The Quantum Stack and the Countdown to Q-Day

Welcome to the Cogs of War podcast. I'm Jonathan Panter, Executive Editor of the Cogs of War Vertical, brought to you by War on the Rocks and Booz Allen Hamilton. Today, we will be discussing quantum computing and post-quantum cryptography. We are joined by Cameron Chera, Ben Gianni, and Dr. J.D. Dalny. Cameron Chera is Vice President of Engineering and Chief Technology Officer at IonQ, a company building trapped ion quantum computers. Ben Gianni is Senior Vice President and Chief Technology Officer at General Dynamics Information Technology. Dr. J.D. Dalny is Senior Vice President and the Leader of the Quantum Practice at Booz Allen Hamilton. Gentlemen, thank you for joining Cogs of War. Thank you. Because this is a complex topic around which there's a great deal of hype, maybe we could start out with our one physicist in the room, J.D., telling us a little bit what quantum means, broadly speaking. Right. So when we go into physics, typically there's several categories of physics. There's classical physics, which is what you think about, most people think about when you hear physics, ball flies through the air at this angle, how far does it go? Thermodynamics, electrodynamics, and then quantum physics. Quantum is the physics of small things, of fast things. And when we talk about quantum in this context, we're saying, how do we harness this corner of physics to provide some advantage? Whether it's an advantage in computing, or an advantage in sensing, or an advantage in how information is transmitted, what can we do with this? Right. So typically when I talk about quantum, I talk about three fields. Quantum computing, which is what a lot of people think of when they hear this, and that's probably what I would call the default, which is using quantum physics to do very specific things in computing better, not everything. Okay. Next would be quantum communications, which some people would say includes post-quantum cryptography. Meaning transmitting quantum information, which is in of itself a challenge, but also how do we protect the information we're transmitting today from a future in which an attacker has a quantum computer, right? So that's, we call that post-quantum cryptography. And then finally, there's quantum sensing, which is detecting things like magnetic fields or movement, using quantum effects in ways that make the sensor more delicate, more sensitive, but for all these things, they're not going to be that sensitive. And then finally, there's quantum sensing, which is more sensitive, but for all these things, they're not going to be that sensitive, but for all these things, they're not going to be that sensitive, but for all these things, there's nuance and trade-off. And at Booz Allen Hamilton, are you working on all three of those, or do you focus particularly on computing? We focus on all three. We have to, because right now, all three of these big areas are relevant to our customers' missions, and we need to be thinking about how do we bring these technologies to those missions? Because this is, I think everybody here would agree, this is the hard stuff. This is the true cutting edge, and how we learn to harness this stuff for some sort of impact is the value everybody's seeking right now. And what exactly is the value that you're seeking right now? And what exactly does Booz do for its customers with respect to the quantum field? Right. So when we engage, we always think about, all right, mission first, what are you trying to do? And let's think about how quantum can help accomplish that mission. Right now, computing, as anyone would tell you that's in the field, is the big rush is to get computing to usable, right? We do things on quantum computers, and they are extremely real. The physics is happening, and we've had many incredible moments where quantum computing has shown again and again that you better get ready for us, we're coming, right? But at the moment, you're not plugging your problem into quantum computing, and the problem is going away, right? It's still at the research level, and I would say engineering challenges remain. For us, for example, a concrete thing we work on is clients need to move to post-quantum cryptography. They want to upgrade their systems, they want to be safe against that future attacker who has some sort of tools that maybe they haven't told us they have yet. We're always looking for the next threat. Maybe they're holding data now to decrypt later. And so we move them to those new protocols, and we test their systems and those sorts of things. All right, well, we're going to get to this question of harvest now, decrypt later, and what that putative Q-day would look like. But let's go next to the guy, Cameron, who's actually building the computer that could do this. So what do you work on at IonQ? What's the approach? It's pretty incredible. As a non-physicist, as a classic IT guy that's coming in to really help a company scale, I recognize the same thing JD recognized. And I hate to say it this way, but we all grew up with this as science fiction, but it is here. It's very practical. At IonQ, we've been building systems literally for over a decade. So when you think of the qubits and all the things we've talked about that are necessary to get to what is happening now, which is Q-day, it's becoming very much a reality as JD laid out the landscape for us. So we've been building systems. We're about to deliver a 256-qubit machine, which gets us one step closer to this Q-day, which is when post-quantum cryptography becomes very influential because the current levels of encryption we have will be broken. It's not if, it's simply when. And there's new attacks already emerging. You very astutely talked about HarvestNow, DecryptLater. We're starting to see TrustNow, ForgeLater emerge with digital signature and those types of things that could affect crypto markets, other e-signatures that are required. So it's a very real field. BQC, QKD, which is quantum key distribution, is a very real thing now to protect systems. In the compute side, we're making great strides. There are, as JD offered, limited use cases, not broad operational application yet. But when you marry it with AI, classical computing, and the current infrastructure that missions use, it becomes a very powerful tool. So we are embarking on this journey rather quickly. And then on the sensing piece, gravimeters, atomic clocks to get better than the timing we have now, timing is to be exact. We get better sensor agreement in the multi-int that we collect for sensors. We start to get better alignment and coordination as far as order of battle. So there's a lot of advantage that missions can look at today. It is still frontier tech. However, it's giving us an advantage. And that's the real key, is the decision advantage. There's so many things I want to ask you about what you just said. But I think maybe for our listeners, it would help. When you say a 256-qubit machine, perhaps you could correct me if this is the wrong way of thinking about it, but how many qubits would be required for a cryptographically relevant machine, one that can run Shor's algorithm and break modern encryption as we know it? So how far away are we, or is that the wrong way of looking at it? That's a brilliant question. Actually, it's a very good way to look at it. I'll look to my cohort here, JD, to help me along. The original hypothesis was you need about 1,000 to 10,000 logical working qubits to be able to crack encryption. What we're finding in practicality, and this is a real key differentiator, with our modality, which is natural trapped ions, we've made a lot of progress in a very short period of time, is it's requiring less qubits, less logical qubits, to have better operational output. So we believe Q-Day is moving left. I think the recent executive orders have also kind of shown a light on that with putting a date and a solid time in the ground as 28. Was IonQ one of the companies that the government took an equity stake in? We were not initially. We are involved right now, and that can happily report. We have successfully acquired Skywater Technologies, which is the world's leading fab for these types of QPUs or quantum chips. And it also has a domestic supply chain. So at the time, we couldn't participate in that phase because we were embroiled in that acquisition, which is fine. These are the things that companies do to build capability on behalf of the United States government in support of the mission. But the modality is working. And to your question, crisply, it's really 10,000 logical qubits is what people initially believed. What we're finding on quantum relevant systems, it's happening a lot sooner than we anticipated. We believe you need a lot less logical qubits now. And the early signs we're seeing with applying algorithms to the actual physical medium that we are working with, we're seeing better than anticipated results. Okay, so Q-Day is moving left in your estimation. And then, of course, we have Ben over at General Dynamics. You guys are all over this, right? You're as we were discussing before, you're primarily focused on assisting your clients with post-quantum cryptography. We are. And as an industrial engineer and a computer scientist at General Dynamics, we worry a lot about the practical application of technologies into our customers' missions. And so that's across the federal, state, local, even tribal governments, whatever's important to them is important to us from a mission perspective. But on the post-quantum crypto, well, we're really looking at the full stack. So our belief is that you've got to have compute and you've got to have sensing and you've got to have communications and networking in a full stack to be able to be relevant in the industry. Because our job is to integrate such things together to make it work, secure them, make them accomplish the mission objectives at the time and place of need. And without working with all aspects of quantum, you won't get that. And especially working what we call hybrid systems. So quantum, classical, also GPUs, right? So CPU, GPU, GPU is the ultimate hybrid that we set up. We see happening. On the post-quantum cryptography thing, it is not hype. It is a mandate. It is real. It is one, the way we think about it is we've got to stop the bleeding. So you mentioned harvest now and decrypt later. Well, you know, for some time now, adversaries have been harvesting information. And so the more time that elapses between us and PQC mitigation means that more secrets are being harvested. And the more recent those secrets are, likely the more relevant they become to national security, even the old secret. CRITSERV sometimes, you know, causing grave and serious damage to national security and economic security. So our belief is that it is here. We need to deal with it now. And that is a reality. And I can piggyback on what he was saying about the number of qubits for to crack crypto. This is one of the big questions. And it's a bit of a gotcha because the only thing we know is the number is going down. OK, because there's as much innovation. Everybody focuses on the hardware. There's as much innovation on the algorithm and execution and software side. That's we're having eureka moments in the field, which I think is fascinating. We're seeing steps down in what we thought we needed to execute these big algorithms at scale. And it's moving everything to the left, which just increases the urgency. What's fascinating about what JD is saying here is now with AI basically at our doorstep, everyone got, I think, caught off guard by that bow wave. Quantum is the next one, but this one is going to be far different because it's not just going to be software based. That quantum leap I jokingly talked about is happening in a very real sense. We have to think about this of going when we were doing punch cards to then the first semiconductor, the 8008, right, from Intel. That was a true leap forward because then from that point, it was frontier tech. But then the generations of development have brought us back. And that's what we're seeing. We're now using AI to assist us in algorithm development, software development, applying it to the quantum challenge. And then on the machines we've built prior, we're leveraging that quantum technology to help us advance quantum technology faster. So we are really in this incredible innovation loop and do loop of innovation that is happening more rapidly than I think we've ever seen in the last 50 years. But in your case, Cameron, the natural trap die on, full disclosure, GDAT has a collaboration agreement and a strong relationship with IonQ and College Park. One of the reasons for that is that we saw the trap die on modality of quantum compute, maybe first to market and first to a production environment. Do you think that's true based on where you're at and why do you think that is? So we offer that, you know, thanks to the government and a lot of the government programs that this was all born out of, Chris Monroe, one of our co-founders in 1995, was the one to really start looking at natural trapped ions. But to complement your point, Ben, which I think is really powerful, it's not just about bringing quantum computing to the market. There are other companies very successfully looking at it, but they've not graduated from a lab environment to actually building machines at an operational scale. I think the advantage of the natural trapped ion modality is I don't need cryogenic chandeliers. I don't need exquisite things in the data center. I'm not a suspect to vibrations and reverberations and all these other things that when you look at other modalities, you're not a suspect to vibrations and reverberations. But when you look at other modalities, as they add up, tend to be more exquisite in nature. We all know from supporting the mission environment pretty proudly in our careers, they don't live in an exquisite environment. They live in an environment where we constantly have to talk about, is your tech ready for an operational setting? And that's very, very different. And as a company, we think about how do we commercialize this? How do we industrialize it and allow it to be ready for that operational setting? Because that's the most critical component. And it might be worth just giving a few seconds on what modality means. So when we talk about quantum computers, right, we talk about doing things with quantum systems, meaning we have some objects, maybe they're bits of light. That's a photonic quantum computer. Maybe they're trapped ions and an ion is just a atom with a charge. Maybe they're neutral atoms, right? And maybe they're little loops of electricity. They call that superconducting. And those are the ones you typically see. And then you have the quantum systems, right? And then you have the quantum computers. And if you see a picture of a quantum computer, you see these big golden chandelier looking things. Yeah, I've seen that. That was in black. Yeah, that is a dilution refrigerator. Big show, by the way. Which is basically a tool that cools a volume of space, very, very cold, millikilvin temperatures, colder than outer space. And all of these things are to say, if I want to make a qubit, a quantum bit, I need a quantum thing, and then I need to be able to control it. A lot of companies like IonQ, others have said, we're throwing our weight behind a specific modality, right? And that is super noble. I want everybody to win. I want quantum computers to be real. And at the end of the day, if you think about modern computers, you just use your iPhone or use your Android. They're not actually fundamentally different. They're silicon, right? We've determined the best way to make classical computers. There's a lot to unpack in that sentence, but basically, we're not fundamentally changing the core physics, right? Whereas with quantum computers, we haven't, I think, fully gotten to one winner. And that's why this time is so important. Interesting. So when you hear modalities, think types of quantum computer. So if one modality pulls ahead and wins this race compared to others, be it the trapped ion, or be it the super cooled environments and so on, does that change what applications will be available? I can imagine, let's say, and people talk, at least in my world, a lot about quantum sensing. Would we be able to have submarines that can navigate in the absence of their traditional inputs, right? And so something that is controllable at room temperature, right, might be more applicable in that environment. But are there benefits to these other modalities? Oh, yeah. I mean, we could do a whole podcast on the puts and takes of these different modalities. But I think for the listeners, the jury is out on an absolute winner. Let's put it that way. I think there's very good arguments for different modalities, and people feel very strongly about why one versus the other. Oh, this one doesn't have to be as cold, or this one has a longer coherence time, which is kind of how long it stays quantum. But I would say, in the end, technology in this space is going to go to be a magical black box that you plug numbers in, and out comes an answer that helps you with your problem. That's where we're going. Kind of think of it like ending up like a smartphone. That's a gross simplification. But ultimately, that's what we want as a field. We want a usable device. And so what modality wins, or if a modality wins, because that may not happen, and several may do very well, is yet to be seen. But the applications in the end are all going to be dependent on what algorithms we figure out we can run on them that have advantage. And to that point, if you think about where compute is needed, where quantum compute is going to be needed, where there is high-rate data that needs to be turned around quickly for decision advantage, sometimes we're at the edge, which means we're in the field somewhere. And it could be a very hostile environment. It could be hot, dirty, and could be very cold. And so there may be multiple modalities that fit, you know, from the enterprise to the edge of the mission, either in a data center, conventional setting, or at the edge, more unconventional settings. So it is feasible that multiple modalities, you know, get packaged differently, have different cost factors, and have different operating factors. One thing I want to add, though, you touched on something really important. I think JD and Ben began to unpack it. I think this is something that we should maybe articulate for the listeners, because it's a very important thing you've tapped into here. For me, what's most important is the only way to do it. I think the only winner in this has got to be the United States, because a world where the United States doesn't win the quantum race looks extraordinarily different than what we enjoy today. What does that look like? So we know about, we've mentioned harvest now, decrypt later. The Chinese are hoovering up all sorts of data. They beat us to quantum compute. They decrypt all that. Okay. So that's one potential terrible outcome. What other sort of ways might it be disadvantageous to the United States if other countries pull ahead in this race? I mean, we can all probably address this, but cybersecurity is one of the areas where as a nation, U.S. government and military intelligence communities have their Title 10, Title 50 of the U.S. Code statutes to say, "We have the authority to conduct information warfare with cybersecurity." So if you look at what's happening with AI accelerating pace and the quality and the creativeness at which cyber can be prosecuted both offensively and defensively, quantum might. It might be the same thing. And I agree with Cameron 1,000% is the U.S. has to win this. So one of the things that probably should happen is that we develop the most powerful quantum computers and we show what they're capable of doing, which is doing damage from a cyber perspective, but then choose not to use it as a deterrent. And I think like we have the nuclear deterrent, quantum deterrent comes to my mind as a national capability that we and only we should be getting all we can. Trevor Burrus: That's a good point. I think it's a good point. It's always the most powerful, the fastest, the cheapest, the most usable. Tom Hanks: Look, deterrent is good. We always know peace through strength has been proven through generations that it works at times and not at other times. To Ben's point, post-World War II, when the United States set up, let's say the Navy, the United States Navy to help patrol the oceans, what benefit did that offer to people outside the United States? It allowed Europe to do commerce, Asia into Pacific. It allowed Europe to begin to reestablish economies and all these other types of things. When you think of winning the quantum race, it is exactly as Ben described. It's being able to set the tone globally that is also for economic development, for safety, for sovereignty, because it's not just about the United States' sovereignty. It's about our allies' sovereignty and the other things, the other people that want to participate in a positive way in the world. So that's really one element about that race. To unpack the second part of your comment about, does a modality win? Not necessarily, but there will be a modality that will turn into what we know in classical compute today. about compute, we don't say it's a CPU, we say I want AMD or Intel. When we think of GPUs, we go, I need an Nvidia card, right? That will occur in quantum. And what will drive that is the pervasiveness of the adoption. And can you drive the cost to a consumerized nature so that it could be consumed more broadly than the other modalities? I think those are the elements you might be seeking in that answer that I think is really going to drive the stronger adoption of which modality will be adopted further. But those initial customers are likely just going to be government customers, national laboratories, and academic institutions, correct? Believe it or not, we are seeing uptick in financial services, healthcare, and other industries call it Fortune 1000 enterprise customers. I initially thought, I'll be very candid with you, when I joined IonQ, I thought this is going to be where God is going to take us. And that's not going to happen. And that's going to be where God is going to take us. Has to send the demand signal. They have the most budgetary expanse and resource to be able to afford these. And I'm so pleasantly surprised that I've learned one of our objectives is to drive the cost of the bill of materials down so that it can be commercialized, afforded by other industries across the market segments. Automotive is a great one. Aerospace and defense for computational fluid dynamics. We're seeing a whole host of use cases and industries emerge. This is fascinating. And I agree with Cameron. I tell my team all the time, if we have one of these companies, we're going to see a whole host of use cases and industries emerge. One motto, it's the U.S. has to win this quantum race, right? But for like what Ben was saying, for deterrence, for cybersecurity reasons. But just like Cameron was saying, you will pay, the loser of this race will pay a compounding economic cost. Your innovation will fall behind. Imagine if one side had computers and the other didn't. It'll be that sort of offset, right? So your ships won't be as good. Your pharmaceuticals won't come out as fast. One of the things I think that bears laying out very clearly is when we talk about quantum computers, we're talking about quantum computers. We're talking about quantum computers. We're excited about what they can do. But the ladder of applications is as you go up, it's less and less clear if they're going to provide advantage. But one of the things for sure they will do is crack crypto. For sure, they will help with molecular simulations, which we would use in something like developing new materials, pharmaceuticals, really important stuff, right? And so when we think about winning quantum, there's just so many reasons and axes to make it a national priority. And we're behind it as well. Yeah. So that leads us to our government taking a strong interest, which they have, but also backing that up with budget. Probably some mandates, like we're seeing with other areas of cybersecurity. And we're seeing it in Department of War, where there's a mandate for post-quantum cryptography. We need to see that across government at all levels. We have to defend our networks first. And then we have to work on the rest of the technology stack because there are clear advantages with compute. And if we're going to have quantum computers, then we need to be able to write application programs for them. So we need to make that easy too. We haven't talked a lot about the software stack that goes along with the quantum computers. That's also an area where we have to prepare, like we're preparing kids coming out of college for cyber jobs. Now we're preparing everybody coming from everywhere with AI-related jobs. Being able to program quantum computers is probably the next thing on the list. Well, I hope maybe then any of you could comment on the talent question, because this is one of the things that we saw with AI. Huge, huge disruption, whereas people who are trained in traditional computer science are now concerned that they may be losing jobs as Claude gets better at coding. But people who went through the AI ML machine learning programs now can't get offered enough money by the frontier labs. What has your guys been experiencing as you try to build out the teams in your respective companies with finding people who actually understand what is an inordinately complex academic area? I would offer you as the number one employer, at least in Maryland for quantum, and we hope in the United States in the coming weeks and months, the talent gap is going to be reasonably significant for the next several years. And there's a very good reason for it. There's not a lot of people graduating with PhDs or masters in physics that could even understand the science behind quantum computing. Of that subset of people, there's basically about 100,000 people or less in the United States. And then as you break those demographics down, you get to the teens that have any interest in getting cleared and helping the government mission. So this kind of puts a context of what we're talking about here. The good news about quantum, unlike AI, it is going to be a massive job generator. It is not going to be a job loss thing. It is going to create tens of thousands of jobs and not in the way that AI is done. These data centers are going to be far more efficient. Our quantum computers can literally plug in and draw as much power as your household dryer. So we're talking about a fundamental shift here, now, not to be disparaging of GPUs. We need all of these different types of compute in the data center. What we're saying here is, is we can make them more efficient because we can train models and certain elements of the models with half the power. So you can refocus the GPUs on those extremely critical mission use case scenarios where they offer incredible advantage. But the job creation piece is going to be pretty extraordinary. It's why we partner with all the R1 institutions, all the universities across the United States, we're building academic curriculum. We're going all the way down as far as the K through 12 environments to start really stimulating, but now the steam, because we recognize you do need a little bit of the artistic component, because to your joke earlier about Black Mirror, it's not really a joke. The artistic piece allows our brains to think about the physics in that type of very abstract way. And look, I grew up as a Star Trek and Star Wars kid. We are going to see a day where we will be launching the USS Enterprise out of the state of Texas. That'll happen in our lifetime. And it'll all be a result of quantum because of the material science advancements, the advancements in computational fluid dynamics, engine and thrust, all these different things. They're all a very real possibility now within our grasp. And it's all going to come down to the talent, partnering with the institutions, partnering with the universities and our allied nations, because there's not going to be enough organic talent in the United States to do this. Yep. And I'll comment on talent if that's okay, too. I cannot hire quantum talent fast enough. It is one of those things where we're a huge employee as well in this space. And we have found, though, that when I started building the quantum team at Booz Allen, I was hiring people like myself, other physicists. And you definitely need them. You know, physics has a massive role in quantum. But we found as we've scaled out, you need cyber, you need engineers, you need this huge interdisciplinary team to really bring quantum to the missions we want to bring it to. And I'm sure you guys would agree that that has been the magic is what is the interlock of all the different specialties and trades and what we've done to get quantum effective. And it's been really fun to build it. So that's a brilliant point he's making, J.D. You have a real world example. So we partnered with a classified customer, and they're doing hybrid workflows right now between AI and ML. And what was fascinating is when I went to my engineer, my field engineer who was in physics, had a PhD in physics, when he looked at the problem, he's like, "I don't know if I can actually solve for this." It wasn't until I married him to a classic IT person that has done dev sec ops in the community, is very familiar with Penny Lane and all the other things we do there, along with all the other softwares we use like Kubernetes and everything else in the dev sec ops community. It wasn't until that partnership occurred that we had the spark of innovation that actually solved the problem the customer's been working on for four years. We had a similar scenario doing an order of battle management scenario for a hackathon. And it was extraordinary to J.D.'s point, when you get a physicist and a classical person together looking at the world through two different lenses, that's where all of this is going to accelerate. So it's a multidisciplinary, multiple talented team that you need to pull this off. One of the first decks that we did at GDIT, we began thinking about quantum years ago is, let's work with our learning and development team, the trainers in the human resources area. Let's put curriculum together for every employee in the company who deals with technology in one way or the other. So to understand the words that we're using right now, what is the modality? What is quantum? What are the packages that program quantum computers? How are those different than current packages that we use now? And that's converging and getting a lot closer. What does PQC mean? What's atomic clocks have to do with technology stack? And so what we found is that some of the best quantum people were people that were already in the company that knew how to solid IT foundation, solid cyber, solid AI foundation, pick the stuff easily, and then go work on the projects, the various projects that we have. And we've uncovered some really interesting use cases where we've been able to prove the efficacy of quantum over classical computing. And as things get complex in the machine learning area, especially with complex images, like seeing through the dust and the snow and the rain and the leaves, and trying to understand what's out there and what's moving, those sort of use cases, clear distinction of benefit with quantum computing. Well, this is really unusual, for a national security podcast in as much as you've all given me a great deal of hope, and particularly hope that technology is moving in the right direction and we're doing the right things about talent. So just to balance that out a little bit, let me ask each of you, what do you think some of the holdups, the biggest challenges are that we'll see towards progress? It could be a particular material in the supply chain, or it could be, let's say, a lack of government funding. What concerns you going forward? Being actively engaged in multiple government engagements right now, there's been a few lessons learned. We are maniacally focused on domesticating the supply chain. supply chain for quantum. We think that not only being a good steward of the science to help the country win the race is important, but being a good steward of the entire ecosystem and supply chain has to come through it. It's why to compliment Ben's point, we're not just a compute company, we're an entire platform company. We needed to bring compute, sensing, network and security, and now the fabrication of the semiconductors with the Skywater acquisition. But the raw net precious source and natural minerals are going to be required. You still have to source from allied nations until some of the policies catch up with the technology, which is a traditional thing that happens for government. And that's happening. We needed to happen faster because that can hold up a lot of supply chains. To compliment JD's opening salvo on the modalities, there's still some engineering challenges that have to happen. And there is a talent gap there to be able to engineer. And then we need to start thinking of how do we bring classical compute people in that have engineered these technologies. These platforms using other technologies like CPUs and GPUs and start to see if we can accelerate that. And I think from a collaboration with Five Eyes allies, I think it's going to be extremely important. And we need to lean back into that community because the U.S. although needs to win this, it's going to require, I think, a whole team to go do this. Procurement traditionally has been interesting following the DoD 5000. I know Ben and I know it can be painful on Mondays and very easy on Thursdays, right? You know, with JD's, it's the way it goes. We need to make sure we are paying attention to how fast we can allow the government to acquire this because speed is going to be critical. That once the technology is available, we have to get operational as fast as possible. We have to be judicious and pragmatic and secure, but we have to move fast. I think one of the biggest problems we're seeing is there's just a thousand fires being lit. We work with a ton of Fortune 50 customers in addition to the government. And you have big fans of Quantum and people who have nothing to do with it, right? There is a lot of real good effort out there. But I think a holdup is going to be once it becomes obvious, and I know to us it already is, that this is inevitable. This is the thing you must do. You got to move your cyber. You got to start thinking about compute for your problems, whatever it is, sensors for your device. Everyone's going to rush. And I think a lot of right now, if we want to talk national security stuff, the supply chain is an interesting challenge. A lot of people have a fully vertically integrated supply chain or they make their entire thing or most of it. And while that is great for prototyping, it's not great for manufacturing at scale. And it requires people to play nice. You know, we're both going to go in on a foundry that's going to make these pieces. And right now it's a bit of a horse race trying to figure out who's going to win in what field of Quantum. So I think we've got a couple of issues in pulling together, not only as a nation, but as an allied group to say, we got to win this race. We got to win. So what do we do? And this is commercial. This is federal. This is everyone. But that's a big one. And of course, we already talked talent, but talent is going to be a perennial problem. Because just as soon as we get enough talent for today, the needs of tomorrow are already on the horizon. And do these programs even exist? A lot of people got into Quantum through a more door. Oh, I did math. Oh, I did physics. Not, I've done nothing but design quantum chips since I started college or maybe even before, right? And I think that is, we're going to get quantum native workforce, but yeah, you put the filters on your search and you've got, that's interested in national security, that's willing to live in this area, and you have a very small population. Well, to recap, just to say that we need to defend our networks. That's everything in our systems. That's everything from our phones and laptops that we do our banking on, because those are vulnerable to quantum decryption. We need to build better networks. Those can be quantum networks and quantum encrypted networks that shut themselves down when they're attacked. And we need to get the quantum compute advantage. We need that as the deterrent. So all of those things need to happen quickly and they need to happen for every level of government and every computer user and digital element of our. Our entire economy, not to mention our national security systems. I'll leave you with one point that I think we all remember. If you remember after the pandemic, there was a computer chip shortage. And if you remember the secondary tertiary effect of that, price of cars went up, price of milk went up, price of everything that we had to buy went up because of the lack of simple computer chips to satisfy or satiate what we needed for the digital economy to keep moving along. To enhance your point, without focusing on it now and continuing to say it's a nascent technology, it's not until '27 or '28, I would offer we're already behind. Because that same moment that we experienced with the chip shortage will 100% happen in quantum if you don't start planning now. These machines can only be built so fast. There's only so many people that can build them in the world. And even if you get to manufacturing scale now, there's only so many things that can be produced based on your supply chain and what the supply chain could offer. So that's why I'm grateful for this podcast. I'm grateful for these things, because we have to create awareness that if you're not starting now in the second half of '26, you are already 12 months behind in your quantum aspirations. Yeah, I like that. Best time to plant a tree is 10 years ago. Best time to start building a quantum foundry. So. That's right. I sense we could keep talking all afternoon. And I also sense you guys will be in high demand on podcasts in the coming years. But for now, thank you for joining Cogsworth. Thank you. Thank you. Thank you. Thank you for listening to Cogs of War. Don't forget to sign up for our free newsletter that comes out every other week, warontherocks.com/cogs-of-war. Keep reading, keep thinking, and stay healthy.