• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed



  • Powered by FeedBlitz

« The nano-Rorschach | Main | Assessing Nanofactory Capability »

June 13, 2004

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Janessa Ravenwood

I'm glad to see you recognize that any regulations enacted must have teeth or they are worthless (whether or not I agree with any regulations is a separate issue). To confirm, if North Korea gets unrestricted nanotech (and at some point, they will), do we start bombing?

Chris Phoenix, CRN

Janessa: I think there's a pretty good chance of delaying nano-anarchy for five years. And I have no idea what happens more than five years post-nanofactory. I couldn't begin to hope to plan for it. People will have to make plans during those five years. But five years is a lot better than five months!

Mike: If your plan works, which it might, then I'll be quite happy with it. A nitpick: your fourth point seems unimportant. The chemicals are for efficiency, not control. It's not that hard to turn grass clippings into chemicals. And a more serious point: "conventional" hardware is no longer conventional when it has many orders of magnitude better performance. I suspect that just a few tons of carefully designed nano-weaponry--even if entirely electromechanical--could destroy the human race, if there were no countermeasures.

Janessa: "if North Korea gets unrestricted nanotech (and at some point, they will), do we start bombing?" This is a question that CRN has been asking for a while--except that North Korea can be replaced with any nation. And if we get unrestricted nanotech, does someone else start bombing us? And if two get it at the same time, do they get into an arms race?

Nukes are easier to stockpile than to use. Nano-weapons are easier to use than to stockpile. We're in deep trouble.

Chris

Janessa Ravenwood

On the attack issue, it's not quite that simple. Attacking the U.S. is guaranteed suicide. The reverse (as recently demonstrated - TWICE) is most certainly not true. So if we get it first, I in fact expect no large-scale attack. On the other hand if someone else gets it first we might attack them (more likely we'd just steal it and save ourselves the trouble).

Chris Phoenix, CRN

Mike wrote: "We need to be clear on what our objectives are here. Do you want a regulatory system that would prevent your crackpot neighbor from killing you and everyone in your neighborhood? Sorry, that would require a system that would be too restrictive. Or do you want a regulatory system that would prevent any terrorists organization or rogue nation from destroying a large country or even all life on the planet? That we can agree on."

I'd like to agree with you on both points. But... What worries me is the idea that the terrorist and the crackpot neighbor may be the same problem, given a sufficiently powerful technology. Whatever the mental pattern that makes people go berserk or amok, it seems to include "kill as many as possible." If the hypothetical berserk neighbor could take out a whole city instead of just a block, my impression is that he would. (Any psychologists reading this who can supply more information on this?) And MNT has the potential to be *extremely* destructive.

But I want to take a few steps back. What is my overall goal? It's to develop some kind of policy that allows for threading a course between several Big Threats that can be placed in rough chronological order. I think the first one is unstable arms race. The second is massive oppression from governments. The third is massive criminal and terrorist activity. The fourth and fifth are societal meltdown and grossly irresponsible use of products. If AI counts as a massive threat, it's probably an early one.

We have a few Big Options. Space flight: free, restricted, or none? Nanofactories: free, restricted, or none? Surveillance: (many possible degrees). Large-scale psychiatric intervention: (many possible options).

We have a few Big Questions. Which will be easier: defense or offense? (I'm pretty sure offense wins.) How quickly can new products be developed and deployed? (It'll vary widely, but may require only weeks for some products. The biggest delay will be writing embedded software.) What difference can various kinds of AI make?

The diversity of Big Threats implies that there are constraints: you can't let a genie out of a bottle to solve one problem if it'll make later problems unsolvable. For example, unrestricted nanofactory access might avoid massive government oppression, but lead straight to massive criminal activity. On the other hand, black markets are a genie that's very easy to release and hard to stuff back.

Please, everyone, comment on my lists of Big Threats, Big Options, and Big Questions. If we can agree on those lists, then we can start to narrow down policy choices.

Chris

Chris Phoenix, CRN

Janessa: For a nation, attacking the US is suicide. For a network organization, it's not. If I were a fanatic medieval group whose only hope of global significance was to keep the West from developing a post-crude-oil economy, I might act just as Al-Qaeda did.

Chris

Janessa Ravenwood

Well, hopefully new nanotech resources will grant us enhanced abilities to help kill every last one of them first, especially if we get it before they do (so that's a scenario I'm rooting for).

Janessa Ravenwood

Big Threats:
* Unstable Arms Race: No reason it need be too unstable. I think it might be great for the economy to have a new cold war going on (say against the Chinese). It would pump lots of $ into nanotech R&D funding and as we won the last cold war hands down I feel pretty confident of our success in another.
* Massive Govt. Oppression: We seem to have a good track record on avoiding the worst of this. Short of outright rescinding the Constitution, I’d say the worst Orwellian scenarios won’t come about in the U.S. Possibly in the rest of the world, but it’s the U.S. I’m concerned about first and foremost.
* Massive Crime Wave: Possible, but unlikely. What is the objective of the crime? Robbery or violence just for the sake of violence? If it’s robbery, MNT is likely to take away a lot of reasons to do so rather quickly as you won’t need money for a lot of goodies. The stupidest thing you could do here would be to restrict nanofacs from manufacturing illegal drugs. If the junkies can get their fix, drug-related crime virtually disappears (as does the DEA who’d have a fit trying to enforce drug laws with no more dealing or importation going on) and this takes a big part of all crimes committed. Expect a massive black market in unrestricted nanofacs to pop up overnight if such a colossally stupid restriction is in place on personal nanofacs. If it’s just violence for violence’s sake, well those are thankfully fewer in number than the robbers and it’ll be up to the police to deal with them like any other violent nutjobs.
* Social Meltdown: Could you be a bit more specific here?
* Grossly Irresponsible Use of Products: And here, too?

Big Options:
* Spaceflight – free.
* Nanofactories – very minimal restrictions. If such does not appear likely, free. I'll take my chances with free over totalitarian.
* Surveillance – unless there’s good reason to believe you’re a terrorist, no spying on citizens (carte blanche for spying on non-citizens is always OK by me).
* Large-Scale Psychiatric Intervention – Yikes! Who gets to decide what “correct thought” is and who gets brainwashed for not conforming to the standard? Head for the hills on this one…

Big Questions:
* Easier: Offense – “the art of destruction is infinitely easier than that of creation.”
* The Rest: Unknown. I defer to some of the other here for these.

Karl Gallagher

I think defense is going to be stronger than offense because of surveillance. If we can carpet the world with cameras it'll be impossible to launch an attack without the target seeing it coming and taking countermeasures. Sure, your state-of-the-art nanobots can do major damage, but if you can't take control of the target you haven't gained anything. This is referring to war as opposed to terrorism, which has a different dynamic.

Janessa Ravenwood

I guess I WOULD be classified as a subversive in such a society as I would be trashing the cameras any way I could. Well, I suppose maybe depending on your definition of carpeting the world. Any camera looking into my house is GOING to get trashed and I don't about whether that's legal or not.

Karl Gallagher

I'm working with the "Transparent Society" scenario. Janessa--if you could pull up a view of any window in the city (including the mayor's office, the police cellblocks, and all your neighbors) from your PC, would you tolerate cameras at your window or would you be willing to be the only person in town who can't be checked on for dangerous activity?

This doesn't change the offense/defense question--I think you could get a camera net good enough to detect planned attacks without getting to everyone's bedroom, every street corner and some key offices is probably sufficient.

Mike Deering

BIG OPTIONS


  • Space Flight - Human expansion into space has so many advantages that delaying it would be counter-productive.   As you point out in another post, anything that you can do in space with nanotech can be done on Earth too.   The only difference is the difficulty presented to surveillance by the speed of light limitation.   Unless you are going to have ubiquitous surveillance on Earth this difference is not significant.   And if defenses to nanotech weapons stays ahead of offenses then there is even less reason to restrict space travel.   This is not a simple question of which is easier, offense or defense.   You have to take into account the advantages that a globally organized, well funded, high priority defense development program will have over a renegade group or individual.   The resource imbalance between the good guys and the bad guys will be significant.   The advantages of human expansion into space include but are not limited to, separating people who can't get along, giving people something worthwhile to do, learning new stuff about planetary systems and space objects, and of course the eggs out of the basket analogy.   The last one, though not an attractive one to need, is still a significant advantage.

  • Nanofactories - They should be restricted enough to make it difficult for them to be used in a WMD manner, but unrestricted enough that they provide economic and physical independence.   I think a careful balance of built in restrictions, regulation, law enforcement, and multi-tiered nanotech defenses can achieve objective.

  • Surveillance - Many people feel strongly about the right of privacy, for reasons don't understand, which makes ubiquitous surveillance politically impossible.   If ubiquitous surveillance were imposed in a non-political fashion, by a governing body not subject to the citizenry, there is the grave danger of government abuse if not implemented in a transparent manner (ubiquitous access).   In solving the problem of surveillance you are trying to satisfy three competing constraints; weapons development detection, privacy, and government abuse.   Many complicated schemes have been proposed to balance these issues but they are all clouds in your coffee (I got that from a Carly Simon song, I'm so vain I probably think this thread is about me), because the argument can be made that nanotechnology is so small and versatile that any amount of privacy is equivalent to no surveillance.   Therefore we are left with only two options: the transparent society of David Brin, or ineffective public surveillance only.   My personal vote would be the former but I think the latter wouldn't mean total failure either.


    BIG QUESTIONS


  • Defense vs. Offense - If you don't want to go all the way on defense, don't bother.   Half measures will not be effective.   We need a truly multi-tiered defense system.   The first tier is government activity in technology restrictions, regulations, law enforcement, and a proactive terrorist offense program.   But all government activity in all its forms whether through direct action or mandated regulation is merely the first tier and insufficient to the overall objective.   It must be complimented by the other two tiers.   The second tier is open source public effort defense technology development and defense hole patching.   The third tier is personal involvement in your own nano-safety.   You must take personal action to insure your nano-safety.   The first two tiers can't do it for you.   People who will not take reasonable actions to protect themselves can not be protected.   A related issue is defensive strategy which also happens to be three-fold; harden the target, use early warning and containment systems, and preventive measures.   If you take into account the advantages that a multi-tiered, three-fold strategy, high priority defense program will have over a renegade group or individual, then defense can stay well ahead of offense.

  • Product Development - The speed and ease of product development will be directly related to the sophistication of the CAD/CAM software.   This software is advancing in complexity and capability exponentially along with many other aspects of technology.   Of course, the advent of artificial general intelligence (AGI) will provide a quantum leap in this capability allowing non-specialists to design products from scratch.

  • Effects of AI - The advent of AGI will provide intelligent control of these technologies to the common man.


    BIG THREATS


  • Arms Race - I don't think avoiding an arms race is possible.   Therefore I think it is imperative that we take the necessary steps to insure that we win the arms race.   The arms race is only unstable if it is balanced.   One winner is a very stable situation.

  • Totalitarianism - This is the biggest risk we face, in my opinion.   No government can be trusted with sole possession of a technology this powerful.

  • Terrorists - See Offense vs. Defense above.   Incentive for most crime will necessarily drop if we relax the restrictions against victimless crime and material scarcity is abolished.

  • Societal Meltdown - Unavoidable.   Most current societal structures will be obsolete after the Singularity (the simultaneous development of MM and AGI).

  • Environmental Abuse - can be avoided with reasonable regulation and law enforcement.

  • Bad AI - Bad problem.   See my upcoming paper "Telluris of Cyborgs: Safety Passage through Great Filter of Evolution" co-authored by Pavel Vassiliev and Mike Deering, to be presented at TransVision2004 in Toronto in August.   Copies available in Russian and English.


  • Janessa Ravenwood

    [sigh] Another David Brin acolyte. I have no idea why anyone would ever believe that mass surveillance would ever be anything but one way – the authorities looking at us. I’m firmly convinced – as a political junkie who tries to understand how government actually functions in practice as opposed to theory – that believing you could ever convince the authorities to allow you to watch them is incredibly naïve. There is NO way they would ever allow it. Don't believe me? Try to introduce that legislation and get back to me on how long it took to die in committee.

    And I’m QUITE OK with being the only person in town who can’t be checked on for dangerous activity. Cameras at my window will be smashed – and I SINCERELY doubt that I would be the only such person in town doing so. That’s the kind of thing that’s SO outrageous that it invites political “regime change” at the next election and inspires voter initiatives (at least here in CA) real fast.

    I MIGHT be OK with the street corner thing but only if the situation were to grow VERY desperate. I think we have too many cameras now.

    Mike Deering

    Janessa, I agree with you that a transparent society ala Brin or any form of ubiquitous surveillance is a political non-starter.   I just mention it as a practical measure. If done properly it would work.   But I am under no illusions of it ever being done, properly or otherwise.   That is why the arguments against space expansion are spurious.   Any threat that can come from space can come from Earth, developed in secret.

    Janessa Ravenwood

    Actually, my response was to Karl Gallagher above - you must have hit POST just before I did.

    Brett Bellmore

    I tend to agree about the political infeasibility of Brin's transparent society. Congress carefully controls where CSPAN's cameras can be pointed on the floor of the House or Senate, why would anyone expect them to permit uncontrolled cameras in their offices? There's just WAY too much public officials do which they can't afford to let the public see, let alone allow to be recorded in a fashion which would constitute admissible evidence in court.

    If it were to happen at all, it would only be as a result of a rather violent revolution. Because politicians are aware that that level of transparency in their day to day activities would probably cause one, anyway.

    Karl Gallagher

    Janessa: believing you could ever convince the authorities to allow you to watch them is incredibly naïve

    The same authorities we can fire from their jobs? Used to be the authorities were hereditary, now they serve at our sufferance. We've already got a start on this with FOIA and web access to Congress's activities. If politicians learn they can win elections by promising more transparency, we'll get transparency.

    Janessa Ravenwood

    I wouldn't hold your breath waiting. And if neither of the (top 2 party) candidates in an election feels like focusing on that, who will you vote for? Unless you can convince a LOT of people that this is a REALLY BIG social issue - more than taxes, abortion, drugs, social security reform, wars (pick any other big controversial topic) - then it's going nowhere.

    Brett Bellmore

    Yes, the same authorities who have manipulated ballot access rules and campaign regulations to achieve a 98% reelection rate. And who just made it illegal for people to get together to buy airtime to say things about them, at least when it's close enough to an election to influence the outcome.

    We've got the theoretical capacity to remove them from office, but don't confuse that with free elections. And, yeah, there's FOIA, but you can't FOIA things that don't get recorded in the first place, which is why you shouldn't expect cameras in Congressional offices any time soon.

    michaelvassar

    Janessa, are you a patriot? I agree that the US can beat China to MNT if we want to, but we're not even trying, and if we did get it, what then?
    Since the US has the world's largest absolute prison population except for China and the largest fraction of its population in prison except for China, I am skeptical of your claims about us avoiding oppression. The US is definitely a freer than average country, and really was once the freest country in the world, but that is just because the standards were very low. Anyway, MNT changes the whole meaning of government, possibly in difficult-to-predict ways.
    I agree normal crime is not a problem, and that violence for its own sake is a very rare motivation. Also, with MNT, only intelligent violence is likely to matter, and the only really smart terrorists I am aware of (ever) are Kazinski and Lennin. However, hackers may be a huge and hard to control problem. Letter-bomb spam anyone?
    Space flight + nanofactories :-0 Are you SURE you want to enable 9/11 with comets? It basically makes planetary existance impractical in the long term.
    Free nanofactories mean goo is an issue, but probably a minor one due to the difficulty and the blatant suicidal nature of goo. If you want to die, you may want to take out many people with you, but you don't want to spend five years learning better engineering techniques and another five designing a weapon. Nano-enhanced plagues are more likely, especially during the first decade. After that, nano-defences can probably stop them.
    Moderately restricted nanofactories = free nanofactories, as you well know.
    Surveilance: I'm sympathetic to Brin's Transparent Society, though I don't think it's enough
    Psychiatric intervention: I share your feelings. Definitely not, at least not without decades to evaluate the real consequences.

    Mike

    You know, it shouldn't be so hard to enable independence with desktop factories that are NOT nanofactories. Imagine the following. Moisture condensor+water purifier, refrigerator, 10kw solar generator, 100 kw/hour mechanical energy storage, yeast + plankton culture tank, greenhouse, food processor, nanotube shredder/blender/hypercentrifuge for recycling, waste incinerator, 3D printer, chemical plants to make printer feedstocks, 10kw of mixed LEDs for illumination, displays, VR, greenhouses, etc.

    You don't get GAI. Imagine what bacteria would do if they had obedient human servants!!! Some AI technologies may be safely used, but GAI cannot be unless a comprehensive safety theory is developed in advance.

    Arms race: I agree, but you are walking into...
    Totalitarianism: Why oh why do techies NEVER distinguish between totalitarianism and other non-democratic systems of governance? It is important to prevent totalitarianism, but a nanotech monopoly, while they may be somewhat oppressive, will have NO MOTIVE to become totalitarian. Totalitarianism is about the discovery of techniques for total control of the lives of the citizenry. With MNT, such control is irrelevant. Such governments don't NEED soldiers or grunt laborers.

    All: As for transparency, it's happening. Think Reality Television. The first time a reality TV star used their popularity to get into congress wouldn't change much, the third or fourth time would. When half of the officials were doing it, any opposition candidate would gain a weapon against their opponents. "what's he hiding"? Unfortunately, it looks like MNT is developing faster than entertainment/security. Full transparency would take 20+ years to develop.

    Brett Bellmore

    "Space flight + nanofactories :-0 Are you SURE you want to enable 9/11 with comets? It basically makes planetary existance impractical in the long term."

    First, speaking of "transparent" societies, last time I looked, vaccum was transparent. As in, you can see what's going on across planetary distances. Nobody is going to be able to go out to the cometary belt, and drop a comet on Earth, without being noticed years before it could strike.

    No, if there was going to be any threat from people living in space with unrestricted nanotech, it would come from far more sophisticated weaponry. Grey goo filtering down like dust by the kiloton, maybe.

    But the larger issue is the whole mindset here. A kind of paranoid feedback loop where, because you don't trust people not to attack you, you want to do things which will GUARANTEE that they'll want to attack you.

    Brett Bellmore

    I mean, seriously; If your neighbor was talking about chaining you to a post, because they were afraid that if you got out of their sight you'd come back with a gun and shoot them, you'd probably start looking under gun shops in the yellow pages. At this rate, the revolutionary underground is going to be fully organized before CRN's regime could ever get established.

    In fact, I think I might start designing some T-shirts...

    Janessa Ravenwood

    Brett: I think you've hit on my mindset for this to a "T." That's exactly the reaction I get - on a subconscious level - when I start hearing about massively restrictive proposals of any kind. Up the jolly roger and viva la revolucion are my instinctive responses.

    Mike Deering

    Janessa, what level of restrictions would you be comfortable with? Do you see any restrictions as necessary for the safety of the human race? What would you want restrictions to accomplish, and what would you want them to allow? If we can come up with a strategy that will satisfy you and CRN, we could sell it to almost anybody.

    Chris Phoenix, CRN


    I'm not sure it matters much whether a government is totalitarian or merely authoritarian. I was using the former word to encompass both, which was maybe sloppy. But any successful authoritarian government would probably have to be pretty totalitarian.

    I'm not sure whether it's possible for a government to be so nice that no one wants to blow it up. That would be ideal, of course. But a government that allows too many liberties will be attacked by fundamentalist fanatics--you can't please everyone. And by the time you defend aganst them, you've taken away enough civil liberties to threaten the Janessas and Bretts.

    The fact is, any MNT-enabled government would be able to choose how totalitarian to be. I think Mike's statement that totalitarianism [authoritarianism?] is the biggest risk deserves major attention.

    So why not have multiple governments, for a checks-and-balances effect? Or why not have no government, and let non-governmental groups build whatever power structures they can? Because I think Mike is right about arms races not being stable. I've posted my reasons a few times. I've created a new post/thread for more discussion of it: "Can an MM Arms Race Avoid Disaster?"

    Mike wrote: "Terrorists - See Offense vs. Defense above. Incentive for most crime will necessarily drop if we relax the restrictions against victimless crime and material scarcity is abolished." This, I have to disagree with. Crime motivated by material need may drop. But there are other motives for crime. Ideology; vandalism; or simply being a control freak. There are whole societies out there which are based on the idea that too much freedom is undesirable or even evil. Some of these societies are sources of terrorism.

    Chris

    Brett Bellmore

    Here's the deal: I'm not at all sure that it's possible to avoid a bad outcome on Earth.

    First, barring a nuclear exchange resulting in mankind's extinction or return to the stone age, we ARE going to develop nanotech, or some equivalently powerful technology, in the fairly near term. MNT isn't the only way to achieve self-replicating industrial infrastructure, after all; It could be done with something near conventional technology, and a bit better automation to take humans out of the loop, and that is definately the direction things are trending. It's only a matter of time before some company "closes the loop", and manages an automated factory that can build all it's own parts. The original vision of the Von Neuman machine, aka the "clanking replicator".

    And while nanotech could provide some really nifty capabilities, the top down approach is ultimately going to be able to get close to matching them. While biology is going to provide some very nasty weapons indeed for those so inclined. And the cost of the equipment for genetic engineering is dropping all the time, while software tools become more widely available.

    The whole trend of history, since we got out of our hunter-gatherer rut, and started advancing, has been towards more power in the hands of the individual, for creation AND destruction. While people haven't been getting proportionately more durable. I for example, am not particularly violent, but as a well educated engineer, if I REALLY wanted to, I'd have no particular trouble accomplishing the death of thousands. And I'm scarcely unusual in that regard. Essentially the only reason the human race survives, is that the vast, and I do mean VAST, majority of people are actually quite rational and non-violent.

    But a vast majority is not 100.00 repeating percent. There are irrational, violent people out there. Some of them are even quite intelligent and well educated, though that's, thankfully, unusual. And the rising curve of individual capacity IS, I fear, eventually going to cross some threshold, where we can't endure the tiny fraction of violent loons multipled by the vast destructive power everyone has within reach, if they only want it.

    There are three elements to the problem; The first is rising capabilities, and I don't see where that can be dealt with, or even should be, since almost everyone uses those increased capabilities for good. Competition will keep that curve rising.

    The second is the terribly degree of interconnectedness of modern industrial society, which is designed to work very efficiently when things are going well, but is also terribly vulnerable when things don't work well. And here there's room for improvement. Frankly, there's no good reason that our cities have to be hours from thirst, days from starvation, if the infrastructure goes down. We've got UPS for computers, we could redesign our infrastructure to be more failure tolerant. Heck, it's as simple as keeping some bottled water and non-perishable food in the closet, to as complex as building houses like nuclear submarines, so that if somebody releases something deadly, you can hole up while it's dealt with. And nanotech would certainly help in that latter regard. We could make every room a bomb shelter, every suit of clothes NCB gear!

    And the third aspect is that fixed human fragility. And here, again, nanotech can help, by things ranging from artificial immune systems, to physical enhancements, to distributed personality backups such that nuking a city would only cause property damage, but no "deaths". We don't HAVE to pit ever improving weapons against unchanging people.

    But, regulation? At one extreme, nanarchy, every loon has their own army of T-2 terminators. At the other extreme lies a dystopian nightmare of liberty eternally extinguished. And there doesn't necessarilly have to be a stable, acceptable middle ground.

    Which is why I'm so big on dispersing mankind into space as the earliest oportunity. Job one is ruling out human extinction. Making things turn out nice on Earth is at best job two.

    The comments to this entry are closed.