jacket to avoid being eaten by the immune system anyway. Its internals are either going to be similar to biological machinery, or harder substances like diamond. In the former case they would be pretty hard to differentiate from the body, and in the second case you would have to work hard to make them destructable.

"modified shielding which is immune to magnetics, RF, and radiation"

First I have to point out the obvious. That statement is so flawed that you might as well say they'll operate on super-conducting monkey-colliders. Or maybe you should read Microcosmic God by Sturgeon and see what happens when the lifeform the scientist creates is told to create an impenetrable shield.

Why? Because if these little nanobots are in any way going to be controlled intelligently, you're going to have to be able to communicate with them. Tell me, if they're shielded from all outside interference and signals, how exactly do you communicate with them?

Electromagnetic oscillatory propulsion may be unlikely, but as far as I can see it would be one of the most safe ways to do it. For exactly the reason it would be unlikely: Because it would be hard to duplicate, let alone even achieve, without extremely sensitive and specific equipment and an operating environment engineered within strict boundaries. It would make that technology extremely difficult (if even possible) to misuse on any sort of large scale.

As far as "100 mA devices" (and I think you mean uA, not mA) operating on Active Transport, that's all well and good. But I shouldn't have to point out that such devices will inevitably have susceptibilities. It shouldn't be difficult to destroy them alone, again, with the proper interference.

Anything inorganic inside an organic system immediately differentiates from the environment, making it possible to single out and destroy the inorganic objects, especially if they are delicately crafted machines.

That is why virii will continue to be the most difficult to remove pathogens into the far future, because often the only way to kill them is to kill the surrounding tissue, which, if you want to go on living, is totally unacceptable. In this case, nature has already developed an efficient method (if not the most efficient), and we will be hard pressed to do better.

You may hold on to your delusion of world-engulfing, singularity inducing, invincibly progressing nanomachines if you like. I just get the feeling that what you are realling holding on to is more of a neurotic compulsion of misanthropic origin.

It can be done via modified shielding which is immune to magentics, RF, and radiation. This could be done in multiple ways (i.e full body armor or modification of the nanobots themselves)

As for elecromagnetics powering these nanobots…that's highly unlikely. They will most likely be powered via ATP, as they will only require a few milliamps to run. The Japanese have already done research on this area, and currently have 100 mA devices running in and on the human body powered by the blood stream of the individual.

Tell me, how do you circumvent a safeguard such as radio resonance-harmonics propulsion? If the devices were constructed properly:

• As soon as you left the staging area the devices would no longer function.
• Widespread broadcasting of the frequency wouldn't be high enough power to sustain them.

• Or, let's even get more inventive and imagine them working in a different way. You are in a machine. On each side of you are electromagnets which oscillate between on and off in a designed manner. The oscillation of the magnetic fields is what is used to power the nanobots. Perhaps you would need more than two opposing angles of magnetic attraction, but you get the picture (I hope). So, Kadamose, tell me, how are you going to produce exactly the correct oscillating magnetic field surrounding each and every person's body to allow operation of these nanobots? Please, tell me, you say "one way or another"… So give me one.

Built-in safeguards WILL be circumvented, in one way or another; it's just a matter of when.