How can we deter a future conflict over Taiwan? On the newest Mission Matters Podcast episode (links in comments), Matt and I sit down with Eyck Freymann to discuss the evolving U.S.-China-Taiwan relationship, the challenges of deterring Chinese aggression, and the technologies that could shape a future conflict in the Indo-Pacific.
On the episode, we discuss:
🇹🇼 Why Taiwan matters: The strategic importance of Taiwan goes far beyond semiconductors, and its future has enormous implications for the U.S. alliance system in the Indo-Pacific.
⚔️ Deterrence and gray-zone warfare: Why China may prefer a gradual campaign of economic coercion, cyberattacks, and other sub-threshold activities over a conventional military invasion.
🚀 Emerging defense technologies: The critical roles of autonomous maritime systems, resilient communications, integrated air defense, and AI in a potential conflict over Taiwan—and why the lessons from Ukraine don't always translate to the Pacific.
🏭 The defense industrial base: Why the U.S. and its allies need to rethink defense production and build integrated, allied supply chains to compete with China's manufacturing capacity.
One of my biggest takeaways from the conversation was Eyck's argument that deterring a Chinese invasion of Taiwan is both possible and achievable—but deterring an invasion alone isn't enough. The U.S. also needs a strategy for responding to the growing range of coercive activities that fall below the threshold of conventional warfare.
As always, please reach out if you or anyone you know is building at the intersection of technology and national security. And please fill out the form below if you are interested in working in defense tech (in or out of the Shield Capital portfolio).
You can listen to the podcast on Spotify, YouTube, Apple, the Shield Capital website, or right here on Substack.
Transcript
Maggie 00:45
In this episode of the Mission Matters podcast, we sit down with Eyck Freymann to discuss strategies to preserve peace and protect US interests and values in an era of systemic competition with China. On this podcast, we often discuss the kinds of technologies that we’re building to improve our national security, but we rarely dig into the details of what the adversary actually looks like and why we need these new technologies to remain competitive. So this is an episode I’m really looking forward to. I just read Eyck’s newest book, Defending Taiwan: A Strategy to Prevent War with China, and learned quite a lot about the actual tactics and situation on the ground of what it might actually look like if we had some kind of a conflict with China.
Matt 01:27
In addition to his most recent book, Eyck is the author of several other books, including The Arsenal of Democracy: Technology, Industry, and Deterrence in an Age of Hard Choices, and One Belt, One Road: Chinese Power Meets the World. This conversation today covers everything from the current state of US-China-Taiwan relations, different pillars of deterrence to prevent a conflict with China, and the role that emerging technologies would play in a potential future conflict. Eyck, thanks so much for coming on the podcast.
Eyck 01:56
Thanks for having me on. Let’s get into it.
Maggie 01:58
Yeah, well let’s just start with a basic question. Why does China care about Taiwan, and why should America care about Taiwan?
Eyck 02:07
Well, thanks for starting there. That’s the way to start. The first thing to say is that both China and America cared about Taiwan before Taiwan made semiconductors. In fact, we almost went to nuclear war over it twice, in the first and second Taiwan Straits crises. And the reason has to do first and foremost with Taiwan’s geographic position. Taiwan sits right in the center of the first island chain, which is an archipelago extending down from Japan all the way to Southeast Asia via Taiwan and The Philippines. There are not that many navigable waterways between those islands. That partly has to do with the topology of the seabed. It’s known as the Ring of Fire, right? There’s volcanoes. It’s a place where two tectonic plates collide. And as a result, the United States learned this geography lesson in the Pacific War and the Korean War, if our friends and partners are all astride this first island chain, then a Chinese or any other continental Eurasian navy will have a very hard time breaking out into the wider waters of the Pacific, crossing the ocean and threatening the homeland. But if China can somehow flip Taiwan and use its Pacific-facing coast as a forward base to project naval power, surface ships or submarines forward, they can effectively surround Japan. They can surround the Philippines, they can surround South Korea, and these are economies that ultimately are almost entirely dependent on imported energy, food, and other inputs to survive. So a world in which Taiwan falls by force or coercion is an unmitigated catastrophe for the entire alliance system in the region. In fact, the current status quo in our relations with Japan and the others becomes unsustainable. So that’s why China wants Taiwan for geostrategic reasons. It obviously has this enormous propaganda value for the CCP because not to retread the history in too much detail, this is the unfinished business of China’s civil war. This is the question of whether there is such a thing as one China that includes Taiwan, and whether the Chinese Communist Party is the sole legitimate government, or whether there is some second China, a free China, which is richer and more technologically advanced, and just better in a lot of ways, or whether Taiwan got separated at some point along the way. So for Beijing, this is an incredibly sensitive issue that relates to the Communist Party’s legitimacy, Xi Jinping’s personal legitimacy within the regime, and also this is this massive opportunity, because if Taiwan can be made to submit, the dominoes might very quickly fall, and then China’s pathway to national rejuvenation and eventually regional and global hegemony would be well underway.
Maggie 05:15
One of the things I find really interesting is actually learning about the domestic situation in Taiwan. So can you tell us a little bit about some of the different political opinions in Taiwan about China and its relationship with China?
Eyck 05:29
Well, how much time do we have? I’m kidding. I’m kidding. So, it’s hard to answer this question without drawing caricatures because Taiwan is a thriving democracy with an incredibly robust and vibrant political debate. And they understand China reasonably well because they can absorb CCP material in the original language because they’re flooded with propaganda, from every direction, because they’ve been living under threat in one form or another for the last 70-odd years, and because they’re not stupid. Like everyone else, they can see that Donald Trump is unpredictable, and they wonder what that means for their security and their personal future. So, Taiwan is divided between two basic camps that they call green and blue. The green camp is the Democratic Progressive Party, it’s the current ruling party, which basically takes the view Taiwan is an independent country. We became independent in 1949, and we’re not going to rock the boat, but we are in fact an independent country with our own distinct traditions, heritage, history, culture, and we have a place in the US-led security architecture, and the United States wants to protect us for the reasons we’ve discussed. And therefore, if we keep the Americans on board and we play our part and we are a good partner, the Americans will deter this for us, and they will not abandon us. And then there’s the Blue camp, which is the Guomindang. This is the party of Chiang Kai-shek, the party that fought the Chinese communists on the mainland in the 1930s and ‘40s and then decamped to Taiwan after, which sometimes gets branded as the pro-China party. That’s kind of an oversimplification. The KMT would say, “Who are we kidding?” We speak Chinese. We eat Chinese food. Most of us practice Chinese religion or Chinese philosophy in some form. Most of us have ancestry on the mainland, an ancestral hometown on the mainland. We are Chinese, right? There is one China that includes Taiwan, but for historical reasons that we all understand, there is a free part of China and an unfree part of China, and the free and unfree parts have different definitions of what the ruling framework for China ought to be. But on that basis, we can talk to the CCP. We can deliver each other’s mail. We can trade, and furthermore, the Americans may not be reliable. The Americans may not be coming. The cavalry may not be there, and if that’s the case, we don’t wanna be used as a pawn in someone else’s game, and we don’t want Beijing to get the wrong idea that we’re going to provoke a crisis. We need to do diplomacy with Beijing to reassure them that the best way to handle this is through grown-ups sitting down at a table. And now I’m caricaturing, and within each of these parties, there’s a huge range of perspectives which range from the loco-coco, independence now, dare Beijing to attack us, which is a rounding error of the population, less than ten percent, not zero, loud on Twitter, but less than ten percent. These people have blown up my DMs since I’ve written the book. And then on the other side, the loco-coco or paid off united front, unification now, we can become a glorious province of the PRC and advance Chinese communism. Basically, no one in Taiwan believes that. But plenty of people in Taiwan believe, due to China’s size and scale and geographic proximity, that some accommodation has to be had. So these two camps totally hate each other. They inhabit completely different information environments. It divides families. It divides communities. And unlike in the United States where there’s this, you could call it a deep state, or you could just call it a centrist core of technocratic professionals, which is most of our military, most of our intelligence community, a lot of our diplomatic corps, people who basically consider the partisan stuff to be noise, a distraction from the mission, Taiwan is a younger democracy, is very polarized across all of its institutions. They don’t really have nonpartisan think tanks, for example. And both sides caricature the other side and try to get the Americans, in some cases China, to intervene in Taiwan’s politics to their advantage. This is a very noisy and challenging information environment, and the reason I’m harping on this is because Taiwan’s going into an election. In 2028, January 2028, they will elect their next president, and it will probably be the incumbent, Lai Ching-te of the DPP, the green side, versus some yet to be determined KMT. And there will be a temptation for some in the United States government to just assume, well, the DPP is the pro-America party superficially, so they’re our team. And one of the arguments in the book is this stuff is way more delicate than we would think, and trying to meddle in Taiwan’s politics is more likely to backfire than to work. We should respect that the people of Taiwan can make their own choices, that they are sane and don’t want their kids to grow up under communism, and that they will make good decisions for themselves. But you have to remember that the people of Taiwan will respond to the signals they get from the United States, and if they believe that the United States does not have their back and will not be there for them, then they may decide that the only solution is some kind of negotiated surrender. And the goal of the US policy should be to back up Taiwan, whoever they happen to democratically elect, to allow this process to play out peacefully through diplomacy and not in a more violent or coercive way.
Matt 11:43
So speaking of the US position on Taiwan, help us understand, like let’s suppose the US decided that the risks of conflict with China weren’t worth it, and that, whether it was the military conflict or just an economic crisis that coming to Taiwan’s defense, in an invasion or blockade scenario was just not worth the cost. Help us understand what would be the worst case scenario for the type of outcome that could lead to, from a US perspective. What would you want the US government and citizenry to have in mind, as it evaluates that decision and thinks about what its posture will be toward Taiwan?
Eyck 12:31
Assuming we’re unwilling to fight?
Matt 12:33
Yes.
Eyck 12:33
Well, if you’re unwilling to fight, you’re choosing between bad options, because you definitely don’t wanna get backed into fighting a war that you don’t wanna fight. So if it turns into a crisis and you’re intending to fold, the more chips you have on the table when you eventually fold, the worse a position you’re in. And in other words, if you’re holding a bad hand in poker, fold early. Everyone knows that. On the other hand, the United States has maintained a, what we call this One China policy. Technically, it’s a One China policy guided by the Three Communiques, the Six Assurances, and Taiwan Relations Act. It’s a whole litany to recite, but we have this policy that has grown over half a century that we have reiterated a zillion times, and that is part of what creates stability in the region. So if we were to change our policy in an arbitrary way, it would scare the daylights out of all of our regional partners. It would make them doubt whether our promises to them, which have never really been tested, are still good, and it may invite provocations and further tests, not just from China against Taiwan, but from North Korea on the peninsula or in contested maritime space, from Russians all around the region. The Russians are out there on a regular basis harassing our Japanese friends. China may see whether it extends into the Southern Ocean. They recently fired a nuclear-capable ICBM towards Australia. So the possibility that walking away from Taiwan could actually set in motion a catastrophe that brings us in is something we can’t ignore, and that’s one of the reasons for strategic ambiguity because, frankly, different presidents have different views on this, and we don’t have a window into the heads of past presidents, and it’s possible that some of them would have fought for Taiwan and others actually wouldn’t. But maintaining the current ambiguous position seems like the wise thing to do in that scenario. But obviously you don’t want to be led to the brink and then have to walk down in a humiliating way.
Matt 15:06
Yeah, I mean, in the book, you draw a really interesting historical comparison to talk about US resolve, during the Cold War, to Berlin, and the symbolic positioning that Berlin had in the Cold War, and how it’s analogous in some sense to the position of Taiwan today. I was wondering if you could explain that comparison and what we can learn from it as the US kind of continues or modifies its strategy and adapts to the new situations going forward.
Eyck 15:34
So there’s lots of ways in which Taiwan is not like Berlin, that we have to get to understand the limits of the analogy, but let’s talk about the ways in which they are potentially instructive. So Berlin is this weird case at the end of the Second World War because it’s in the heart of East Germany. It’s an exclave. It’s in the heart of East Germany, but it’s divided into four parts, a Soviet sector, an American, a British, and a French. And over the course of 1945 to 1948, it is unclear whether the Soviets will do as they’ve promised at Yalta and actually allow these Eastern European countries that were liberated from the Nazis by the Red Army to elect their own leaders. And it becomes clearer and clearer that Stalin is not intending to do this, and the Western Allies decide to pursue increasing integration for their sectors in Berlin and turn it into more of a garrison. But it’s very hard to defend because it’s a pinprick on a map completely surrounded by Brandenburg and East Germany, which is controlled by the Red Army. And the Soviets believe it should be trivial to take it through blockade. In 1948, they try. And the Truman administration, with help from the British and others, decides to launch this airlift to prove our commitment. And various other contingencies are discussed that would involve the use of nuclear weapons, but the United States decides we’re not gonna go there yet, but we’ll threaten the Soviets that if you shoot the planes which are resupplying Berlin, then you’ll likely have nuclear war. And in 1948, ‘49, only we have nukes. So there’s lots of reasons why we would not wanna fight the Soviets, but we do happen to have nukes and they don’t. And we supplied Berlin, and it is an epic public relations victory as well as an operational success. Because not only do we supply Berlin through the winter with just a never-ending flood of planes landing, like, on the tarmac in Tempelhof, like a Swiss watch, but crowds of German schoolchildren, you know, stand and wave flags for them. And it’s clear that we are the good guys and the Soviets are the bad guys, and that we’re gonna continue this, and they give up. Now, it’s tempting to think, “Oh, awesome. So if China blockaded Taiwan, we would just do that.” Wrong. And the reason is the distances are much greater, there’s 10 times as many people, and the GDP is much bigger, and a lot of what they do is, like, industry, like making chips and things that the world needs, which is super energy intensive, and a lot of Taiwan’s energy system runs on gas, which you can’t fly in a plane. So for various reasons, it’s much, much harder to resupply Taiwan through an airlift than it was for Berlin. The other way in which the Berlin crisis is revealing is in the late ‘50s through 1962, under Eisenhower and Kennedy. At this point, the Soviets have nukes, and they decide to try again. And Khrushchev’s idea is, “Well, now that I have nukes, you wouldn’t dare.” And he ends up linking Berlin and Cuba, and he pushes us to the brink of a Cuban missile crisis, believing that if he stares us down over Cuba, he can also make a deal over Berlin. And that doesn’t work in part because he underestimates our willingness to go to the nuclear brink. There’s a bunch of lessons you can take from this history, but one of the big ones is Taiwan feels a bit like Berlin in this respect, and if you’re not prepared to go to the brink, you’re lost. But it is much more likely, given the fact that the US and China are nuclear powers, that Taiwan’s fate is determined by a crisis than by a war. Because if we do go to the brink, given the unimaginable stakes, it is quite possible that one side or the other just backs down with all of the reputational disaster that follows from that.
Maggie 19:52
So coming back to the present, luckily today we are not on the brink, but that’s not to say that there isn’t anything happening in—
Matt 20:00
We’re getting closer.
Maggie 20:00
In conflict. Yeah, so can you give us what is the state today of conflict between Taiwan, China, the US, and, you know, in particular, what’s happening in the gray zone kind of beneath the threshold of a hot war?
Eyck 20:17
There’s a few different levels to this. Let’s talk at the conventional military level, which is where most of your listeners are probably paying the most attention, which is this question of deterrence by denial. China’s building this capability to try to show that they could take Taiwan through an amphibious invasion. Now, an amphibious invasion will probably have multiple parts and multiple sequences. It would probably begin with cyberattacks and bombardment to soften the target, to trap some of Taiwan’s air force in their mountain hangars, maybe to disable some of their surface fleet, some of their air force. It would try to degrade Taiwan’s command and control infrastructure, probably try to take out some of their leadership, sow chaos, all of that. But then ultimately, if China wants to take Taiwan through an invasion, they have to deliver some number of tens of thousands of men across 100 miles of very challenging water, either by ships on the surface or by air. And this is very challenging to do because 10, 20, 30,000 men have to be resupplied with a whole lot of stuff, like millions of tons of stuff: vehicles, fuel, spare parts for the vehicles, medical, tents, other kinds of equipment, blood. And these things have to be, seriously, these things have to be delivered on schedule in what may turn out to be dozens or hundreds of crossings back and forth across the strait. And there’s only so many of these vessels that can do that job, and if a certain number of them are taken out of commission, damaged, destroyed, sunk, then you can have a scenario where China ends up with several thousand men on the island, but no ability to sustain them. They get massacred, and this is the worst possible outcome reputationally for China. So, the contest that the United States and its partners on one side and China on the other are engaged in on the deterrence by denial question is trying to change each other’s risk calculus or assessment of how likely it is that this could succeed. And so much of this depends on assumptions, on assumptions, on assumptions stacked on top of each other. You know, how much operational surprise do they get? How resilient is Taiwan’s communication system when certain nodes get whacked? What countries allow the US military access basing and overflight? And on and on and on. But both sides are presumably studying this with an incredible degree of granular detail, and they’re plugging it into simulations, and they’re running those simulations. They’re all probabilistic, and they’re running the simulations zillions of times and then seeing what percentage of the time it comes out with a win and with a loss. Obviously a real war, way more complex. These are not crystal balls, but just they’re indicative. And what the US and its partners are trying to do is multiply the sources of catastrophic risk that this kind of operation would face. And it’s even if Beijing has an assessment, it is something-something percent, they’re gonna have confidence intervals on that number. So it’s partly about shaping what that number is and partly about shaping their confidence intervals based on classified capabilities that you may or may not have. But that is obviously not the whole story. There is a whole other basket of activities that China’s doing to Taiwan short of kinetic attack that they hope will get Taiwan to submit without fighting. And I think obviously China would prefer to get Taiwan for free than to go through all of this cost and challenge and risk in an operation that could not possibly have 100% chance of success.
Matt 24:15
Where are the conflicts in Ukraine and the Middle East instructive for how to think about deterrence in Taiwan, in the Taiwan Strait, and where are they not instructive?
Eyck 24:26
Oh, that’s an interesting question. So I think the danger of over-indexing on the Ukraine case is that we assume that Xi Jinping’s personality is like Vladimir Putin’s. It’s not. It’s totally different, and his political system works differently, and that it’s a binary between whether he invades or whether he doesn’t. Let’s talk about personalities. Vladimir Putin is a risk-taker. He’s a gambler, and he’s invaded countries before. He invaded Georgia in 2008, and he got away with it. We let him get away with it. And then he took Crimea and started what was essentially a low-level war in Eastern Ukraine in 2014 that dragged on for years and years. And we also basically let him get away with it. We put sanctions on, but we did not resist what he was doing from a military point of view. So he did learn from those two experiences that he would probably not face serious military costs directly from NATO if he moved again against Ukraine. And he also revealed that that’s part of his pattern of behavior, so it should not have come as that much of a surprise when he went for the whole thing in 2022. Xi Jinping has a different personality. His whole modus operandi is different. He is more of a risk-taker when it comes to taking out threats to his rule at home, taking out rivals. But that’s because the stakes if he gets that wrong are existential for him. When it comes to moving in foreign policy, he tends to act in a much more gradual and systematic way, trying to change facts on the ground over time. Look at how he handled the South China Sea. Look at how he handled Hong Kong. In fact, he probably believes that he was extremely restrained letting the Hong Kong protests develop as they did over more than five years before he finally decided to put the kibosh on it in 2019. So everything we know about Xi Jinping’s mode of operating suggests that he would prefer a slowly tightening noose than some blitzkrieg over Taiwan. There’s also a difference between land war and maritime war, which is pretty significant. And maybe this is a separate question, but clearly drones matter a lot. Unmanned systems are changing the battlefield in Ukraine, and there will be significant read-throughs to the Indo-Pacific. But it’s just different when you’re operating across an enormous body of water, and the battle space is hundreds of times as big, and you’re routinely using either carrier-based aviation or very long-range precision missiles to deliver mass to target. Command and control, your ability to see and understand a battle space that is hundreds of thousands of square miles of size, is an entirely different technical problem than the one Ukraine and Russia are facing in their land war today. Relatively speaking, C4ISR matters much more, and the lesson from historical maritime war seems to be because of that. If you can see your enemy and deliver an effective first shot, that can be it. So maritime wars, once they start, tend to have much more decisive conclusions that come very quickly. And you can tell a story either way about what that means. But it does suggest that a lot of the things that we believe we have to build in Ukraine to protect the rest of Eastern Europe may not be exactly the things we have to build in the Pacific.
Maggie 28:34
Yeah, I think you were reading my mind for the next question, which is one of the major themes, of course, coming out of the conflict in Ukraine and the Middle East is the importance of commercial space like Starlink, the importance of drones, the importance of AI for targeting, the importance of electronic warfare, et cetera. What role will some of these emerging technologies play in a future conflict, potentially in the Indo-Pacific between China, Taiwan, and other allies?
Eyck 29:03
Well, immensely important role. Let’s start with the invasion scenario. So Taiwan has to... Let’s assume that Taiwan is being invaded, they’re being bombarded. They have to sustain the resilience of their command and control, so they actually know what is happening and they can coordinate their forces, and then they have to identify which are the most dangerous approaching craft, whether in the air or on the surface, and then they have to take them out with an economy of resources because Taiwan during an invasion is gonna be very hard to resupply. Ukraine has depth, they have geographic depth, and they have a land border with NATO, and Taiwan is something like one seventeenth the size. So they will have to conserve their forces and use their missiles, unmanned systems, and other fires in a much more disciplined way. So that means in a context where Taiwan doesn’t have access to Starlink, the technologies that will enable Taiwan’s armed forces to have truly resilient C4ISR are fundamental. There’s a number of reasons why Taiwan doesn’t have Starlink. One is that they don’t trust Elon, and they probably have good reasons not to trust him, given Elon’s recent comments in China. But the other thing is, unmanned systems where unmanned surface vessels are probably gonna matter relatively more. There is a maritime fight in the Black Sea. The Ukrainians have been very effective at it. But if you assume that China will generally, in invasion, try to take most of the men across the strait on slow-moving ferries, the question is how you can create risks and complications for those ferries at a relatively low cost. And an indigenized supply chain to make USVs of various sizes is, I think, 100% essential, mission-critical for Taiwan. They’re doing that. They’re ramping, and if Taiwan can do it right, they could potentially become a provider of those capabilities to others in the region. And I could go on all day about this. I think these are two of the big ones. The third is air defense. Taiwan is close enough to China’s mainland bases that they would be saturated with missile and drone strikes, just enormous strike packages against their critical infrastructure, against military targets, possibly against civilian targets. And force conservation is gonna be an incredibly important principle in absorbing that, those opening barrages. Taiwan cannot afford to expend its precious air defenses on cheap, low-value drones. And having an integrated air and missile defense system is non-negotiable for them. Probably exploring some creative counter-UAS capabilities as well. The good news is that Japan needs exactly the same thing, South Korea needs exactly the same thing, the Philippines needs exactly the same thing. So we have a basis, if we can get our act together, to create one great big allied market and create an overwhelmingly powerful demand signal for whatever producer anywhere in the world can generate these capabilities. And I’d like to see us move in that direction.
Matt 33:05
Yeah, are we moving in that direction, or what are the hindrances to having a allied demand signal for these capabilities across the Indo-Pacific?
Eyck 33:18
Well, I think we’re not yet at a point in the American debate where we’ve accepted that we’re just not gonna make most of this stuff ourselves. I mean, it seems obvious to me, and I argued in my previous book, The Arsenal of Democracy, that literally the only way that we can get the capabilities we would need to fight and win a war with China is gonna involve buying at least components, if not finished stuff, from Japan, Taiwan, and South Korea, ‘cause we can’t make anything in this country. And if we make anything, it comes in late and over budget and often poor quality. And with no offense to the many exciting defense techs that are popping up everywhere, including at Stanford, I mean, there was a wonderful Wall Street Journal piece the other day about, I think it was in the Journal, about trying to make the board game Monopoly in the United States, and the conclusion was it cost twice as much, the pieces were crap, and they had to buy the dice from China. So if you can’t make a Monopoly set, why would we think that we can become a globally competitive drone power? We’re just not going to be. But if that is true, then the implications for our strategic posture are enormous because it means you are gonna become dependent on someone else for a capability that is life and death in fighting any future war. So which of our partners are we willing to become that dependent on? And our partners are gonna ask the same question. So I think there’s an overwhelmingly strong case for involving the Japanese. But do we want the Taiwanese to play a role in this as well? We already depend on them for TSMC. We haven’t spent much of this conversation talking about chips, but, like, clearly one of the reasons we care about Taiwan is that they make all the GPUs that are powering AI. So if we’re depending on them for the GPUs, why not also depend on them for the motors or the actuators or the sensors or the other ingredients of making a drone? It’s one thing to say that on a podcast. It’s another thing for the Pentagon to put in writing, “Yeah, we’re gonna become strategically dependent on Taiwan to make drones in the United States.” But the co-production thing is the only way to do it, and this is a conversation that has to include at minimum the Indo-Pacific allies plus Taiwan, but ideally also the British, whichever European militaries are interested in being involved, because they’re struggling to make drones too, the Ukrainians, who have their own supply chain challenges, who would struggle to do bilateral cooperation with Taiwan but are totally happy to access our capital markets, and frankly the Canadians too, because if the club is that big, maybe the Canadians will come to the table. So the approach should be you get all of the allies together and you say, “We want to become an integrated block to make drones so we don’t depend on China for the drone supply chain. Who are we comfortable including in the club? Do we not wanna become more than 70, 80% dependent on any one place? Okay, fine.” What follows from that in terms of setting specifications to make parts interoperable across different kinds of unmanned systems? Who’s gonna be allowed to buy from whom? How are the co-production supply chains going to work? It just seems obvious to do the economy of scale, but actually accomplishing that is a diplomatic lift, which is easier said than done. And that is why it hasn’t happened yet.
Maggie 36:59
Can you tell us what is the current state of the Chinese military’s investment in some of these emerging commercial technologies, and how do we, you know, the US and Taiwan stack up?
Eyck 37:10
Well, that’s a broad question. So China is investing in a whole array of technologies. Now, how good those actually are is another question. Now, I like to say the CCP loves things that it can count, and it hates things that it can’t count. And the way that their system works is they’ll have a meeting and they’ll decide we’re gonna make X many missiles, Y many drones, Z many submarines, and then they go out and do that. In the same way that every year they have a meeting and they say our GDP is going to be 5%, and if their GDP is actually only on track to be 2%, they dig holes and fill them up again or build bridges to nowhere. That’s how their system works. These numbers are inputs, they’re not outputs. So there is this lingering question about quality that applies to everything that they do. And quality is harder to measure, and this is a source of insecurity for them.
Maggie 38:16
I mean, we hear stories, right, of missile silos filled with water instead of gasoline, right?
Eyck 38:21
Well, if you wanna think about, like, what levers we can pull to make Xi Jinping think twice about the assuredness of his military forces. Like, you’re not gonna make him doubt that he has X many drones, Y many missiles, Z many submarines, and you’re not going to make X plus one, Y plus one, Z plus one because they’re the Chinese Communist Party. So if you can’t beat them in quantity, you must beat them in quality, and you must inject uncertainty into his thinking about the extent of that qualitative gap, that you must show him that you lead technologically, and you must make him insecure about what that lead means, and so he doesn’t know how much quantitative overmatch he needs to compensate. That is, by the way, to put it in a different way, what we did in the last 15 years of the Cold War that beat the Soviet Union, is that starting in the second half of the ‘70s, heading into the ‘80s with a bunch of visionaries, Harold Brown and others, we decided that we would bet on going to space for a lot of our ISR, and this was not a sure bet at the time. It was a moonshot. GPS was almost canceled multiple times. And this is a whole saga which I talk about in a previous book. But the fact that we did it, and then we displayed those capabilities to the Soviets by doing deterrence patrols that shocked them, that we could just go dark and pop up someplace else, made them unsure about what it would actually take to catch up and conclude the only possible approach was arms control to rein in our capabilities. We need to do something similar for China. And arguably, if you want to get China to the table in AI arms control too, we have to do something like that. But in cyber, in other domains where quality is hard to measure, it’s very important to signal our lead when we have it. And we need to invest in the sources of our technological strength because they are marshaling a lot of resources to catch up.
Matt 40:41
One major stakeholder in all this we haven’t talked about much is NATO and Europe. Obviously that’s a frequent topic of discussion and debate today. What role do you see NATO playing, especially now in deterrence in the Indo-Pacific, not just militarily, but kind of across the different pillars you talk about in the book, and especially economically? And what’s your outlook on their role in all this?
Eyck 41:11
NATO remains fundamental because there is one continental scale land war that is happening in the world right now, and it’s on NATO’s eastern border. And if we don’t win it or settle it on favorable terms, it will spread. There is no doubt in my mind about that. Putin will not stop. Therefore, NATO needs to remain active. NATO members need to pay their way. More important than spending the money, they have to spend the money on the right stuff. They have to move towards more efficient co-production of the critical stuff that’s needed. They have to have a better pipeline to get the learnings from Ukraine and integrate them into European militaries. I mean, who among us really believes that if, God forbid, Putin were to roll tanks across the border, that NATO members would fight as effectively with drone warfare as the Ukrainians? I’m not convinced. Who among us believes that if North Korea were to try this, try out some of these tactics, techniques, and procedures on the South Koreans, that the South Koreans would be prepared to fight as effectively as the Ukrainians? I don’t know. So NATO has a role to play in bringing these learnings that Ukraine is acquiring and paying for in blood, to the rest of the coalition. I think NATO also has this globally stabilizing role, even if NATO allies would not be directly involved in the conflict in the Pacific, because China doesn’t want to initiate a chain of events that breaks their relation with Europe. But for reasons of geography, for reasons of politics, for reasons of economy, the European Union is not really much of a player in the Taiwan issue, and they are not part of the core coalition that would have to go toe-to-toe with China if it came down to it in the Pacific. So we should not delude ourselves, and we should also not pretend that you can just copy and paste the NATO model to the Pacific. We should be wiring together our Asian allies more effectively. I hope that’s come through in the conversation. I really think we should, and I think we can, but the way to do it is not, like, to build a cathedral and then try to wire it. You build the wiring first, and then if you want to later, you put some masonry and stained glass windows in. And so the interesting thing to me is what the US, Japan, and Australia, for example, are doing on a trilateral basis to conduct joint fires, and we’re demonstrating that in exercises like Balikatan. Bringing the UK in through the AUKUS pact, which is not just about subs, but it’s about hypersonics and all these emerging technologies. How do we fold Japan in or get Japan to a place where they can join that pact? And then because this is not an all or nothing proposition, how do we wire our respective systems so we can collaborate with the Indians where convenient, even if it’s just for ISR? How do we wire ourselves up to talk to the Vietnamese? Obviously, we will not integrate with Taiwan’s military the way we do with Australia’s, but how far can we go? Rethinking this as a sliding scale. A sliding scale or as a ladder that you can build, climb step by step, I think that kind of federated architecture is much more attractive to everyone, including us, than a NATO model, which has benefits but is also kinda cumbersome, right? Because NATO has to move by consensus. So NATO is wonderful. We gotta protect it, we have to preserve it, and in the Pacific we need to build something a little different.
Matt 45:11
Yeah. One other topic I wanna make sure we went into a little bit more depth about was AI. I mean, you have an entire section of your book where you talk about the role of AI in strategic deterrence specifically. And obviously the US and China both depend on Taiwan for the most advanced chips. I think most of our listeners know about that at this point, so we don’t need to talk about that in a ton of depth specifically. And I think you wrote that 99% of NVIDIA’s most advanced GPUs are made in Taiwan. But this particular issue strikes me as one of the most confusing parts, one of the most confusing dynamics of the Taiwan scenario, just given that the government’s understanding of the role of AI is changing really fast and the models themselves are continuing to improve, like, super rapidly. So kind of across the board, and it’s a broad question, but if you were briefing US policymakers, what framework would you use to explain what role AI is gonna play in all this, especially as it’s just evolving so fast?
Eyck 46:13
Okay. That’s a nice way to end. I wanna make three points. First, I wanna talk about strategic deterrence and how it’s become fuzzy in a very scary way. Next, I wanna talk about hybrid war and the gray zone. And then third, I wanna use those two things to talk about AI. The strategic deterrence used to be simple. If I have nukes and you have nukes, I can nuke you, bad for you. But if you can absorb that and nuke me back, it’s bad for me. And if both of us have a secure second strike, it’s in neither of our interests to move first. So I just explained that in 15 seconds. Tom Schelling won a Nobel Prize for explaining that and other things, and building a game theoretic architecture around it, and that was nuclear strategy basically for most of the Cold War. Now, that doesn’t end nuclear competition. We’re, you know, competing with the Soviets over intermediate range missiles and air defense, ballistic missile defense all through the end of the Cold War. But basically, in a land where strategic deterrence is nuclear deterrence and second strike capabilities exist, it’s simple. We don’t live in that world anymore because our nuclear command and control systems are now multimodal. They’re based on space-based architectures that can be attacked through all kinds of means, computer systems which can be attacked using all kinds of means, and the number of tools that you can potentially use to go after an adversary’s critical infrastructure, including chem, bio, and other things, are becoming easier to access. And as cyber enables all sides, including non-state actors, to reach into others’ critical infrastructure and tinker, it raises real questions of what strategic deterrence even means in the 21st century. And that’s a very scary question, and it brings us back to some debates in the Cold War about what tactical nuclear weapons might do to strategic stability, like whether it might be tempting to use a tactical nuclear weapon that could take out just a building or a city block. Maybe it won’t be such a taboo if it can be targeted in that way. Second point, hybrid war. We are not at war with China and Russia, but we’re not at peace with them either. Hybrid war is relentless. It’s happening globally. It basically is a euphemism for acts of terrorism, and it’s happening all over the place. China cutting undersea cables in the Baltic, cutting undersea cables connecting to Taiwan, using lasers to jam satellites. Russia sending explosive packages to major German cargo airports. I mean, I could go on. This is ramping up in a number of domains simultaneously, and it’s really scary because what it is really targeting are the sinews of commerce in the commons, particularly the sea lanes through which the energy and people and supplies on which our economy depends for its lifeblood depend. And we see an example of this in Hormuz. Hormuz isn’t fully open, and it isn’t fully closed. And what do you do with that if you’re no longer in a world of binary open and closed? This is how China plans to move against Taiwan, through an all-around strangulation campaign that targets the shipping and eventually aims not to shut it down but to take it over, to dictate the terms on which it can flow. We do not yet have a doctrine for how we deal with that, and the proof that we don’t have a doctrine is that the bad guys’ use of it is going parabolic. And if it continues to go parabolic, eventually basic functions of the global economy will stop working. So this is not a peripheral question. This matters just as much as deterring an invasion. Now, how does AI play in? AI just turbocharges that problem because it offers all of these ways to make trouble for other countries in ways that are actually really damaging or socially destabilizing, but hard to attribute and below the level of kinetic, actually kinetic attack. You can potentially deliver strategic effects with AI. An agent that goes rogue, “Oh, so sorry it escaped from a lab and took out your power grid. Sorry about that.” I mean, it introduces so many hundreds of scary scenarios at the same time, and we should think of it as an arms control problem from that point of view. But I think we also have to be realistic that the Chinese and the Russians think they’re winning in this fight and won’t change their minds unless they think we have big-time qualitative overmatch. So I think AI is very scary for this reason, because it makes this definition of strategic deterrence fuzzier because it intensifies the hybrid war. I don’t see a clear solution other than pacing the frontier.
Maggie 51:54
Eyck, one last question to close us out. What is the thing that you most see startups and other American technology firms and policymakers most misunderstand or get wrong about the situation with Taiwan?
Eyck 52:07
That we can deter Xi Jinping, that this is not a lost cause. How do I know? Because we’ve deterred him so far. The guy’s been in power for, like, 14 years. Every single day he has woken up and said, “Not today.” So if the question is, can you deter him from pressing the button? This is the big one. When the missiles fly and we’re gonna whack American bases, and we’re going to blockade the island and send the global economy into free fall, I don’t think he’s gonna press that button. Not because he wouldn’t under the right circumstances, but because the United States and Taiwan and our other friends and partners are stepping up and continue to step up so that he doesn’t see a clear pathway to success. So if that’s your definition of victory, I think we are succeeding and we can continue to succeed. What people don’t fully understand is that that is not the whole story. This hybrid gray zone question of the boundary between peace and war getting blurred and pushed, we lack a doctrine for this, and we lack an understanding of the kinds of effects we can deliver in response that will be equal and opposite and proportionate, that will make our adversaries think twice. Some of these may not be kinetic military in nature. Some of them may be economic tools. A lot of them may be cyber-related. Some of the most valuable in China’s case may be information war capabilities. But until we have a doctrine and then a set of capabilities that we can deploy in service of that doctrine, this feeling that the international architecture that holds the commons up is breaking down, that’s gonna accelerate. Our supply chains are gonna get less secure. Our societies are gonna get less resilient. And this is ultimately something that the Pentagon has to lead on by clarifying that this is a priority. But if and when they do, and I hope they do, we are gonna need industry to step up, with ideas, with innovations, with products that can be made available to the allies to respond.
Maggie 54:33
Well, thank you so much for coming on the show. I think our listeners probably learned a lot about the current state of Taiwan deterrence, and all things China. So thank you so much for your time.
Eyck 54:43
Thanks for having me on. This was great.
