Angry Planet - How 3-D Printers Could Spread Weapons of Mass Destruction

Episode Date: December 6, 2019

We are living on the precipice of a manufacturing revolution. 3D Printing is more than just a hobby your weird rich friends use to make guns, it’s the dawn of a new process of prototype and creation. Additive manufacturing, if you believe the hype, will make it easier to make everything. Including weapons of Mass Destruction. But how real is that threat and how close are we to making chemical weapons at home for fun and profit?Here to help answer that question is Grant Christopher. Grant is a senior researcher at Vertic, a London based think tank that works to strengthen verification of international agreements. He was part of the research team at CERN that discovered the Higgs Boson, and he’s the co-author of a research paper titled WMD Capabilities Enabled by Additive Manufacturing.Support this show http://supporter.acast.com/warcollege. Hosted on Acast. See acast.com/privacy for more information.

Love this podcast? Support this show through the ACAST supporter feature. It's up to you how much you give, and there's no regular commitment. Just click the link in the show description to support now. You know, there are things that you can make with additive manufacturing. You can't make any other way. So it's really hard to understand where that's going to go and how it's going to be used. You're listening to War College, a weekly podcast that brings you the stories from behind the front lines.

Here are your hosts. Hello, welcome to War College. I am your host, Matthew Galt. We are living on the precipice of a manufacturing revolution. 3D printing is more than just a hobby your weird rich friend used to make guns. It's the dawn of a new process of prototype and creation. Additive manufacturing, if you believe the hype, will make it easier to make everything, including weapons of mass destruction.

But how real is that threat and how closer we to making chemical weapons at home for fun? and profit. Here to help us answer that question is Grant Christopher. Grant is a senior researcher at Verdict, a London-based think tank that works to strengthen verification of international agreements. He was also part of the research team at CERN that discovered the Higgs-Bosen, and he's the co-author of a research paper titled WMD Capabilities, Enabled by Additive Manufacturing. Grant, thank you so much for being here. Thank you. Absolutely pleasure. All right, so let's do some basic definition stuff up at the top, what exactly is additive manufacturing? What do we mean when we say that?

Yeah, it's really an umbrella term. It means quite a lot of things. So picture two things. Picture Michelangelo is David, and then picture a model you made of Legos when you're a kid. So additive manufacturing is like the Legos. You take some smaller things and you build up a bigger thing. from them. And in additive manufacturing, you'd glue them together. You'd use a laser to melt the bricks together, basically.

The David is more an idea of conventional manufacturing, subtractive manufacturing, where you have a block of marble and you reveal the structure of the David from inside it with some skill. or in modern manufacturing you could do it with a machine tool. So additive, you build up from smaller things, conventional, you take larger things and shave off pieces. And when we say 3D printing, can you use that interchangeably with additive manufacturing, or is there any kind of distinct difference between these two? I tend to use them interchangeably. There may be some, you know, edge definitions where people stick with, but I don't really see an issue with one or the other.

All right. Great. One more definitional question up here at the top, which I think is kind of a theme or a broad framework for how we can kind of look at this issue. And that's wicked problem. Can you explain to the audience what we mean when we talk about a wicked problem? and how it applies to specifically additive manufacturing. Yes, I mean, I would think of wicked problems more in... You combine several forces or several big concepts and how they relate. And you change factors in one, and it really doesn't... It's hard to understand how that will affect the big picture. So nuclear disarmament is a wicked problem, for instance,

because of the technical side of nuclear disarmament, the political side, and then the environmental factors around it. So the politics may be hard at any time, but the background politics might change significantly from one year to the other. In additive manufacturing, well, it's technically very difficult, depending on the process you're talking about. what you might picture before you started this podcast of being additive manufacturing. So a garage-based plastic printer is very different to actually solving an engineering problem using a three-quarters of a million-dollar metal printer.

Getting those things to work and to print as you want them is extremely complicated. then getting that onto a, let's say, a military system is also extremely difficult, or a civilian nuclear reactor. Those things need to be checked very carefully that they're not going to break. They need to be as reliable as a part manufactured any other way. So that's kind of the wicked side. And then the politics angle, you know, the underlying factors, I don't know. know. Those we haven't really gotten a good example of yet.

Right, because all of this stuff is still so new. I mean, the Pentagon just announced that they were kind of going to be using it officially just a few months ago. Right. This is not something that's widespread, not widely adopted. Everyone's still kind of, we're still kind of in the hype figuring things out phase of additive manufacturing, correct? Yeah, and it goes into these categories of emerging technologies. So what are the other technologies kind of thrown in this mix? Anything with it, and when you're talking about researching emerging technologies from the military sense, anything cyber, anything AI, blockchain, you know, all of them. additive manufacturing is a little different. It's still very much emerging, but you are seeing applications.

So on the emerging side, you would say technology is emerging where it's still changing over time and you haven't seen the most important application of it yet. That's kind of like the hand-wavy definition, I think, can be quite useful in a conversation to define it. So with that, right, you could think of the internet itself.

So when I was in college, the internet didn't have a lot to do with YouTube, but I think it would be almost absurd to think about the internet without its video component today and social media component. And those didn't really exist. So the internet was a continuously changing technology. but we're in like a certain period right now where a lot of important uses on social media and video messaging and people being uploaded video content for additive manufacturing we are in a phase where there is a significant active metal manufacturing industry there is a lot of of interest in trying to manufacture in in different ways so using one of the additive manufacturing techniques that the hobby one that the the plastics hobby one to

make things like explosives and magnets and just print any material that you can you can melt and extrude so we don't really know where any of this is going and there isn't an industry leader right there isn't an industry leader that is gobbling up all the smaller companies, our Apple, Microsoft, or, you know, pick your industry, there's still quite a competitive marketplace. No one's technology has proven to be a lot better than everyone else's yet. So I think that's another important factor in the emerging technology stakes.

So I want to point out that, again, I don't know, I think some of the stories that filter down to most people about additive manufacturing 3D printing are, you know, people using them to make guns or look at this cool sculpture I made with 3D printing. But you do have big companies with big money who are figuring out some interesting applications. The two that really stick out to me are Raytheon, who can produce something like 80% of one of their missiles using 3D printing. And the Los Alamos National Laboratory can manufacture high explosives with,

this tech, correct? Yeah. I mean, I think these are absolutely fascinating, right? So the Raytheon example, so they're a company out in California, they're trying to print most of the parts from missile.

So this is not, you know, T, ill, gray, hot, you know, press a button, print a missile. This is printing component separately and then assembling them. But the most important question about this, is it economically viable or is it a prototype?

Is it a, you know, not quite a publicity stunt, but is it a proof of concept? Or is it really a revolution in manufacturing missiles and something that's going to be economically competitive. If that's the case, then that is a really big deal. From the work I've done, I wasn't necessarily looking at the economics, but I was looking at, you know, what is possible. Additive manufacturing is hard and slow. So it's often spoke of as you're able to rapidly prototype. So that means you can iterate on designs quickly. That doesn't mean that. the process of printing itself is quick. Printing like a 20-centimeter metal part in one of these expensive metal printers takes

the best part of a week. And unless there have been advances on making this process more reliable, but chances are that your part might have broken while it was printing, these parts need to go through a very extensive quality control process. and additive manufacturing is hard and still it's hard to print something on the same machine twice this was something that I learned in speaking to practitioners a couple of years ago and I think still is largely true this will not be true forever right if you are interested in this industry this is one of the big problems you want to figure out how to make this more reliable

in general whether you're printing missiles, guns or anything else. On the gun side, so one of the first guns printed was plastic and there have been some of the pioneering folks out there that have tried to design metal guns. I wouldn't recommend trying to use or fire these for the same reasons I noted above unless you've got a CAT scanner that's willing to, that's capable of internally verifying that your gunwork split apart and seriously harm you the first time you try and use it.

That being said, it has been a major concern for security. And I mean, it is one of the things that consistently gets coverage and gets spoken about in when discussing additive manufacturing. I find it quite curious that it's a subjective conversation in the United States, but for Europe at least, where there are much low rates of gun ownership and much tighter controls on who can own weapons. It maybe is more of a concern.

What is it about the machines that makes it hard to print the same thing twice? the underlying physics of rapidly melting and cooling a 16 micrometer metal powder and then layering surface by surface over and over again it's extremely complicated so much so that the US national laboratories have invested significant resources in trying to understand the process they've been using the high-speed cameras that have been used for nuclear weapons

explosives testing, like that level of great equipment to understand the very quick, the very rapid heating cooling processes that go on when you put a laser on some powder and it melts and then solidifies again. It's very easy to form.

It's very easy to be warped. It's very hard to get definition around corners and overhangs. It's very difficult to understand the best way to print something. So think about going back to constructing something with Lego. You've got one laser and you're making a, I don't know, let's say, picture of a, Lego spaceship or something like that. That laser has to scan over that each layer of powder of each point that you want to stick together.

It's got to melt the entire thing. Between one layer of powder and another, you've got, say, 60, 100 micrometers. So you're doing that over and over and over again. That is complicated, and it's not obvious that all your metal powder is going to going to glue itself together. There are going to be maybe holes in the middle if you don't control the process properly. So when you put

a lot of stress on it, i.e. try and do some ballistics, the whole thing might just break apart. We're going to pause for a break reel here. We are on with Grant Christopher talking about additive manufacturing or 3D printing and it's possible use in creating weapons of mass destruction, which honestly

at this point sounds like a pipe dream. All right, War College listeners, we are back on with Grant Christopher talking about weapons of mass destruction and how the creation of such may be aided by 3D printing. Sir, just before we broke, you've been making it sound like there's nothing to be worried about. Everything's going to be fine. This technology is years away from producing anything of value, let alone, you know, a nuke or a sarin gas. factory. I think it's a little bit more complicated. So if we look at the most advanced applications of 3D printing as it goes into aerospace,

the military, and any of the high-end users, we did talk about someone is trying to print missiles. So don't discount that. If this couldn't be done, then that company wouldn't still be operating, right? secondly um g e have a factory in alabama um

they have certified aircraft parts they have jet engine nozzles they've printed tens of thousands of these at this factory so this is something that you can do at some scale if you know the secret um you know i haven't been allowed to roam around their IP

and exactly figure out how they solve this problem um but there are solutions to be to be had here, I do know that, you know, there's still some tips from dentistry. So they took some advances that had been done in printing of particular metals for medicine and dentistry and you kind of use those adaptations to get a march, get a couple year head start on other people trying to develop new materials. So this may not be here right now, but it may be coming. The other side of things is it's got to be really clear.

To make nuclear weapons, you have to have fissile material. You may be able to work with fissile material in printers. You'd have to think carefully about criticality, but you can't manufacture it in the printer. You've got to manufacture it the usual way. you've got to use enrichment or reprocess plutonium coming out of reactor. And you can't easily 3D print those things. That probably isn't possible. So the impact is going to be some other different ways.

So when it comes to weapons of mass destruction, how do we get them? So nuclear weapons, you need to have uranium. You need to do some chemical processes to it. and then you need to enrich it. The way that we currently control that process, so we try and keep tabs on which states are doing it, the IAEA inspects them, and we also have export controls.

So they're not necessarily to completely halt the spread of all the material and equipment and metals and machine tools you need to carry out these processes, but they're an inherent. part of a regime to monitor and make it difficult for people to do this. Addisive manufacturing could undermine and get round things. So, okay, you can't get hold of something the normal way. Maybe you try 3D printing it instead. So countries like Iran, North Korea, maybe they're looking at advances in the West and seeing how this go. We don't have a lot of evidence for that.

but we wouldn't necessarily expect to see much evidence for it. They're going to be keeping their card to quite close the chest in this respect. But nearly every country in the world is looking at

how to use this for their military. They're looking at civilian applications, civilian nuclear applications. They're going to be looking at manufacturing weapons, manufacturing delivery systems.

So that's just on the nuclear side. On the chemical and bio side, there are other emerging technologies that make 3D printing a bit more of an interesting place. So it's the gene assembly technologies and your chemical printing technologies that are quite interesting

that make additive manufacturing more of a more of a player um so what i mean by that so you've got to you've got to have the the virus or the the thing you're going to use as a bio weapon you've got to you've got to have the chemical but to make your chemical chemistry lab to make your bio lab um you need to work with special materials um you know no one's trying to print with these as far as my studies have shown but the special materials you need to assemble these laboratories are being worked with. Moreover, if I'm, like, saying a university researcher in a lab, I'm going to have had all kinds of, you know, tools and custom things I'm going to build in my lab.

Every one of these university labs is going to have a 3D printer or two to make custom tools. So there's going to be a lot of proliferation of expertise. have to work with these things. That's just on the kind of the lab research level. So I mentioned a little before about the companies that are working on 3D printing, just like any new area. Think of like the early part of the web. Like companies are popping in and out of existence,

like nobody's business. So there's plenty of expertise floating around, you know, that maybe that can be funneled off for perhaps nefarious purposes. This has been an off-studied thing in proliferation analyses of, you know, where do people go that were involved with weapons programs? It's like, well, where do people go that were involved with advanced machine tooling applications that have floating expertise in the world? And a third point or a fourth point, I'm not sure which one I'm on.

if you go online and you're like okay I want to get a 3D printed like metal cap to give to my kid or something these machine shops are all around the world and you can send a design and they'll print it for you

and send it to you so I would want those to be monitored and looked at very carefully that they know exactly what they're printing and who their customers are because they're not all in Western Europe

in the United States. I think those are the kind of main points like really underlining that unlike quite a lot of advanced technologies, key developments are not happening and not restricted to

the West. So Europe, the United States, North America and East Asian allies, there are key developments in China, key developments in Russia, and key developments elsewhere. The US and Europe and Germany in particular certainly leaders in this technology, but there are lots of interesting areas to be explored, and the Russian National Atomic Company, Ross Asim, has and is developing 3D printers. China is leading in quite a few areas.

They've developed titanium parts for their aircraft using 3D printers. So this is happening and this is happening globally. And it's not so easy to keep tabs on how things are going to be developed. And things aren't necessarily going to be developed in a way that we understand based on our experiences with the technology. It's completely changing the supply chain. Right, which is the supply chain, I think, is one of these really boring, unfortunately, topics that's actually very important. Because kind of as you've alluded to, part of the way the international community keeps tabs on all this stuff is by monitoring the supply chain, right?

Exactly. So I think that's the big story of additive manufacturing is that it hides the supply. chain from the international community? It certainly could. In terms of, you know, there are a lot of the big manufacturers of the machine tools themselves, of the princes. They're still in

Europe and North America mostly. But you can have, you can manufacture locally things that you wouldn't necessarily be able to manufacture locally before. Exactly. What do you think is the timeline on this. Do you have any sense of how

how worried we should be and when things will get when things will get worse from a proliferation standpoint? Do you have any idea or is it all just guesswork right now? So I've been looking at this for about four years and I've seen real wide eye-opening changes in that time.

Things do move pretty fast. So, for example, you can print Teflon now. That's quite an interesting material from a nuclear proliferation point of view. It's resistant to uranium hexafluoride. So that's normally the chemical form of uranium you do is for enrichment. You know, it wasn't invented for that purpose. It was invented by a company that makes a lot of Teflon. And they just thought, we're going to invent this new printing technique.

So new printing techniques and new materials are being invented all the time. And I see applications and relevant applications that surprise me. What is the national security community doing? Well, we're not behind. but if there was an urgent requirement to act, it would probably be at the level of export controls and monitoring. But, I mean, I don't think we've missed the boat. I don't think we're in trouble.

But yeah, then again, on this chem bio side, I have a lot of uncertainties there about how 3D printing is going to help these kind of chemistry in the basement guys. So that definitely is a not very well understood issue. Do you think the threat is greater from state actors or non-state actors, these chemistry in the basement guys?

Or do you think they're just different kinds of threats? So if we think about ISIS, we talk of them like a non-state actor, but they were essentially they're a bit more like a state actor right they control territory they had um scientists working for them and and they were working on working with chemical weapons um for small groups of people i mean i think that the critical technology is going to be um the the chemical printing and the bioprinting type of technologies uh which is you know they're very very important, but they haven't necessarily been my area of focus. I'm a physicist by training,

not a biologist and a chemist. So coupling those together, you have some uncertainty, right? You've got two technologies developing at the same time. So it's kind of hard to figure out where they're going to land. So pulling all the way back to the beginning with AI, AI is going to be used very heavily in 3D printing, not just to improve the reliability, but the designing of components. And this may allow, you know, deployment in weapon systems from non-state access

this may really level the playing field in terms of design in a way that we didn't really understand. And in my first couple rounds of looking at this, I really missed this fact. And it took a colleague to point out this concept of generative design, which will be used in all design areas, will be particularly interesting for additive manufacturing because additive manufacturing just opens up a completely new design space. You know, there are things that you can make with additive manufacturing. You can't make any other way. So it's really hard to understand where that's going to go and how it's going to be used.

So that's, I mean, from an analyst, it's an interesting problem from a security point of view. You certainly would want answers to these questions. All right. Let's back up just a minute. What are the things that can be manufactured this way that can't be manufactured any other way? Right. So you can make things with very complicated internal geometries. So what you would do the normal way

is you make 30 different pieces and you stick them all together. Additive manufacturing, you do it all in one go. So complex internal cooling systems, things like that. You can make new parts that are just lighter. So you use some algorithm to tell you, okay, here are all the places in your part that you didn't actually need. like the physics and engineering algorithms tell you that these are superfluous

so you cut all of them out unfortunately we can't share pictures this is a podcast but you end up with kind of weird HR Geiger type looking things like very strange almost organic looking designs because you've just designed something

really differently so that's you know taking one material the other new things are because of this the metal printing with rapid heating cooling a lot of the thought so far has been kind of going

against you're trying to reproduce a material you already know how to use with a 3D printer so you're kind of pushing against the technology so rather than do that you could kind of work with the technology and use that rapid heating cooling of your lizard to be advantageous.

You could use it to blend different materials together. So it becomes very easy to have a material that slowly changes composition as you go over the full course of the material. So I don't know. Thinking of my, I'm sitting right in front of my

computer here. So you think of like the part of the computer right at the top of the keypad being titanium and at the bottom it's made of a completely different metal and in between you've got like a little bit of both so you're blending them together. Additive manufacturing allows that kind of material, a functionally graded material.

So there are just new types of things that can be made and how those are going to be applied, I do not know. It sounds like a new forge. But anyway, tell me what generative design is, and why AI is an important factor in it? Yeah, so I'm quite new to generative design.

So a colleague on one of my paper friends really highlighted this. So you let the computer design the part for you. You kind of tell it roughly what you want and it solves all the physics and

engineering. And it can give you something that didn't necessarily, you're surprised how it looks, you're surprised how it is, but it does the job as it was asked. So, a simple thing like a

bracket, you could generate it or design it, that's quite easy so you're a bracket that holds, holds two corners of something together, it attaches two pieces together, fine. What about if you have a more complex problem as we, as we master this,

technology, like a component of a weapon, rocket fuel, nozzles, any kind of weapons part, you know, we come back to the gun problem in a world where we have more reliable 3D printing, in a world where the technology is more accessible, you can, you know, maybe mail order it easier, where everyone has access to the software on their home computer, that's suddenly a bigger problem. And the timeframe question you asked me is difficult. But I think it depends on, you know, we can track how AI develops and how generative design develops

kind of independently of additive manufacturing. but this is a 10 years kind of thing, 510 years kind of thing. And 10 years isn't that long ago. 10 years was 2009, and it's like the first year of the Obama presidency of something. So it really isn't that long. The other thing that strikes me about all this is that we tend to be when dealing with new technologies and new threats reactionary. Yeah. So do you think that it's going to take something bad happening before we, I mean, even figure out how to handle it or even really know what we're looking for? Yeah. I think, you know, how long does it, when do people actually kind of, the English phrases, pull their finger out and actually deal with the problem?

Like, when do you get the political will to actually address something, a Sputnik moment or, um, um, You know, any other crisis you care to name. So hopefully, I mean, I can convince you that I'm not the only person looking at this. I'm in the NGO community, so I have very little influence with government. But, you know, we, you know, we can publish good research and try and make sure we keep tabs on the problem. But, yeah, it could be a real issue because I don't, I don't have access to all the proprietary information. and I don't understand the problem completely and probably can't. And this isn't my primary area of research anymore.

So it really could be something. And I also think that there are other emerging technologies that have really risen to the top of that floated to the top of the pile in terms of people think they're much more threatening. So anything with the word cyber in it is like catnip to the national security community at the moment and artificial intelligence. So we just need to make sure that we don't neglect very important technologies like additive manufacturing that

like you say it's it's about the supply chain um that's where the big impact is going to be felt um so looking like beyond wmd there's going to be a significant impact in uh in in in militaries and how they how they supply themselves and how they um keep inventories of of spare parts and things like that there's going to be um you know possibly a quite big effect on military logistics from this kind of technology. So there will be big impacts. Grant Christopher, thank you so much for coming

onto War College and walking us through this complicated and emerging threat. Thank you very much and apologies for being so technical, but it's a technical problem. That's it for this week, War College listeners, thank you for tuning in. War College

is me, Matthew Galt, and Kevin Nodell, was created by myself and Jason Field. We will be running episodes Throughout the rest of the holiday But, you know, it's Christmas It's downtime Everyone needs a break

Probably going to hear some reruns in the near future We are still working on new episodes though Including that Metal Gear Solid episode Which I promise Is really coming If you like what we do Please like and subscribe On iTunes

Wherever else fine pods are casted Leave a review We do read them And it does help other people Find the show You can follow us on Twitter at War underscore College. I am at MJG, AULT. Kevin is at KJK Nodell.

We'll back next week. Stay safe until then.