Zyvex Labs Gets $9.7M for Molecular Nanotechnology Research Tuesday, Oct 7 2008
nanotechnology 10:01 am
DARPA gave it to them:
“RICHARDSON, Texas, Oct. 2 — Zyvex Labs today announced the award of a $9.7M program funded by DARPA (Defense Advanced Research Projects Agency) and Texas’ ETF (Emerging Technology Fund). The goal of this effort is to develop a new manufacturing technique that enables “Tip-Based Nanofabrication” to accelerate the transition of nanotechnology from the laboratory to commercial products. Starting with the construction of ‘one-at-a-time’ atomically precise, ‘quantum dot’ nanotech-based products in volume at practical production rates and costs. Harnessing this capability will position the United States and Texas with the fundamental technology to develop next-generation quantum dot applications for military and commercial applications such as advanced communications, metrology, and quantum computers. The spin-off nanomanufacturing capabilities from that early application will result in revolutionary nanotech products in follow-on development.”
In my opinion, current advocates of molecular nanotechnology (MNT) aren’t doing enough to address the risks. Christine Peterson of the Foresight Institute advocates an open source physical security model, which is helpful, but should be accompanied by more specific recommendations to form a seed around which further ideas can accrete. The Center for Responsible Nanotechnology has laid out the technological specs of MNT and called for more discussion, but has provided little in the way of concrete recommendations. Ray Kurzweil seems to just think that everything will pretty much automatically turn out fine.
Nanofactories (manufacturing units based on MNT) will need to have extensive, unhackable built-in safeguards in order to be safe. If they can be hacked and these hacked nanofactories cannot be recovered, that could be very bad (significantly worse that terrorists getting weapons-grade uranium). That’s a phrase I’d like MNT advocates to repeat publicly: “terrorists or tyrants getting their hands on unlocked nanofactories would be far worse than weapons-grade uranium”. Unlocked (or poorly regulated) nanofactories would be able to build devices that enrich uranium many times more effectively than current centrifuge technology. That’s somewhat of a problem, unless we plan to gather up all the uranium on the planet and keep it locked up in vaults.
If unhackable nanofactories cannot be built, then to push ahead on the technology would be irresponsible. Mainstream “experts” will be saying this in 5-15 years, but I’m saying it now.
October 7th, 2008 at 11:39 am
We must plan for the scenario in which nanofactories arrive before friendly AGI, but IMO, without higher-than-human intelligence, it will be very hard to avoid existential threats for long once that tech is out there.
Anyone who has read papers on security by Bruce Schneier knows that it’s almost impossible to make something un-hackable, especially if the gains of cracking the security measures are high and the attackers have lots of resources (in this case, everybody will want to get their hands on it, including intelligence agencies and corporations with lots of resources).
I’d have to think about it some more to see if I’m not missing something big, but with my current state of knowledge, I’d be in favor of a temporary relinquishment of fully mature self-replicating nanofactories until after we have FAGI, unless the best way to get there includes the use of nanotech; f.ex. to scan brains or build more powerful computers, in which case, it should be used only for that project (which would also be risky — as soon as it is known that the tech exists, it will become a target). But we should be able to do what we need to do without von Neumann self-replicating machines.
October 7th, 2008 at 4:18 pm
I firmly believe that there is nothing that can be built by human intelligence that is not hackable by human intelligence.
That said, I think it’s irresponsible to not pursue it. If we don’t, someone else will. Perhaps even those same “terrorists”.
What’s coming will put EXTREMELY powerful technology in the hands of everyday people. I’m not worried about some “terrorist”. I’m worried more about the lonely nerds I know who will be sitting in their basements hacking matter.
There’s no guarantee we’ll survive the coming changes, but it certainly won’t be boring!
October 7th, 2008 at 5:38 pm
First of all: it’s $10 million - that’s bugger all from a budget of billions. If they manage to kill us all on that budget, then they deserve an economic award as well as a Nobel Prize.
Unhackable? Not only is such a thing not possible, it’s also highly undesirable. I’d rather that there are many eyes on the code (regardless of whether they are good or ill) than a single gatekeeper. Security through obscurity doesn’t work. Security through ceding your freedoms to government really doesn’t work (except for the government).
Besides, the real utility in nano is that it is a manufacturing technology - it’s virtually useless on its own. A tool that you cannot use is a pile of junk.
“terrorists or tyrants getting their hands on unlocked nanofactories would be far worse than weapons-grade uranium” - the only thing I can say about this statement is: what a load of rot. If this were a truly valid concern, then the governments that are complaining about nukes might consider getting rid of the thousands they have sitting around “in case”. The fact of the matter is that even if terrorists managed to successful cook off a nuke in a populated area, it would be a tradegy, but it would be survivable (for a nation, the world and the species). If a government decided to use nukes, then we are talking endgame. As an example: if India and Pakistan decided to go at it, and they both have only dozens of nukes, both countries would be obliterated and probably the whole of asia and a large part of the middle east as well from fallout. Given global weather patterns continental America would probably cop a large amount of fallout also. Forget about ever eating seafood again - it will all be glowing. Sure, we are only talking about the suffering, disease and death of 2 billion plus, so it’s not the end, but it could escalate to it very easily). So why is that ok (and entirely undiscussed) but we are all supposed to be in fear of a statistically unlikely terrorist attack?
Terrorists have far greater success with teaming basic explosives and religious indoctrination to create walking powderkegs - why would they need uranium? Uranium is a nightmare to work with, especially if you are trying to make a weapon. The opportunity cost is too high to make it worth the effort (the only viable way to use nuclear weaponry is to buy it off the rack. There are plenty of people who would gladly sell a perfectly serviceable bomb for the right price). Getting people to blow themselves up is cheap and requires almost zero technology. They didn’t bring down the WTC with nukes, they did it with box cutters (it was a truly elegant attack, cheap and easy to implement, and a massive payoff (In terms of terrorism only, 4K casualities is an embarassment in conventional warfare) - I cannot help but admire the thinking behind it).
As for gathering up all the uranium on the planet and keeping it locked up in vaults, have you tried to buy any lately? Locking it up in vaults is exactly what we do with it. I challenge anyone to get their hands on so much as single gram of weapon grade uranium or plutonium (bearing in mind that you require multiple kilos of the stuff to make a useful bomb).
Here’s the statement that I would like people to repeat: “Terrorism is not an existential threat, you need to be more afraid of the health implications from eating poorly than from terrorism”.
Listen to Kurzweil, don’t listen to Chicken Little (because he’s got a lousy track record, or depending on your viewpoint, an excellent negative record. If a person is always wrong, then at least you know what isn’t going to happen). The highest probability is that we will get it to work, and that we will make a mess of some kind doing it. We won’t wipe ourselves out and we will figure out how to clean up the mess and prevent it from occurring in future. Electricity, petrol, natural gas and medicine can all kill you - and we use them everyday, everywhere without a significant bodycount occurring.
Getting things wrong is easy, getting things so wrong that you wipe out the human race is not so easy - nobody has managed so far, and it’s not through lack of trying. If screwing up so badly that we kill everyone is a risk that results in great scientific breakthroughs, that is a risk that I’m happy to take. The alternative is to just give up, to just stop the process of discovery - and, to me at least, this is totally unacceptable. Being safe isn’t actually that useful - it’s risk that brings rewards. It is only by taking the risks on the hackable technology that we will gain any measure of safety - not by waiting around for things to be safe enough, because they’re never going to be safe enough. Dangerous is another way of saying “very useful” - knives are sharp, fire burns, etc. Their utility is in their danger - you can cut yourself, but you can also cut other things. Does that mean we should all have the plastic kiddy scissors? Or should we learn how to use knives properly? I know which option I believe is of more use.
There must be something wrong with me, I look at science and think “wow” - I don’t think “OMG we’re all going to DIE!!!”.
October 8th, 2008 at 9:49 am
Hear hear Stuart. An open approach to nanotech development will allow a gradual adaptation to its threats.
October 8th, 2008 at 12:41 pm
Michael writes, “Christine Peterson of the Foresight Institute advocates an open source physical security model, which is helpful, but should be accompanied by more specific recommendations to form a seed around which further ideas can accrete.”
Since it is an open source style project, everyone should feel free to help supply such recommendations! ;^)
For background, see http://www.opensourcesensing.org
–Christine
October 8th, 2008 at 2:27 pm
Don’t network them, then.
Movieland: “Oh no! Hackers are attacking our mainframe! Quick, activate the antivirus executable!”
Realworld: “Oh no! Hackers are attacking our mainfraime!” |pulls ethernet cable out of socket| “Whew.”
October 8th, 2008 at 3:06 pm
Stuart,
I’m not saying that this specific effort will doom us all, but it’s one step in the development of nanofactories, which could be dangerous unless adequately regulated by national and international law.
The fourth paragraph of your comment begins with my statement that unlocked nanofactories are as dangerous as enriched uranium, but instead of arguing that they aren’t, you go on to argue that nuclear war is more dangerous than nuclear terror. I agree with you on that point. Why did you switch to talking about nuke war vs. nuke terror when your objection was originally about nanofactories vs. uranium?
Osama bin Laden has called acquiring nuclear weapons a religious duty. If he could have detonated a nuclear weapon in New York, he certainly would have. Luckily the government, top defense analysts, and both US presidential candidates take the possibility very seriously, even if you don’t.
Unenriched uranium is extremely available and can be bought on eBay. Centrifuges to enrich it are not. When centrifuges or other enrichment devices (many possible methods exist) become cheap, enriched uranium will become much easier to make. Restricted supply will become a thing of the past.
Those predicting doom through scientific causes are rare. Those predicting doom through theological causes are common, because it’s helpful for brainwashing supporters, and there are significant incentives for acquiring new converts. Those predicting possible doom through scientific causes have little to gain from doing so. The motivation comes from scientifically informed genuine concern.
With regard to your last paragraph, it could also be construed as an argument against the Chemical Weapons Convention, Biological Weapons Convention, Outer Space Treaty, and the Comprehensive Test Ban Treaty, all of which ban specific technologies because of their unacceptable danger to human life. The diplomats, political leaders, and scientific experts who draft and sign these treaties obviously would disagree with your implication that risky technologies should always be pursued.
I look at science and think “wow” too. But I also know enough about science to realize that certain scientific advances can be grave threats. It was such sentiments by Albert Einstein that led Roosevelt to develop the atomic bomb before the Nazis did, possibly saving the entire free world. I just examine the evidence and go where it leads me — motivated neither by a habitual pessimism nor reckless optimism. This is called realism.
October 8th, 2008 at 3:29 pm
It is 9.7 million that follows previous $15 million funding for another aspect of zyvex’s three part plan to get to Atomically precise manufacturing.
The Atomically Precised Manufacturing Project currently consists of three coordinated efforts: Micro Automation, Molecularly Precise Tools, and Patterned Atomic Layer Epitaxy. This new funding seems to be focused on molecularly precise tools.
1. Molecularly precise tools [9.7 million for the tooltips]: Zyvex undertook the APM Project in order to deal with the significant limitations that current scanning probe tips and other molecular manipulation tools have placed on science and technology. We are convinced that nano and molecular manipulation technology will not get out of the research labs until molecularly precise tips and other tools are developed. We believe that molecular pick and place will not be viable until dependable molecularly precise tools are available.
2. Atomic layer deposition builds amorphous materials; atomic layer epitaxy (ALE) builds crystalline materials. [previous $15 million was for this part] Start with a protected (passivated) surface: every available bond has a hydrogen atom. If you deprotect the surface, removing the hydrogen, then you can deposit a layer of atoms. If you choose the right precursor gas, you add only one monolayer which is protected as it’s added. Then you can deprotect and add exactly one more layer of atoms. There are a number of precursor gases available. There are literally hundreds of systems to grow things with atomic precision in one dimension.
3. The Micro Automation effort is a result of the realization that affordable molecularly precise manufacturing for many products will only be possible with massive parallelism. Parallel micro-assembly (being supported in part by our NIST-ATP) will develop both the system architecture needed to handle parallel assembly, and the assemblers at the micro scale required to deal with the output of large throughput molecular assemblers.
Zyvex and the universities are also putting some other millions of their own in play for this work. It is a 3-5 year vision and target. It is pretty much focused on working with silicon right now.
$2.5 million for Philip Moriarty to check if Freitas and Merkle diamondoid tool design work.
A lot of difficult work for 5+ years to push these science boulders up the hill. It is when things get to a tipping point and the follow on work becomes easier where the control and safegaurds will matter a lot more.
Funding of real atomic precise capability develop and diamondoid and other capabilities need to gather more of the several billion in what is nanotechnology funding. Right now it is about 1% of the “nanotechnology funding”.
>Electricity, petrol, natural gas and medicine can all kill you - and we use them everyday, everywhere without a significant bodycount occurring.
Actually over 3 million people die each year from diseases and sickness caused by air pollution (world health organization figures.) On average about 14 years premature deaths. Coal and fossil fuels for electricity and transportation.
10,000 or so deaths per year from mining the coal for 50% of world electricity.
1.2 million people die from car accidents worldwide each year.
Significant bodycounts of the total 55 million deaths per year.
Medical mistakes cause 44,000-98000 deaths in the USA alone. 7000 deaths from drug errors.
http://www.fda.gov/fdac/features/2000/500_err.html
October 8th, 2008 at 4:34 pm
I call shenanigans. Nuclear weapons, thanks very much to the poor security measures of the good ol’ CCCP (USSR), have been available on the black market for less than $10M USD, for several decades now.
There’s a very real reason why so-called “weapons of mass destruction” (a weapon is a weapon; death is death) aren’t used in terror. Anyone who uses them for terror purposes will find their cause annihilated by the public reaction — and they know it.
October 8th, 2008 at 11:57 pm
THANK YOU, Brian Wang, for your contribution here. Splendidly concise yet informative.
IConrad, of course, is spot-on right in his observations. This really is the game-theoretic stategic bottom-line.
Most if not all terrorism these days is false-flag black ops anyway, funded ultimately by the “Intelligence” Agencies (cf. the work of Web Tarpley, among others). (There are several tens of billions of bucks “unaccounted for” annually—and this sure as hell ain’t merely bureaucratic incompetency) False-flag terror = Crisis & Leviathan: A pretext for further skipping down the slope toward a Police State (Patriot Act, Military Commissions Acts, good-bye Habeas Corpus, etc., *ad nauseum*). The Power Elite(s) want a global (Corporate State) police state. Fortunately, there is, of course, a countermovement to this. But the Power Elite(s) would like nothing better than to control and totalitarianly-utilize both [meso-size, i.e., Moravecian] robotech, as well as nanotech. We won’t let this happen, of course ;)…
Ciao…
October 9th, 2008 at 4:05 pm
Richardson is only about 30 or 35 miles from my house…yeah, there’s cool stuff happenin’ in DFW…but I concur: Ideally, we need FAGI developed first, or at least concurrently.