David Ulevitch, Katherine Boyle, Ben & Marc
Why Silicon Valley turned against defense (and how we're fixing it)
You can watch this conversation on YouTube here.
Erik Torenberg: Well, now that we have the American Dynamism crew, I thought we’d start with a bit of an opener. I want to ask Ben and Marc to tell us about the moment when we got conviction that this was worth creating its own category fund around. Ben, do you wanna take the first step?
Ben Horowitz: Yeah. Well, I would say that Marc and I were of the view in an abstract sense that we had to integrate American business and American defense back together because we were falling behind. We knew the companies that we were investing in and we knew the companies that we were buying missiles and aircraft carriers from, and they weren’t of the same caliber in terms of where they were on the technology curve.
And so we knew it was a national security problem for the country, but we didn’t have a plan. And then as I recall, David said there’s this woman, Katherine Boyle, who I wanna recruit, and she has this very, very good idea. Can you guys talk to her about it? And that’s when she said look, we should call it American Dynamism when we should build this thing. And it was a very easy sell for us, I would just say.
Silicon Valley’s Historical Relationship with Defense
Marc Andreessen: Yeah, and maybe I could just take a step back. Some of you on this call are too young to remember this, but there was an earlier era in Silicon Valley, and it actually wasn’t that long ago.
Ben and I were both lucky enough, and maybe of the right generation to be part of it, which was Silicon Valley and the United States. Say the mission of the United States as a country, and then sort of the submissions of not just economic and technological success, but also military success, and intelligence success.
Our ability to propagate our way of life, not just in the country, but better around the world. There had actually been a fairly tight degree of integration of cooperation, and if you actually trace the history of Silicon Valley back before even World War II, there was a lot of the original defense technology work that happened in Northern California where, where we sit today. For sure through the fifties through to the nineties, it was just as a given that this was sort of inherently an alliance with a shared American mission. Many great American tech companies including all the big names, from the fifties through the nineties all had very big government businesses, defense businesses, were very important to the national mission. And then, Ben and I in the nineties, when we built our company, Netscape, that was always front and center for us.
My first sales call to the Pentagon was in 1994 when I don’t think I actually owned a suit yet, so 31 years ago. Our company Netscape ended up being a major supplier of technology.
Ben: Both Marc and I had for years, top secret clearance, because of those deals.
Marc: Yeah. So we’ve been working in the Pentagon, and in related agencies, and on the civilian side also for a very long time. And we just considered it kind of natural. I’m really proud of the work that we did there, and it turned out to be a very big part of our business also.
And then it extended over time to also governments outside the US that were American allies. So for us, this is always very natural, and I think even everything I’m saying is almost non-controversial up until probably the 2000s. Then basically for the last 15 or 20 years, for a variety of reasons, including reasons that people feel very strongly about—we could talk about for a long time—just a much greater degree of contention and escalating in some cases to just outright hostility kind of developed, in both directions from Silicon Valley.
So let’s say large components of Silicon Valley, not all, but significant companies for sure, you know, hostility towards DC and hostility towards any sense of national mission, national purpose. Quite frankly that feeling was quickly reciprocated by Washington back to Silicon Valley, and there’s plenty of people from Washington in the last 20 years who’ve made a point of how much they hate us. It just felt to me like we entered this de-evolutionary spiral over the course of the last 15 or 20 years. It’s probably been sort of the peak hostility disconnection was maybe in the late 2010s, early 2020s.
The Google Maven Watershed Moment
Ben: Yeah. I think the Google Maven project was kind of a watershed moment where people had to take sides, just kind of intellectually. When Google won a government contract for AI to be used in drone technology or something like that, and then they basically pulled outta the contract due to a protest from a bunch of their employees.
Marc: Yeah, and Ben, I think you confirmed not just a mild protest, but visceral, emotional.
Ben: Oh, I remember Sundar described it to me, it was like a revolt from the international AI community. So it was pretty wild, and certainly unprecedented.
Marc: I think that sort of thing sort of bottomed out there. Obviously there are new companies that have played a big role here, but some of us like Ben and I, who were involved in this in the old days, it was a chance for us to take stock.
Where things were, how things had gotten that bad, what went wrong, and then start to at least try to figure out kind of how to fix it. That’s why I was so enthusiastic when David and then Katherine, developed these ideas and this new mission for the firm.
I thought it was just like an absolutely fantastic idea at the start and still do, but I just wanna go through that, which is there’s a back to the future component of it, right? Which is what we’re doing and what others are doing in this space.
It is new and bold and exciting and different, and very different than the norm five years ago, and that’s all real. But at the same time, there is a historical resonance, which is also a return to the original values of Silicon Valley, which is something I find to be just incredibly positive and important to be a part of.
Ben: Yeah. And two footnotes to that that I think are important. One is, it actually started with the universities. So Professor Frank Terman and William Shockley who were developing these technologies and, to support the war effort, you know, basically encouraged their students to not pursue PhDs and spin out and build companies to support the country.
Which is an ironic turn because those same universities kind of spawned the ideas, I would say, that led to their revolt at Google. And then the other footnote is Google/Alphabet has definitely since reversed their position and now do work with the US government and the Defense Department, so we’re in a happy place there.
The Silicon Valley-Washington Divide
Erik: And just to tie the threads together, can we spend just a couple minutes explaining what are some of the headline reasons that the hostility developed, so that we can then get back to the present and describe what changed? Marc, do you want to take a first stab at the high level topics?
Marc: Yeah. Although I want to hear from Katherine, I think in particular on this, because she was both an observer and a participant in it for the last decade-plus. Let’s start by saying there are serious and legitimate differences in philosophy and worldview.
I happen to be watching a lot of movies with my 10-year-old and we watched a movie, Real Genius, last night. It’s a movie from 1985, actually made by our great friend Brian Grazer. It’s one of my favorite movies, an incredible movie, but it was sort of an artifact of the post-Vietnam era in which, spoiler alert, the kids in the movie are going to basically a thinly disguised version of MIT and they’re working on lasers, and they discover to their enormous horror that they’re working on a weapon.
The sort of assumption in the movie is if you’re working on a weapon, you’re evil. They don’t even pause for a moment to ask the question of, well actually, could this actually be super helpful? Or how does this fit into strategy? They just immediately are like, “We gotta...” It was of a moment, sort of very much of a moment post-Vietnam, that the military was evil in general.
There’s been some of that in the ecosystem for a long time. By the way, Ben mentioned Stanford, the pioneering role of Stanford in all of this. Stanford was actually forced in the 1970s to divest all of their military R&D and any R&D that touched the military into something called the Stanford Research Institute. So many top American universities, forever basically since Vietnam, have not allowed ROTC or intelligence agency recruiting on campus.
There’s like a ‘60s dimension of this post-Vietnam thing. And then in the last 15 years, I think it just all got kind of wrapped up in the broader national political drama of our time, basically people feeling like they have to take sides on all kinds of political issues, and beyond that, feeling like political issues are not just political issues, they’re full moral issues and that you’re immoral if you’re on the other side. There’s also a post-Iraq component to it, which is, can we really trust the system that took us into Iraq? Which obviously had a big impact on our national politics. And so that stuff all just kind of snowballed in the Valley leading into what Ben described in the late 2010s. I think there’s a variety of different responses you could have to that. One is you could say you believe that the post-Vietnam generation or the post-Iraq mentality is the correct mentality. You could have that point of view. I don’t have that point of view, but you could.
Yeah, so I think there’s lots of points of view that you could have on that. I just think in the heat of the moment, I think it’s very easy to get wrapped up in the specific, let’s say, politics or morality of a specific situation. It would just come back to the kind of big picture thing, right? Which is: is America in aggregate a force for good in the world? Is it important for the world that America succeed? Is it important that, broadly speaking, American values, Western values, are kind of preeminent in the world? Is it important that the world not devolve into totalitarianism? It was important during the 20th century Cold War that the world not devolve into complete Soviet communism everywhere. I don’t think any of us want to see a world in which we end up with kind of the same scenario with Chinese communism.
And so I think you can just kind of have these big picture views of the importance of the overall thing. I think you can also just say you can have a lot of respect for the men and women in uniform who put their lives on the line every day protecting and defending us. I think it’s a pretty reasonable stance to say that they deserve to be supported by their fellow countrymen like us.
And then you could just also have a view of, as the cliché goes, we live in a society, and part of the society is that the government is going to have certain capabilities and it’s gonna be better if they’re better and worse if they’re worse. So I just think that sort of broader view was getting increasingly washed out by some very specific heated political topics.
I think now is a good time for a lot of people to really kind of take stock of this. Maybe some of the super heated kind of emotions of the last eight years are fading a little bit right now, and some people are saying, well actually we need to think about what the next 30 years looks like and we want to work in a different way.
Cultural Influences
Katherine Boyle: I’ll add two points to that, because I think Marc’s point on culture plays a huge role in this. There’s two things I’ve noticed since moving to the Valley. I moved to the Valley in 2014, but I moved from Washington DC. One of the things that I noticed when I moved is that if you are in Washington DC, you ride the metro, metro stops at the Pentagon, every time you ride the metro, you see men and women in uniform going to work.
So it is a daily part of life that national security issues... when people talk about technology in Washington, they’re talking about aerospace, they’re talking about defense. They’re talking about the companies that they see every day, the billboards for which are very different than the billboards you see in Silicon Valley.
So I think that’s a huge part of it. Just culturally, people are very accustomed to national security missions and needs, and people in uniform in Washington DC, and there’s none of that in Silicon Valley. You could go a decade in Silicon Valley without seeing a person in uniform, so I think that’s a part of it.
But to Marc’s point on culture, I’ve always said that I think The Social Network had probably the greatest influence of the 2010s in terms of being a defining movie of what technology means. And when you look at the Bob Noyce generation of working with your hands, building the physical world, where those engineers came from, versus the story of a Harvard dorm room that inspired generations of college students who were studying computer science in their dorm to come to Silicon Valley. Those are two very different populations and they symbolize two very different things.
So what I think is interesting is probably the 2010s were really solidified by this sort of Harvard-esque Social Network culture that is totally different than the culture of people who are building satellites or building anything in the physical world. And we really didn’t start talking about how those people are necessary for Silicon Valley technology until Marc’s essay “It’s Time to Build,” until 2010... until we didn’t have enough PPE for COVID. And so I think it was really just sort of the kind of, almost the logical 180 to: “What have we done here? We’re not all Harvard grads building in the application layer when we should be thinking about the things that we actually need in the physical world.”
Personal Backgrounds and Regional Perspectives
Marc: Yeah. If I could just, Katherine, I think that’s a great point. I’ll just add one kind of personal component to that, which is Bob Noyce, of course, grew up in rural Iowa, and was of that world, kind of World War II, contemporaneous generation. He undoubtedly knew many service men and women; he and his family undoubtedly knew many soldiers who went to war and didn’t come back, and others who came back wounded.
I grew up in rural Wisconsin. Actually, I grew up in rural Wisconsin, kind of post-Vietnam. Kind of a very similar thing, if you look at where the sort of war fighters in the American military come from, it’s sort of the center of the country in a lot of cases. And so if you grew up in one of those environments, you just realize that this is an integrated part of your society.
Maybe I come from a little bit more of that background and then I’ll tease Ben a little bit. He comes from, of course, the polar opposite, the People’s Republic of Berkeley, in which I’m guessing probably the uniforms you saw people walking around in into your youth were not camouflage, they were something else.
Ben: That is definitely true. Although, you know, I ended up knowing a few people who did go to the service, probably because I played on the football team. I ended up knowing a few people who did go to the service. And I mean, I think that’s one of the things that always struck me as incorrect. I’ll just say about the kind of anti... “we’re not gonna sell stuff to the US military” and so forth.
It’s like so here are friends of ours who are risking their lives to protect us and we’re not going to make sure they have the best, safest equipment and technology available? And that’s your moral position? That always struck me as very weird.
From ICE Protests to Palmer Luckey
Erik: Yeah. I remember a few months ago, Paul Graham was trying to critique Palantir, I think for involvement in some sort of military…
Ben: ICE.
Erik: Yeah. It is just indicative of 2010s insults that don’t work anymore because the culture has changed so much. Within just a few years Palmer [Luckey] could be let go from Meta for sort of political contributions to presumably the “wrong side” and now is a hero for his work in saving American defense. It’s just indicative of how much the culture has changed in the last few years.
Marc: It turned out... I’ll just bash in the counter-arguments for a moment. Having been trying to be fair earlier, I’ll now be totally unfair. A lot of the arguments against, like, they’re very thinly substantive. They’re emotional in nature and very thinly substantive. And so my favorite story, Erik, of your example was there was a big protest against tech companies working with ICE in the first Trump term in 2016, 2017. And there was sort of this revolt where a bunch of tech CEOs kind of did this very posturing thing where they said, “Of course, we’re never gonna work with those horrible people,” who are trying to deal with everything on the border.
I remember actually a friend of mine was running one of those, at the time, one of the most culturally advanced, technologically advanced coolest tech companies in San Francisco. And they had a contract, they had their services used by ICE along with thousands of other organizations. And you know, his employees did the big kind of outrage, “How can we possibly do this?”
And so he actually went out, this is I think 2017. He went out during the heat when the thing was maximum level heat intensity of that time. And he went out and actually got a former director, actually he got a guy who had actually been in the Obama administration, basically the head of border security in the Obama administration, to basically come in and give an hour long seminar at the company: Here’s what’s actually involved with the border. Here are all the things that make it complicated. Here’s all the stuff coming across, not just people, but also children and sex trafficking, and drugs, and all the other aspects of it. He came in to do an hour long talk and the entire company was invited and three people showed up.
I think that level of emotionality, like, it had its moment and I think it’s just over now. And I think where hopefully we and hopefully a lot of others are just kind of back to basics, back to fundamentals, that these missions actually really do, independent of the political mood of the moment, these issues matter. These people matter, these organizations matter, and they deserve to be supported.
Building the American Dynamism Category
Erik: I’m curious for your reactions listening to this conversation as we’re talking about the culture change. And I also want to ask more broadly, you know, we’ve been doing American Dynamism for a few years now, and there’s been a number of companies that have emerged. There was this question for a while as to: would this be just a category where there’s one big winner, which you’ve led for us? Or would there be enough diversity to really justify a whole category against it?
David Ulevitch: Yeah. I think that Ben talked about why Silicon Valley’s roots kind of came from this, and Marc mentioned it as well, but we all have our personal experiences. Maybe it was Ben and Marc at Netscape when they started working with the government and they said, ‘Hey, wait a minute. Everyone sort of has this moment where they realize that your customers deserve the best technology, and if your customers are the government, or if they’re public safety officials, or the first responders, they deserve those best capabilities.’
To hold those back from people just seems wrong, right? And if you’re one of those people that doesn’t wanna support the police, you would at least be able to believe that the police deserve the best technologies to be the most objective and to be the most effective. And so I think for me, for all of us, we sort of come to these moments where we realize that the emotional part of what happens on the border and some of these other places is just not tied to the reality and the facts that we actually have. And we actually spend time with the customers and the people that use our technologies, use the technologies we’re investing in. And in fact, you actually end up coming home at the end of the day feeling really good about how we spend our time and that the time we’re spending working with all these people to support these companies, support the customers that they serve, is time really well served.
The other thing that’s happened, and I’ll get to the Anduril question in a minute, is that we’re at this technology watershed moment where AI is transforming everything from the way computer vision is used in technology capabilities, how analysis and investigative tools are being transformed, how autonomy is transforming the way that we think about robotics and machining, things that will have profound implications on manufacturing.
And so there’s just all these opportunities that have been created that the legacy incumbents are not gonna capitalize on, right? They have innovator’s dilemma issues. They don’t have the best talent, they don’t have the startup mentality. And so there’s always opportunities that just keep getting created at this moment in time.
And I think when you pair that with the absolute necessity that Marc sort of alluded to, do you believe that America’s a force for good in the world? Do you believe we should have a public safety service that is provided by the government? And do they deserve the best technologies? When you pair sort of the demand with this supply opportunity, we just have this moment in time that we have to capitalize on right now.
I think you see companies like Anduril being early in that focused on defense. But there’s gonna be lots of other companies. There’s obviously more than one defense prime. In fact, there used to be many, many large defense companies. They’ve consolidated over time. And I think that’s actually been part of the reason that that’s led us to the place we are in today, where there’s been a lack of innovation. There’s been a lack of competition on price. There’s been cost overruns, there’s been delays in the production process and the defense industrial base.
And so there’s gonna be more than one Anduril, they’re gonna be focused on different areas. But even more than that, what we recognize when we look at these companies, and I think we look at the environment that we’re operating in, is every time we dig in, we actually get a more expansive view. It’s like, okay, Anduril is doing all this manufacturing. It’s terrific, but they need motors. Well, who makes the motors? Where do the motors come from? Okay, they come from China.
Well, how do we make motors here? What do we need for that? That brings us to minerals and to mining. And then we talk about energy. Well, we need batteries. Well, where do batteries come from? What are all the ingredients? Well, we need minerals for those. We have this insatiable thirst for energy and power. Well, can we think about power generation? Can we think about power transmission? Can we think about power storage?
And so our vision for what gets sort of included in all this problem solving and where technology can be applied to fix some of these things that are so legacy, but so critical to the foundations of what we do, we sort of keep shifting left and more expansively, and I think we now take a pretty broad view of where we can apply new technologies, new capabilities, advanced software into all these areas.
And so there will be not just an Anduril in defense, but there’s gonna be, you know, the Anduril of energy, the Anduril of manufacturing. In fact, there’s gonna be multiple, I think, in each category. And that’s only possible because of the moment in time we’re in. There’s obviously gonna be more than one big company in any category.
International Expansion and Allied Nations
Erik: I want to zoom out and also talk about how we think about American Dynamism in the context of our international efforts, because you mentioned Anduril for different spaces, but also countries are trying to build their own version of it. And of course Anduril is also working internationally as well. Katherine we’ll start with you. How do you sort of think about American Dynamism, but also, and I’ll pass to Ben, as we’re building our international efforts, our American and Allies?
Katherine: Well, I think Ben has always said every time, and I won’t put words in your mouth, Ben, but every time that he travels, I mean, the thing that people want is exactly what we named the category, which is: How do we build American Dynamism? How do we build Silicon Valley in our own country, right? Like, how do we build national resilience? How do we think about the future of manufacturing? What are the things that are going to supercharge our own economy?
And I think that by calling in American Dynamism and by talking about how important it is to be building in the physical world, I think we kind of captured a few things that are really important to every nation state. One is, you know, we always point to the fact that 25 years ago, if you looked at the market cap of the largest 10 companies in the world 25 years ago, three of the top 10 were American tech companies. Now that is nine of the top 10, 25 years later. So Silicon Valley has truly, I think, transformed with software the size and scale of how large you can build these companies. Many of the companies that are now in the top 10 in terms of market cap didn’t even exist in 2000.
So I think that’s something where other nations look at what Silicon Valley can build, whether it’s hardware, software, whether it’s companies that are building the physical world and say, “We want to bring this to our own country. How do we do that?” And the spirit of American Dynamism is recognizing that it’s not just an economic zone in a silo. It’s not just pure software. Like the companies that are really having the impact of that scale are building the physical world and they’re interfacing with governments, not only in the US but they’re interfacing with governments around the world. And in order to build many, many more of those companies, we have to have sort of a game plan of how you interface with regulators, with governments, and with the international community.
Erik: Ben, what would you add to that?
Ben: Yeah, I think that geopolitically, we’ve kind of entered a new era. And so if you kind of go back post-World War II, we had kind of bombed our big manufacturing competitors, Japan and Germany, kind of out of existence or outta the game for a while, and America became the dominant economic power in the world for decades.
And as a result, we could be ‘Team America: World Police,’ and just kind of provide defense. Everybody in our strategy was like, ‘it’s actually not only do we not need other countries to defend themselves, if they were to do so, it might be dangerous.’ And so we kind of stationed our troops all over the world, and we built this massive military.
Fast forward to 2025, we’re $38 trillion in debt, and there’s a very competitive superpower economically in China who has just a very different idea about the way the world should work and the way their societies should work than we do. And so now we’re in a position where allies matter a lot. And not only do allies matter, but we need them to be able to defend themselves and to work with us to kind of defend our values, our kind of way of life, our view of freedom, and what it means to kind of be a citizen.
And as a result of that, American Dynamism does need to kind of extend out beyond America to our allies because if they are providing their own defense and they’re allied with us, then they need these same kinds of technologies and products. And so it’s just part of the overall mission that I think we have to pursue. It also makes the economics for these companies work better if they’re kind of a larger market for them.
David: For decades we’ve exported the American Dream all over the world. And when you go around the world and you ask people what dream they have, they always say they have the American Dream. You know, it’s not like someone else’s dream or like the Moroccan dream or whatever. It’s always the American Dream.
And I think we need to export American Dynamism in the same way. People love the idea that they can start a business and improve their station in life and be entrepreneurs. And I think that’s one of the great things that America is able to export. And then, yeah, to your point Ben, we want American technologies in our allies’ hands. We don’t want other countries’ technologies necessarily in our allies’ hands. It’s important from a national security standpoint. Whenever people ask me about defense or war, it’s like, look, we can fight over there or we can fight here. I’d much rather fight over there and we want them to have the best capabilities. So it’s important that we make sure our allies have the best capabilities as well.
China Competition and Strategic Advantages
Erik: Yeah, I want to go deeper on what Ben said in terms of going deeper on China. It seems like a lot of the cultural embracing of American Dynamism stems from some of the realities that Katherine mentioned, like Ukraine, like COVID, and of course sort of this great power conflict with China and people realizing the urgency of and need to prepare for it, and it seems to inspire some of the American Dynamism focus and where we invest. So maybe let’s just briefly take stock on what are the areas in which they’re ahead? What are our unique advantages and how does that inform sort of where we’re thinking for American Dynamism? Katherine, do you wanna take initial stab?
Katherine: Yeah. I thought the conversation that you all had in the last episode with Brian and Chris about what does it mean to participate in wars of mass is one that Silicon Valley really hadn’t thought about until very recently. If you go to Washington, there’s all this talk now about what is known as attritable systems, which are systems where you’re not spending millions and millions of dollars building out exquisite systems that only are usable once. You’re building out very cheap drones, very cheap hardware, where if you lose them, that’s okay.
And the reason why that is so important is because I think a lot of the learnings of Ukraine. If you have an exquisite system that was hundreds of millions of dollars and it can be taken out with a thousand dollar drone, that changes the economics of war, and it particularly changes the economics of a war of mass.
So I’d encourage people to kind of understand the sort of complexities of why it matters so much that we’ve exported our manufacturing, that we can’t manufacture these attributable systems as cheaply and as quickly and in terms of as many as say, China could.
But I’d say there’s also two really important things to take from the war in Ukraine, which is that China has been supplying both sides of that war. When you look at sort of the drone warfare that has come out of the trenches for the last three years, China is a supplier in the same way that Russia is a supplier to its own side. But that means that their industrial bases have been built up in a way that the US industrial base hasn’t built up in the past three years. And that’s something that I think a lot of our American Dynamism companies are very cognizant of.
That we’re at a disadvantage because we haven’t been supplying just-in-time manufacturing to the Ukrainians in the way that, say, Chinese parts have been. And so that’s something we very much have to think about when we think about wars of mass: how quickly can we start investing in our manufacturing base so that we can have this just-in-time manufacturing and these attributable systems to compete against a country that has been building for the last 20 years.
Manufacturing’s Future
Erik: Marc, I’d like to hear your thoughts on manufacturing, because we were talking offline a bit ago about how people hope for the old manufacturing jobs back and how it’s harder to bring those back, but there’s an opportunity for more advanced manufacturing. Why don’t you describe the opportunity as you see it?
Marc: Yeah, so I grew up in sort of what was described as a light manufacturing country in Wisconsin. So we were adjacent to Detroit, and we sort of viscerally experienced the decline and fall of Detroit. We were right in the middle of it, but we were kind of around the edges, so I’ve been thinking about this for a long time.
Look, I think the sort of great offshoring of manufacturing that happened over the last 40 years happened as a result of a set of policy issues, a set of policy decisions, like a set of explicit decisions that were made, and we could spend a long time talking about those decisions, but it was a real set of decisions. There are a set of policy decisions that have to be made. and some of them are in the works now to kind of reverse that.
But kind of to your point actually, I’ll quote Heraclitus: “A man cannot step into the same river twice because it is not the same river and he is not the same man.”
So even if you completely cleaned up all the regulatory stuff and legal stuff, which I think we should, and you sort of corrected the incentives that drove everything out, I think what you’re just not gonna get practically is you’re just not gonna get the old factories back and you’re not gonna get the old jobs back in the way that they were 40 years ago when they were lost.
The reason you’re not gonna get that is actually several reasons. One is that the nature of the products themselves has changed. The things that are getting manufactured in the future decades are much more complex and sophisticated and technologically infused and powered than the things that used to get manufactured. So the products themselves have changed. The methods of manufacturing have changed. So there’s a lot more robotics automation now than there was 30 or 40 years ago.
And then the jobs themselves have changed. Look, if you go into a manufacturing plant in China and they’re assembling phones or building bicycles, you are gonna see a lot of people standing in an assembly line doing the same thing over and over again for 10 hours. And there was a component of that to the manufacturing jobs that were lost 40 years ago in the US. But you’re probably not gonna get the bicycle manufacturing plant back that’s going to build bicycles the way they existed 40 years ago, where the plant’s gonna work the same way it worked 40 years ago, and where the jobs are gonna be the same as they were 40 years ago.
It just is cheaper to do that overseas and that’s just the backward looking view. That’s not necessarily the thing you want to get. What you would like in contrast would be something much better.
Let’s just take bicycles as an example. What you actually want is you want to be making electric bikes. You wanna be making e-bikes, which are much more sophisticated physical artifacts that involve batteries and computers and chips, and you might wanna have all kinds of advanced capabilities in the bike just like people have now in cars and motorcycles, and maybe the bike balances itself.
So like a new kind of product that’s much more sophisticated, built in a manufacturing plant that is more sophisticated and has a much greater degree of automation where you don’t have human beings standing in line turning the same screw every time for 10 hours, but where what you have is a large number of jobs that are kind of ‘blue collar plus’ jobs, and then also white collar jobs and all the associated service jobs around those that are building and maintaining that new kind of plant to build that new kind of thing.
By the way, you can see this if you visit a Tesla factory today. You go visit a Tesla factory today, you see this in action. This is how the world’s leading manufacturers actually do this. So I think what you could do is you can leap forward on the nature of the product you’re building. You can leap forward on the nature of the plant that you’re building, and you can leap forward and actually create much better jobs. Higher paying jobs, higher scale jobs, jobs that are frankly a lot more pleasant, that are a lot more interesting.
So I think you wanna have the futuristic outlook on this, and then if you pair that with the regulatory reforms and you solve all the issues around energy prices and natural resources and everything else that needs to be solved, I think that’s the formula.
By the way, if we don’t do that, all of those things are going to get made in China. Which does mean over time, not just phones being made in China, and not just drones being made in China, but also cars and also robots. The great industry of the future that we can spend a lot of time talking about is gonna be robots, AI in mechanical form, which is gonna be, I think, the biggest industry that’s ever been built. Right now, by default, China’s set up to do that. What an amazing story it would be for America in the 21st century, which is we re-industrialized not to build the products of the past, but to build the products of the future, and I think that’s the opportunity.
Erik: Let’s go deeper there. I wanna ask you the same question I asked Katherine about our unique strengths relative to their unique strengths, because I heard you say this framing in another conversation which I really liked, which is: are we gonna win by being more like us or more like them? So why don’t you flesh that out a little bit, Marc?
Democracy vs. Dictatorship
Marc: Yeah, so we actually had this fight during the first Cold War with the Soviet Union. And there was just this raging debate that started in the 1920s, but was fully raging in the fifties, sixties, seventies, including when I was growing up, which is basically: democracy or dictatorship? Free markets or state directed capitalism?
This is a very real debate. One of the sort of little fun facts of how kind of screwy everything got in the 20th century was common American undergraduate economics textbooks into the 1980s would state as fact that communism was a superior economic system to capitalism. And they would have these charts of projected growth rates and they would show that it’s purely inevitable that communism was gonna outperform capitalism because it was a better organized system. It had top-down controls. It didn’t have all this messy inefficiency of free market capitalism.
So a lot of economic experts in the US, including ones with big newspaper columns and big media footprints, were basically selling the siren song of communism all the way through the late eighties.
And the reason I say that is that debate was actually happening all through that period, which is: Are you better off having a Soviet style system where the state directs production? If the state decides that you need to build tanks or planes or cars or anything else, like they’re gonna direct that’s gonna happen. It’s gonna happen with maximum sort of centralized efficiency. It’s gonna happen with the entirety of national resources lined up against it, everything orchestrated in the right way to maximize production.
Or are you gonna have the American system, which has aspects of communism here and there, but it’s much more dynamic, much more flexible. And so in the American system, are you gonna have a higher degree of sort of independent action on the part of individual capitalists, independent companies, independent technologists? Are you gonna be inventing a much broader range of new technologies? Are you gonna be deploying those technologies in much more creative ways? Are you gonna be just much faster on the innovation curve ‘cause you’ve unleashed the sort of inherent creativity of all your people, rather than just ordering them to all stand in the line together?
And then, you know, the capitalist approach is messy. It has boom bust cycles and it has companies go outta business and sometimes there’s too much competition and profits go to zero. But you have dynamism on your side. You’ve got freedom, exploration, creativity, innovation, free market competition, and dynamism on your side.
I think that question was asked and answered in the 20th century quite definitively. You know, it is back today. The new upcoming mayor of New York City is proposing government run grocery stores, the exact same argument is back. And there are people in positions of great power who still wanna have this argument. I think it’s long settled. It does seem to be back.
It is coming back and I think it’s fundamentally the same thing. And by the way, it’s not that the centralized approach doesn’t have certain advantages. If you wanna pump out 20 million tanks or something, or if you want to pump out a billion phones, there is something to a top-down centralized approach. Every company within itself is a form of a dictatorship that does that. So there is some advantage to China at being able to kind of direct all the national resources. And there is something frustrating being on the American side where like, in theory you all share the same mission, but in practice you’re too busy beating each other’s brains out to ever get to the shared purpose.
Having said that, I think we won Cold War I by doing it our way. I think we win Cold War II by doing it our way. Every time somebody stands up and says, ‘We need to get to be more like China,’ I always ask the counter question, where we need more state control, more top down, more direction, more planning—I always just ask the opposite question: Well no actually, what if we take the opposite approach? What if we become more like us? What if we lean even harder into innovation and even harder into creativity and even harder into entrepreneurship? Isn’t there so much more that we could do if we got to be more like us?
Obviously I think that’s what we should be doing. It is interesting that that debate is back though.
Defense Procurement and Five-Year Plans
Katherine: Inside the Department of War, Marc sort of defined the Chinese system, but he also sort of defined one of the major problems with procurement, which is that the Department of War makes its decisions about what weapons it’s going to buy on five year cycles, 10 year production cycles. And it’s very difficult to correct the ship because of the way that we’ve structured our procurement process going back to the sixties and the seventies.
So I think there is something about the, you know, if you look at the last 25 years, the second American century has been a story of American dynamism, innovation, creativity, and how on a global stage that has actually led us from three companies in the top 10 to nine companies in the top 10 that are tech companies.
But the Department of Defense really hasn’t benefited from that. It’s the exact same companies that were founded just after World War I, solidified their dominance in World War II, and then mergers in the 1990s post the Cold War with this view that we’re post-war, we don’t actually have to worry about things anymore. So there really is sort of a Soviet style procurement system that’s still at work inside of the Department of War. And there’s many good people in Washington trying to change that to allow for this American dynamism to actually flow into the places that it needs to exist.
Marc: By the way, just one fun historical note. Katherine, you would know this: Does DOD run on five year plans?
Katherine: It depends on what types of programs you’re talking about, but I mean, it’s not every year that these contracts are renewed. Some are three to five, some are 10. It is a top down system where we’re trying to predict the future in the same way that China does.
Marc: So it’s actually just a historical tweak. If you work with big companies a lot, one of the things you see is they often have five-year plans. They have five-year planning cycles, five year strategies. Actually, the creator of the concept of the five-year plan was of course Joseph Stalin. It was literally the Soviet Union run as five year plans. And actually between the 1920s and 1950s, that idea actually mainstreamed itself into Western society and into western business. And has stuck there ever since.
And by the way, it routinely doesn’t work. Like it routinely fails because the world is too dynamic and that’s not nearly a fast enough iteration cycle. And yet people still stick to the five-year plan thing. It’s like the Soviet communism hand coming out of the grave kind of still strangling our ability to actually advance and get our jobs done.
And look, just to double underline what Katherine said, the battlefield itself is changing. And so one of the things that you see, for example, in the Russia-Ukraine conflict is an incredibly high rate of iteration on the battlefield techniques and use of technology on the battlefield. I don’t know whether they have a five day plan in the Ukrainian military in terms of how fast they adopt new technology and adapt what they have. And that’s what they need to be able to fight. And if they didn’t have that approach, they would’ve lost a long time ago.
The Russians are trying to match them on that. And so the old model of World War II era mass machinery and mass men to win wars, those days are long over. Even the modern version of trench warfare, you still have this very rapid iterative cycle time. And again, this should play greatly to America’s benefit, if we unleash our strength, which is if we take our strength of flexibility, innovation, creativity, dynamism, and apply that into the battlefield of the future, like we really ought to win if we’re really able to do that.
Ben: I would just add to Marc’s thinking that going the other direction for the US is literally fighting everything about our culture, everything about our history, everything about our system. And the fact that it has appeal is, I think, like a weird psychological defect that some people have where they just always want central control. And they always think it’s better.
And this is a great investing strategy: if you see a company with a five year plan, short it. It works every time and yet they still do it. It really has no basis in anything. We saw this actually, weirdly, in networking. Marc and David and I all come from the networking field, and when decentralized networks and packet switching came, people fought it to the death. Even after, when Kleinrock mathematically proved it worked, like people would still say it would never work because you couldn’t control it. And of course history has proven them all wrong.
But you know, it’s not logical. It is just like some kind of brain defect that they have in their psychology where they can’t even visualize the fact that a system could work well that’s decentralized.
David: I think we see this over and over again. I’m curious what Marc thinks, but when I think about Project Warp Speed, to me that was promoting capitalism at its best. We said, “Any company that creates a vaccine, we will distribute it.” So it’s not that there’s no role for government, but they really stepped aside and let the private companies come up with a vaccine, which happened much faster. And we had, depending on how you count, three working vaccines in a record breaking time.
If you look at what China did, they did not have that, and they were unsuccessful because they weren’t able to lean on creativity and ingenuity. I think when we look at defense, some of the defense reforms that we wanna see happen are things that actually push more toward entrepreneurship. Not on past performance, but what can you do for me today? What can you deliver today? Can we put out the demand signals and then have you actually deliver it against those? Not look at what you did 10 years ago, 20 years ago.
The idea that the government currently puts out RFPs that have a criteria of ‘what have you delivered in the past,’ no matter if it was over budget or over time, doesn’t lend toward leaning towards startups. So some of the changes we’re pushing for, I think will benefit all companies and certainly will benefit America and actually lean more toward that idea of entrepreneurship and into what we’re best at.
Ben: Yeah. And even an equally good example of that is the whole chip industry in the US kind of came out of the race to get to the moon, where the US government specified, ‘Look, if you build a chip with these characteristics, we’ll buy it.’ So the government can set rules, can create incentives, can do a lot of good things, but centrally controlling everything tends not to work so well, at least for us.
Um, and so the government is, can set rules, can create incentives, can do a lot of good things, um, but centrally controlling everything, uh, tends not to work so well, at least for us
Erik: Quick logistics check. There’s 10 minutes left. I’m planning on closing with the SF and DC because it sort of involves Ben and Marc, unless someone here tells me they want to make sure that we get to something else that uniquely involves Ben and Marc.
Marc: So David and Katherine, are there other AD specific things that you wanna make sure we hit?
David: I mean, the only thing that I would maybe touch on is this idea of sort of shifting left in the technology. So why look at mining? Why is that a venture scale opportunity? Nobody’s applied software and advanced software to the mining process, to the refinement process. We have a long history of it, like going back to cobalt. Nobody’s looked at battery technologies. Why is it so important? Motors, things like that. Why that’s interesting and important, it’s the same reason we do AI infrastructure. We don’t just do AI apps. The same thing is happening when you look at the supply chain. That could be one topic.
Marc: Yeah, Erik, I think we’ve done a lot on the firm politics and stuff in the past, and Ben and I are not particularly trying to create new visibility for us on those topics right now at the moment. So I think we should dig in on that. I think it’s a big deal for us to be doing energy and minerals. And so maybe we can explore that and sort of what that means for the future.
Energy, Mining, and Critical Infrastructure
Erik: So segueing more concretely into some American Dynamism investments. Why don’t you share a couple of the areas we’re particularly excited about or opportunities we’re looking at?
David: Yeah, I’ll start and then maybe Katherine can jump in. I think energy is a great one. You can sort of zoom in at any level you want. At a macro level, we’ve just seen throughout history: When you pour energy into a system, you get just tremendous economic output, economic growth.
If you look at where we are today, not even looking forward, we have this insatiable thirst for energy, whether it’s for AI compute, or whether it’s for people switching to electric vehicles. Or even moving to systems that are just electrified. I mean, even the entire Department of War has multiple initiatives about moving toward electrification.
And so that just has this huge consequence of creating opportunities for power generation, power transmission, power storage. And the question is, are there gonna be opportunities that we think are venture-scale there? We think there are. We’ve invested in nuclear power companies, a company called Radiant Nuclear. We’ve invested in X-energy. We’ve looked at lots of battery technology companies.
And we don’t just think there’s great opportunities here, but even just pragmatically, we know we have three companies now that are on China’s “unreliable entities list.” Which is an amazingly named list. But they’re forbidden now from buying batteries from China. So if you care about the defense industrial base, if you care about anything that requires a battery, even consumer electronics, you need to have battery capability in the United States or amongst our allies.
So we just think there’s exciting opportunities there on the energy side. In the same vein, if you continue to shift left, you look at critical minerals and whether it’s magnets, whether it’s motors, whether it’s copper for manufacturing, whether it’s steel. There are all these areas where there are massive economic spend-centers, trillion and trillions of dollars are spent in mining, in construction, in manufacturing, where software has really not looked into these areas. Nobody’s figured out: can you really apply advanced algorithms to figure out how to do mining better?
And while that sounds like a quote that would be memeable of like, “why that sounds so stupid,” it’s actually not stupid. There’s pressure, there’s temperature, there’s a lot of work that goes into mining. Not just in site selection, but actually in the process. And these are areas where I think we now see incredible talent shifting their eyes toward these areas. And we just think it’s a great opportunity and it will ultimately be good for America.
The New Generation of Defense Companies
Katherine: The other thing that we’ve really noticed, if we go back to sort of our early investments, you know, again, investments like Anduril and Shield AI, where we led the Series A in 2016. We’ve been around this space for a very long time. But those early companies, they had no other choice but to be what’s known as the Prime. They had to compete against the legacy incumbents. They had to try to get their own contracts in working with the Department of War, and it was an uphill battle for them as they were the first.
But what we’re seeing now from our companies, which I always refer to as sort of ‘Defense 2.0,’ are the ones that really started around 2021, 2022. These companies are happy to work with the existing legacy primes. I’d also say that a lot of the industrial base is kind of waking up and recognizing that they need to bring software, they need to bring the best and brightest engineers from Silicon Valley into what they’re offering government, and they’re already on contracts.
So one of the things that we’ve seen, at least in the kind of second cohort that’s come through the American Dynamism Fund. One, is that these companies can grow so much faster and they have a much richer customer set that they’re supporting because of that. So I think that’s something that’s sometimes lost on people is just: How is it possible that these companies can really be working with the government this quickly? And one, it’s because you have a government now that understands why they need to bring these companies in. But the real reason is everyone’s sort of woken up to the opportunity. And there’s a lot of partnerships that are happening among the legacy and the new players and they’re reaching, I would say, extraordinary heights much earlier in their trajectory than the previous generation of space and defense companies in 1.0.
David: I’d also say a lot of these founders are coming from some of these version 1.0 companies, so they’ve seen what success looks like in these areas, at least at some level. If you come out of SpaceX and you were there for 6, 7, 8 years and you saw how SpaceX got built, you have a huge headstart in starting the next American Dynamism company.
Founder Profiles
Erik: You mentioned some winners in the portfolio and others like Saronic and Apex and Base and Flock and many others. As they’ve emerged, what can you say about the founders and the types of founders that are building these companies? Maybe what’s interesting about them culturally or philosophically or their backgrounds or what their understanding is that maybe others don’t fully appreciate. Katherine, what would you say?
Katherine: Well, I’d say what we’ve always pointed to is that these companies are built across America. So Saronic is based in Austin, Flock is based in Atlanta. And if you look at the founders, it’s not your typical, again, to go back to the Bob Noyce, Harvard Dorm Room comparison, it’s not your typical software engineer coming out of a college dorm.
These are people, in the case of Dino at Saronic, you know, he served in the Navy SEALs enlisted for 11 years before going back to school and then founding companies and learning private equity. So there’s something about being entrenched in what the customer needs that I think is very specific to American Dynamism. A lot of the best founders, they’ve either sold to the government before, many of them have clearances before they even start their company. They understand how to speak the language of the customer, which I think is very important for this category.
It’s something that we always really look for when we’re investing in new companies is: How deeply does this founder understand the customer? And it’s very hard to understand the customer if you’ve never served, if you’ve never worked inside of government. And so I’d say that’s actually something we very much test for across these categories is how deeply do you understand who you’re selling to?
Looking Forward
Erik: With just two minutes left, anyone else want to get any last points? Do you wanna say maybe just gearing forward, we’ve covered the past and present of American Dynamism. What can you tell us about what we’re looking forward to?
David: I think that we’re at this exciting moment in time where both the need and the demand for new technologies, whether it’s in public safety or in the Department of War or anywhere else in the government or just anywhere across the United States is greater than it’s ever been.
We’re also at this major transformation moment. Marc mentioned, I think robotics is AI personified in physical form. I think it’s gonna be extremely exciting. I think it’s a lot closer than people think. If you’ve been in San Francisco, gotten in a Waymo, you already know that autonomy and self-driving cars is absolutely gonna be the way forward.
We just see that we’re just sort of entering this incredible moment in time where new technology is meeting these very, very legacy and old categories, and I’m very excited about where it’s going. We see more and more founders entering the area across all these different domains. And for us, to me, it’s just like we’re just in the early innings of America sort of on a major comeback when it comes to introducing new technology into these key areas. I think it’s pretty exciting.
Erik: That sounds like a great place to wrap. This is a great conversation on the history and philosophy and the past, present, future of American Dynamism.
David: Thanks.
Ben: Thank you.
Erik: Cool.
This newsletter is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. Furthermore, this content is not investment advice, nor is it intended for use by any investors or prospective investors in any a16z funds. This newsletter may link to other websites or contain other information obtained from third-party sources - a16z has not independently verified nor makes any representations about the current or enduring accuracy of such information. If this content includes third-party advertisements, a16z has not reviewed such advertisements and does not endorse any advertising content or related companies contained therein. Any investments or portfolio companies mentioned, referred to, or described are not representative of all investments in vehicles managed by a16z; visit https://a16z.com/investment-list/ for a full list of investments. Other important information can be found at a16z.com/disclosures. You’re receiving this newsletter since you opted in earlier; if you would like to opt out of future newsletters you may unsubscribe immediately.