Blake Scholl on Supersonic Flight and Fixing Broken Infrastructure - Live at the Progress Conference

Conversations with Tyler is produced by the Mercatus Center at George Mason University, bridging the gap between academic ideas and real-world problems. Learn more at mercatus.org. For a full transcript of every conversation enhanced with helpful links, visit conversationswithtyler.com. Hello everyone. Welcome to yet another event at the Progress Conference. I'll be chatting with Blake Scholl. Blake has done more than any other human to make civilian supersonic transport a reality. His company, of course, is Boom in Colorado. Blake, welcome.

Good morning. Let's convert some jet fuel into human progress.

There's plenty about boom online and in your interviews, so I'd like to take some different tacks here. This general notion of having things move more quickly, I'm a big fan of that. Do you have a plan for how we could make moving through an airport happen more quickly? You're in charge, you're the dictator, you don't have to worry about bureaucratic obstacles, you just do it.

I think about this in the shower like every day. Um, there is a much, much better airport design that as best I can tell has never been built. Here's the idea. You should put the terminals underground, and airside is above ground, terminals are below ground. Imagine like a design with two runways. There's an arrival runway, departure runway. Traffic flows from arrival runway to departure runway. You don't need tugs. You can delete a whole bunch of airport infrastructure. Imagine you pull into a gate, the jetway is actually an escalator that comes up from underneath the ground, and then you pull forward. So you can delete a whole bunch of claptrap that is just unnecessary. And the terminal underground should have like skylights so it can still be incredibly beautiful. And if you model fundamentally the thing on a crossbar switch, there are a whole bunch of insights for how to make it radically more efficient. Sorry, this is a blog post I wanna write one day. Actually, it's an airport I wanna build.

No one does this, right?

I don't think so.

What's the main obstacle? Common sense or lack of common sense?

No, lack of business model and alignment. Like we've socialized airports. Right. And so— and we've limited their revenue to $5.60 per in-plane passenger. This is why airports trap people in shopping malls, because it's the only way for them to make any money. Right. So you have to privatize all of that infrastructure. You have to— you have to, like, invent a new revenue model for airports. I don't actually know what that is, but it needs to be created. And I think something like VTOL is probably a key enabler because we need places to build airports. That are actually accessible from population centers and that, that there's a whole real estate problem to be solved here. So I think the, I think the design aspects of this are the easy part. The hard part is how do we unlock building and how do we create a revenue model around this when we've got this sort of entrenched model that is limited by, you know, limited by regulation of like literally what you can charge. I think the thin edge of the wedge here is creating supersonic terminals. And so that's something I actually hope to work on is creating a specific terminal that will be around Overture that will allow you to kind of fly through the terminal and optimize for getting to the plane quickly. Like, imagine a guarantee if you're there 15 minutes before the flight leaves, you're actually going to make your flight, like delete all this other buffer time.

So when I move through security, that is not at supersonic speed. PreCheck has gotten more crowded. CLEAR won't actually take my application, and that's more crowded too. How do we make that part of flying quicker?

It's all a farce. I think we know that. Like, personal story. So the first time I got married, I flew from San Francisco to Seattle, got married outside Seattle, then flew through Heathrow to Rome, and we're in Venice, and my wife says, like, "Hey, Blake, could you find the, like, whatever in my purse?" And so I start, like, rifling through her purse, and I find a box cutter. Like, the literal weapon used on 9/11 has been brought through SFO, Seattle, and London Heathrow with nobody catching it. Like, if the best— we are not actually stopping terrorists. What we need is to solve the regulatory problem where nobody wants to— nobody's incentivized to do do anything that's risk-on. Like, you know, if you're the bureaucrat that perpetuates a bit more security monstrosity, like, kind of, you know, it's totally fine. But if you're the one that actually takes a bit more risk and then something happens, then that's career-ending. I think that is a fundamental problem we need to solve. There is a problem that regulations, particularly anything that's got a tie to safety, tends to be a one-way door. Once it's put in place, it's very, very difficult to muster the political will to reverse it. That's a problem we have to solve. If we can crack that one, then we can start rolling back a lot of this stuff that everybody knows does nothing.

But you say you think we should return to the pre-9/11 regime. My friends can meet me at the gate. They don't need a boarding pass. There's no scanner.

Something like that. I mean, the, you know, I think we are actually safer than we were on 9/11 and in two important respects. One is reinforced cockpit doors. Those are a good idea. We should keep those. Another is that the passengers know to fight back. Like, and this effect happened so quickly. It is why the fourth airplane didn't make it into the White House on September 11th. The passengers got the word that, you know, this was not just like a money heist, that they were being converted into a missile. And so they knew to fight back and they won. So those two things did make us meaningfully safer. But like not filtering box cutters and preventing us from bringing like bottles of water, like none of that makes anything safer. You know, so I think we should have, you know, a trusted traveler program. Anybody who flies as much as I do and has never taken an airplane down should be trusted. The complex, difficult parts here are about how you do the political transitions. That's the unsolved problem, not airport security.

I've been to a lot of countries in the world. I'm sure you have too. I've never seen a country that doesn't in some way copy the US system. Why is there so much implicit regulatory cartelization? No one is willing to do anything very different.

Oh, it's not implicit, it's explicit. There is international harmonization of security. That basically results in, like, the worst rules in any country metastasizing. There's actually sort of a reason for it, which is to support transferring and connecting passengers without having to reclear security. So whoever has the most, like, obtuse security rules, they tend to spread globally so that passengers can connect and not get rescreened.

So I'm at the United desk. I have some kind of question. There's only 2 or 3 people in front of me, but it takes forever. And I notice they're just talking back and forth to the assistant. Like they're discussing the weather or the future prospects for progress. Total factor productivity. I don't know. And I'm frustrated. How can we make that process faster? What's going wrong there?

The thing I most don't understand is why it requires so many keystrokes to check into a hotel room. Like, what are they writing?

I have the same question. But you run a company which has internal operations and you send your employees around to hotels, right? You must think about systems more generally, how to make them more efficient. What would you do to the rest of the world that maybe you've already done at Boom?

I think there's something here that is almost like at a deeper, like, cultural or almost like spiritual level. We live in a world full of things that are radically improvable, and yet we've learned to live with most of the problems. Like, like the one that drives me the craziest is actually traffic. Like, we sit literally more than a working month per year. The average American spends sitting in traffic and we all just sort of accept it. You know, we don't, really build more roads because we kind of give away road access for no variable cost. They're always flooded. The traffic engineers have this concept they call induced demand, which is basically like if we build more roads, people just drive more. So we guess we shouldn't build more roads. I think this is ridiculous. The real problem is a lack of a price system. I get flamed on Twitter for this view because I think every road should be a toll road and then we could actually solve traffic. The meta point is we are surrounded by things that we've accepted that could actually be made way better. And that's one of the things I hope that this movement can kind of teach the culture. Is that we should, we should not be satisfied with where we are today, that virtually everything can be made better, and that we should challenge it.

But say for traffic, and I agree with you, but the United States is not Singapore. We have New Jersey. There are all sorts of people who have cars with no electronic transponder or whatever device you want to hook to the car. The car is illegal, the car is not registered, the person is in the country illegally, whatever. You pile on a bunch of things where there's no accountability for actually getting the toll paid, we just let those people through and tag who it is we can, or how's it going to work?

This seems like a very solvable problem, right? Like, you can read license plates, you can mail tickets. I mean, if we went to a—

Well, like, District of Columbia gives me tickets in the mail. I don't— in fact, I pay them, but I don't have to pay them.

And most people don't. I'm going to start mailing tickets to people and see if I get checks.

You will. This is why some people don't pay them.

If we go to this system it will be such a massive percentage better that we can then deal with the cheaters. And like, you know, I'm sure we can find some way to solve that problem. That's the small part of this.

But if I think of District of Columbia, when I get these tickets, there's some zone that is marked as 25 and I'm driving what I think is a perfectly safe 37 and I don't even see that it's 25 and I get this weird ticket in the mail and I'm baffled as to what happened. I just pay it. Shouldn't I have the same concern with tolls? That I don't trust the government in the district, and they're just gonna screw me over because I'm from Virginia, I don't in any way vote there. Like, do I really want them to have tolling authority over me?

The real thing is, I mean, so imagine if we did with food what we've done with the roads, right? Where you can just go in, we pay for the grocery stores with taxes in various ways, and therefore everybody feels entitled to show up at a grocery store whenever they want, take whatever they want. There would be a line out the front door. You couldn't possibly butcher enough cows. The cow, there would be like agricultural engineers and they would talk about induced demand and it's like, you know, it doesn't matter to have more cows, people just keep eating the steak. And then we would have this conversation about like, how could you possibly charge people for grocery stores? Like, how would you figure out who took what? Like, we could sit here and talk about it abstractly, but like, we all know that's absurd because there's a thing called a cash register. And you know, and pretty soon Amazon will figure out how to delete those and you can still get charged for what you take. I think these are very important questions.

But I trust Amazon and I don't trust the District of Columbia. I don't even trust Maryland, frankly. How about boarding an airplane? My sense is in Western Europe, people board an airplane more effectively, more quickly, with no loss of value. How can we improve that in the United States?

It's at the intersection of airport design and airplane design. Fundamentally, airplanes need more doors. That would help. The doors are generally in the wrong quantities in the wrong places. The other thing that we need to do, and this is part of why it's not a solved problem, is we need to fix checked baggage. Because baggage check is unreliable and slow, we have people carrying onto airplanes things they absolutely do not want to carry. And if we fix airports such that baggage check is fast and reliable, then we can stop having carry-ons and we can get on and off airplanes much, much faster than we can today. And that would actually be the biggest win. Like, imagine, imagine an experience where you take your Uber to the airport. The bag that today you would carry on is in your trunk. You step out of the car, someone grabs, maybe even a robot, grabs your bag from the trunk. You don't see it again. After you land, you get a push notification on your phone that says your Uber's in slot 7A, and by the time you get to your Uber, your bag is back in the trunk. So the customer experiences your bag teleports from the trunk of your Uber at your origin to the trunk of your Uber at your destination. Like, that's how this should work. And then you don't carry all this stuff, and it's much faster to get on and off airplanes.

What else would you change, or maybe I should say, will you change, about the interiors of airlines?

Ooh, we— so we're cooking something on this that I can't steal my own thunder on. But the meta-insight is design the interior and the airplane together. I think you get the best results when you've got designers that think like engineers and engineers that think like designers. And then just hold a very high bar. Like, long before I started Boom, every time I got in an airplane, I would ask myself, like, what would this be like if somebody who cared as much as Steve Jobs or Jony Ive had designed it? And it's a very dangerous question to ask because it makes you start to notice things that you can't unsee once you've seen them. Like, okay, here it goes. If you look above the window seat on basically any airplane today, you will notice there is like a dark spot in the same place in every row. This is from where people stand up and they bump their heads and their head schmutz ends up on the underside of the overhead bin. Sorry, you can't unsee that. And nobody ever cleans it. So it's sort of a one-way door. If you care about this stuff, it's all very solvable. I think there is something that we'll reveal, I don't know, maybe a year or two, that is a striking interior that is possible only when you design the airplane and the interior together.

And how will your new interiors improve the quality of the food? It will make that possible too, right?

I mean, get there much, much faster. You can then go to any restaurant you want.

But still, say it's a 2-hour flight. You might want a snack.

Yeah.

You know, on Emirates, I want to eat the food. It's probably better than the restaurant I would have instead.

I think this is an airline problem to solve, really. You know, or maybe, maybe it has to do with one of the reasons it's very hard to change the service model on-site airplanes is the service delivery models are actually written into union contracts. Like, if you remember Virgin America, you could order snacks through the in-flight entertainment screens. And then when Alaska took them, that got undone.. And this was because it required a different union contract. And so that part has to be unwound or refigured somehow in order to be able to change the service delivery models. And then ultimately that impacts the passenger experience.

When we deboard planes, are Americans too polite? I much prefer the Italian method, however you would describe it. But you're just trying to get off the plane. I think that de facto ends up being more polite.

No comment.

No comment. When you worked at Amazon with Jason Crawford, by the way, who's founder of Roots of Progress, what did you learn about speed? So I never feel I'm waiting for my Amazon packages. It's quite reliable. And you were there. What did you internalize from that?

Oh boy, so much. I think the single biggest thing— so I worked at Amazon and then years later I worked at Groupon, and both companies said a lot of the same things about vision and culture, but they were actually completely different. And the study in contrasts was really incredible. And it was a study in long-term decision-making versus short-term decision-making. So I remember back in like, ooh, this will date me. It was early 2000s when CDs were still a thing, but they were sort of starting to be on their way out. And what would happen is at the end of every quarter, the record labels would panic and be like, ah, we're going to miss the quarter. This is going to be terrible. Like, quick, let's stuff the channel with lots of CDs at a discount. And Amazon would run the math and they would sort of do the discounted cash flow analysis and say, okay, great. If we take a whole bunch of like discounted CDs, we will actually be able to sell them. Our profits will go up. It will make our Q4 look terrible because we'll have this glut of inventory, but our Q1 will look amazing. And the long-term cash flows support this decision. And so they would do it every time. And, you know, when I worked at Groupon, I'm looking at a couple of people I worked here at Groupon with, they'll know this story. The second half of my time there, I ran the email business and this was like the most uninspiring job you could have because I was like running the world's largest spam cannon. I'm so sorry. I did quit. But I would get— I would get a call from the CEO and it would come every quarter and it was like, The quarter is soft. Send more email. And I was like, dude, that's why this quarter is soft, because we did that last quarter. And he's like, I don't care. Send the email. Like, fix it next quarter. And we'd like email the same customer like 3 or 4 times a day, you know? And then the conversation would be like, why is everyone seeing the same deals over and over again? It's like, yeah, because we send more email than we have deals. It's a case study in long-term versus short-term decision-making. And the thing Amazon did very, very well is make long-term decisions while being incredibly short-term vigilant. There was operational excellence moment to moment, day to day, while all major decisions were made with a view of the long term. And that is, I think, super, super important.

And how is it they kept that as the company grew much, much larger?

Well, I haven't been there in 15 years, but it was baked very deeply into the culture. In any— customer centricity was so much in the culture that any debate about any decision was kind of resolved through like, okay, what's actually best for customers? And that was like, it was sort of the cultural trump card that was handed to every employee for how to win any debate in the company.

Given that you worked on Groupon, surely in the shower at least, you've thought about whether we can improve the quality of online reviews. Is that hopeless at current margins? A long time ago, some of them were useful. Is there anything we can do there, or we just need to write it off?

It needs to be reinvented. Everything is 4.5 stars, right? Like, I think percentile ranks would be a really useful thing to do. Like, this is in the top 5%, this is in the bottom 5%, and you get away from the everything's 4.5 stars. But then there's a very different problem, which is how you get good quality inputs. And I think that's— I don't know the answer to that, because there's such a financial incentive to stuff the review ballot box.

Given that the British and the French did the Concorde SST to begin with, Is there now in those places residual interest in supersonic civilian transport or no, that's just gone, that tradition?

Oh no, there's tremendous latent interest. I mean, I sense it very strongly in the UK. Maybe it's there in France too, but I don't know 'cause I don't speak French. But no, like there is deep, deep British pride in Concorde and there's sort of a latent like frustration that it went away and didn't go forward. We've hosted a whole bunch of retired Concorde pilots at Boom that actually, a poignant story, The first Concorde pilot to fly the Overture simulator was the guy who did the last commercial Concorde flight, and he flew the simulator with his daughter, who is also a pilot now. And there are a lot of people cheering for a renaissance here.

How is it they beat the Americans to begin with? Because JFK said, "We're going to do supersonic transport." Obviously, the moon thing worked out in some manner, but our SST did not.

So I get shot at conferences like this because I think Concorde never should have been done and Apollo never should have been done. And yeah, I can— I feel that.

But still, we lost, right? There's plenty of races that should be done where we win.

I think we actually won that one because it was a dumb contest. So let me unpack this. So in 1969, we landed on the moon, and we also flew Concorde for the first time. And I think if you had wandered around the streets and you found like the most pessimistic human anywhere on the planet, no one would have told you that in 2025 you won't be able to fly supersonic and you won't be able to go to the moon. Right? So these things didn't actually work. And I think they were— what we did is we built really impressive tech demos, not products. And that's what happens when you have government-spect innovation.

Right?

So Concorde was a joint venture between the French and British governments.

Right?

Like, think about that. Usually when there's a joint venture between the French and British, it's a war. Like, it's actually pretty remarkable that airplane ever worked. But nobody thought too hard about, like, the economics. It pretended to be a commercial airliner. But let's think about it. There are 100 seats on that airplane. They're very uncomfortable seats. Like, we have a couple in our office. The one thing they have that our seats won't have is an ashtray. You could think they're out of the back of, like, Ryanair. And yet, adjust for inflation, a Concorde fare is $20,000. And it's the 1970s. There is no large premium international travel market. So across 27 years of flying, that thing flew 52% full, half empty, even on the most popular route, Newark to London. And so that's why only 14 airplanes ever entered service. It never made any economic sense. Like, supersonic should have started with a supersonic private jet that could take a handful of well-heeled people coast to coast with a small airplane, limited range, you know, and that would have kicked off a whole S-curve of innovation that would have us all going Mach 5 by now. But we didn't. And instead, what we did is like convince the world that supersonic flight was impractical. We made it a competition between nations, so we banned supersonic in the US. You know, that ban persisted from 1973 to July 6th of this year. Thank God it's gone now. Thank you. And so we really— and we had a half century of no progress. So I think these things have to be done in a commercially led way such that you don't create— I mean, think, dump on Apollo here for a second. Incredibly inspiring accomplishment, right? Let's all acknowledge that. Like, what a triumph of humanity. At the same time, we created a supply chain behind it that was cost insensitive and a bureaucracy that wanted to perpetuate its own existence. And then 50 years goes by before— thank Elon for realizing that we need to go tackle economics of space if we actually want there to be an enduring human presence in space.

In 1980, they thought that a route London, Bahrain, Singapore was commercially viable. How could they possibly have thought that? Like, even Singapore then is not extremely wealthy. Bahrain has hardly any people. Like, what was going on in their minds?

Maybe they were smoking something good. I don't know. Like, national prestige goes a long way on these things, right? The airlines— this is the other thing that's kind of messed up in this industry is a lot of the airlines are state enterprises, and they're literally called flag carriers. And so things get done for show, and for show is a really, really bad motivation. You can sometimes get the tech demo, but it doesn't lead to good places.

Now, I've read that testing and construction for the Concorde, that it only took a total of 6 years. That's almost impossible to believe today.

I think it was more like 12.

12, yeah. But that's still a pretty good record.

Yeah, we've gotten really slow. Think about the conversations around Apple today. Like, if this year's iPhone is not that different from last year's iPhone and everyone's unhappy, And yet, does anybody know the year in which Boeing and Airbus last launched an all-new airplane program? 2004. 21 years ago. No new airplanes. And we can't even do the light modified ones successfully. I think part of what's happened here is it's difficult to regulatory certify an airplane, an all-new one. And, you know, it's like, I think Boeing kind of goes to the doctor and it's like, "Doctor, it hurts when I do this." And the doc says, "Well, don't do that anymore." And Boeing said, "OK." And so they literally— like, Jim McNerney, who was the CEO in mid-2000s, literally said, "No more moonshots." And what he meant is no new products. And it's unbelievable. Like, this is the company that used to be literally in the moonshot business. And instead, what we have to do is figure out how to make that innovation loop, that iteration loop, much easier to do. And so that's what I tell my team every day. Don't plan to certify once. Plan to certify repeatedly. Plan to do it iteratively. Let's get with the regulators and figure out how we prove out airplanes much more efficiently than we do today so we can keep changing and innovating and moving much faster.

What did Ayn Rand understand about supersonic transport that we have forgotten? Not you haven't forgotten it, to be clear.

Supersonic is actually mentioned in Galt's speech once.

And that's 1958, right? 1957, I think. '57, yeah.

1957. Yeah, I mean, the passage is about, you know, moral conditions and, you know, motivation. And it basically says, you know, you can't have a socialist create a supersonic airplane. You need to have a capitalist with real motivation. That's not literally what it says, but it's effectively what it says.

The moon program, if I want to defend it, I would say, well, to get the satellite belt up, we needed those advances from the government. You needed capital put up front. It did inspire people. The internet in turn has relied heavily on the satellite belt. Would it really have been better if we hadn't had a moon program?

I mean, we don't get to see the counterfactual, right?

But it would have been more money spent on consumption, right? More popcorn, lower taxes.

I think we have to look for our inspiration, you know, from the areas that developed not like that and understand those histories. You know, like from 1903 with the, you know, the Wright brothers' first flight all the way through the introduction of the 707, we had entrepreneurially driven, commercially driven innovation in aviation. And every generation of new commercial airplane was faster than the one that came before it until we literally banned speed as, you know, a vector of innovation. And every generation brought air travel safer to more people too. Like, that was happening privately. Imagine if that had gone differently and the 707 had been a state-led project and DARPA had been involved and blah, blah, blah. Like, we'd be sitting here asking the same question about, like, jets, except none of us would fly on them.

Why isn't Boom first building a private plane before the attempt at a commercial airliner?

Effectively, that was banned because in 1973, when we put the supersonic ban over land in place, we banned effectively a supersonic private jet, because even on a high-end, you know, international range Gulfstream, 85% of business jet miles are over land. And if you can't fly faster than the 85% use case, there's no market. So we effectively, we banned minimum viable product. And our plan at Boom was to kind of do this kind of leap over what would normally be 1.0 and just start with 2.0 and do international flights where the regulatory issue was not an issue. And I didn't anticipate it would be as easy as it actually was to change that rule.

So are you thinking of changing course and doing both?

Uh, no, we're pretty far down path with the airliner. I hope someone else will do the private jet. If nobody does it, we'll do it eventually. I hope somebody else does.

Now I live on the East Coast of the US. For me, the worst flight is to go overnight to Reykjavik, Iceland, because it's too short. I'd actually rather fly to Istanbul. I'd even rather fly to South Africa, which is an easy flight for me for sleeping. As a customer, am I a problem for your business model?

No, because, because your overnight flight to Reykjavík turns into an actually quite nice daytime flight. If you want to go from the US to Europe, instead of, you know, having a too short night of sleep, if you could even sleep on an airplane, you leave in the morning, you get there in the afternoon, and it's a daytime flight with like great Wi-Fi. So I think what will happen in the first wave of supersonic is that the North Atlantic, which today is red eyes, turns into daytime flights instead, and you save an entire day. And if you look across the Pacific, you actually save 2 days on a round trip. Today, if I go to Tokyo for a Monday morning meeting, I have to leave midday Saturday. I get there end of day Sunday, go to hotel, I try to sleep, my alarm goes off the next morning, I go to my meeting, try not to sleep in my meeting, then I can fly back. By the way, we have a deal with Japan Airlines. I did this a bunch of times. It takes 3 calendar days, and you better not make any decisions the rest of the week 'cause the jet lag is like, really messed us up. And supersonic is totally different. You don't leave Saturday, you sleep Saturday night at home, leave Sunday morning, an 8:00 AM departure from SFO. Arrives at Tokyo 8 AM Monday morning Tokyo time. That's 6 hours later, so it's like Sunday afternoon to us, Monday morning in Tokyo, do a whole day of meetings, overnight flight back, sleep on the flight, you're back in town 24 hours after you left, so there's no jet lag because you never change time zones. It's much, much better.

Should there be windows on commercial flights for, you know, crossing the Pacific? What are they good for?

The only thing good about being in a tube at 60,000 feet is you can look out.

And, uh, there's nothing to see over the Pacific.

So it's actually gonna be great because at 60,000 feet you can tell whether the Earth is curved. So we're gonna do a flight of flat earthers that will be able to settle this.

How is the training for supersonic pilots different? Is it much harder, different at all?

It's a little scary to think, but there is such a thing as like a worst doctor and a worst lawyer. And a worst airline pilot. And we're designing an Overture such that, like, you don't need extraordinary skill to fly this thing. The Overture sim, which I've spent a bunch of time in, it's like going from a BlackBerry to an iPhone. Like, we've deleted all the knobs and switches and buttons, put hardware controls only for the things that are necessary. And there's a sort of— I think what will be seen as a breakthrough in safety, which is about the pilot-airplane interface. And the key thing is how the airplane and the pilot talk to each other in a crisis. It's very normal to get tunnel vision when alarms are going off, you know, things are not right. And the Overture has force feedback side sticks, so you get a whole extra sense of what the airplane is doing in addition to what you see and what you can hear. And I think this is going to be a big deal.

Can you imagine flying with only one pilot rather than two?

I think that's— no, I mean, well, I guess I'd do it personally. But in a smaller plane, I don't think it's a good idea for airliners. What happens when, you know, the one pilot has to pee?

Well, the computer.

Yeah, but then what happens if that's the moment something goes wrong? I don't think single pilot's actually a good idea, and I don't think it'll be a viable transition path to autonomy.

There's so much debate these days about the manufacturing infrastructure of the United States. Now, you work with it on a very, very regular basis. You rely on it. What's your general sense of where things stand and what we should do about whatever problems we might have?

We could probably talk for hours just about this. It's in pretty bad shape. And so aerospace and defense is like one of the only industries that kind of stayed in the US because it had to for national security reasons. And yet it's like really messed up. So like, here's a story. We start quoting out of the supply chain turbine blade for our supersonic engine. And it's a 3D-printed blade with like a little bit of post-processing on it. And the quote is, this is gonna be $1 million for one engine's worth of blades, and it's gonna take 6 months to get them. And I was like, 6 months? Like, what's going on for 6 months? And it's like, well, you have to wait your turn on the machine, and then the machine goes, and then, you know, and then that's in one factory, and then there's like a different factory in like a different state where one post-processing step happens, and then there's another factory in another state where another step happens. And so it takes 6 months and the part spends more time on a truck than on a machine. And I was like, this is insane. Like, what does the machine cost? $2 million. Okay. How long does it really take to do these process steps if you just do them back to back? 24 hours. Okay. Okay. How long does it take? How long does it take to get the machine? 2 weeks. It's in inventory.

And how do we fix that?

So what we did, the most insane innovation in aerospace manufacturing is to do all the process steps under one roof. And so you don't need any trucks. And so that's literally what we're doing. We're building a factory in Denver that's going to be raw materials in one side of the building and jet engines out the other side. The second-order effects of this are tremendous. Like, if you tell an engineer to design a turbine blade, and if they get it wrong, it takes 6 months to get another iteration, they like really wring their hands. They move very slowly. They're very conservative. They can't take any risk, right? But if you say, by the way, if you screw it up, you can get a new part in 24 hours, now there's a lot of design freedom. You can test more ideas. You can move much faster. You don't need analysis paralysis. So that's, that's what we're doing. As I was kind of going through this learning experience, I was like, why is it this way? It just seems stupid. And a meta-principle I've learned that I would suggest for everybody is when something seems stupid, ask yourself what would have to be the case for this actually to be smart. And I found by asking that question, occasionally things actually are dumb, but it's the minority case. Usually there's something you can reverse engineer about how the world works from asking that question. So why would it make sense to have one factory for one process step spread out all over the country. What would cause that?

Congress, right? Congress!

Yes, so this is a congressionally optimized supply chain because the way you get a defense program to become a program of record is you maximize the number of votes, so you put one process step in every congressional district. By the way, this will be very, very terrible if we actually go to war.

How do we fix that systemically? We're gonna have Congress no matter what, so you and Jennifer Pahlka get together and come up with what kind of plan?

Nobody wants to be the only one who cuts things in their home district, right? So I think— I also think about this in the shower. Maybe I shower too much. The whole Congress runs together on one platform and they all agree to be in on something, I think might actually be really important. It doesn't necessarily have to be partisan. Like, imagine if we had a bipartisan campaign for the next Congress and like everyone agreed on a few things that are like, "Hey, we're going to all be in this together. Maybe we're going to balance the budget, and maybe we're going to change defense procurement in some way that it fixes this effect." I don't think that's what the exact solution is, but I think a congressional team all in on the same agenda— because everybody could say, by the way, "Obviously we can't be porking every district," but so long as that's a district-by-district decision, it gets perpetuated. I think there needs to be a team or a theme, and I think this actually could be sold to the public, is at a certain level, it's obvious that it needs to be fixed and obviously good for America.

Now, while you've been a private pilot since you were an undergraduate, you were not an aerospace person per se. So what can you teach us about what you've learned from mastering technical subjects pretty quickly but also reliably?

I think we tend to get taught that we go to school and we study something, and then we kind of get a job and we go deeper in that. And then the way you really win is to become an expert. And by the way, by the time you become an expert, you're almost certainly useless because now you're like steeped in the status quo. And I think there is a massive advantage in switching domains and doing so quickly because it forces a focus on first principles. And I think so if you're passionate and if you have a good internal sense of when you're clear versus confused, my experience has been I could actually learn a whole lot pretty quickly. When I was in the first year of Boom, kind of trying to figure out whether this made any sense, I kept a confusion list, and whenever I didn't understand something, I would write it down. And my goal was to take one thing off the confusion list every week. What actually happened is that list grew without bound. But it was actually a really important skill because I learned how to tell in my own mind when I was clear and when I was confused. And I was confused a lot, but I could tell. And if something turned out it mattered, I could go study it and I could study it until I actually got my head around it. You know, my favorite interview question as I was hiring was, teach me something. So you didn't get hired in early boom unless you could teach me something.

How do you use LLMs in your current workflow?

Ooh, so many ways. And we're just starting to discover how powerful they really are. One thing LLMs are very, very good at is filling out regulatory paperwork. We can like smile and say, that's great. But actually the second order effects of this are really important. The way airplane certification worked historically is you'd have to go produce a very long, say, test plan for how you're going to prove that your airplane is safe in lightning. And this document might stretch on for like 100 pages. It has to have all the regulatory citations in it. And you have to hire a kind of engineer who is very happy spending 2 months writing a 100-page document for lightning strike protection. And then, by the way, if you ever want to change anything, nobody wants to change anything because it means another 2 months of work writing this document. Well, it turns out a small prompt and a RAG with all the federal regulatory guidance in it can produce this thing in like, you know, a few minutes. And then a pretty creative engineer can edit it and fix the hallucinations quickly. So this is what we're doing. And what it allows us to do is go from like large armies of sort of like, you know, turn the crank kind of talent to small teams of incredibly creative talent that are no longer change-averse because change is inexpensive. This is a big deal. Many, many things about what we're doing at Boom and how we're doing it are about reducing the cost of change. About reducing the cost and time required for iteration. And LLMs are a really potent weapon for this, along with a few other things like vertical integration and embedding software engineers in hardware teams.

What is it you like about the violinist Anne-Sophie Mutter?

She does a beautiful job playing some of my favorite classical music.

And what's that like? I love Brahms, the violin concerto, Beethoven violin concerto.

All good stuff. Don't listen to it enough.

Last two questions. First, this is October 2025. Again, there's plenty of you out there on YouTube, but if you would just give us an update in the last 6 to 9 months, what has changed for Boom and the prospects of supersonic commercial flight?

I think that it's blown wide open. It feels like the year that everything started to work. So in January, we broke the sound barrier. February, we did it again. That's when we announced the boomless cruise. That week, So it was February 10th, we announced Boomless Cruise. I flew to DC that night. By the time I landed, I had an invitation to the West Wing. Thursday of that week, the airplane model had made its way into the Oval Office. By the way, it's still there, like living rent-free in the Oval Office. And it was 115 days to repealing the ban on supersonic flight over land via executive order on June 6th. The barriers are all removed. The technology is there. The regulatory environment is there. It just throttles all the way forward, like, let's go. Like, there's lots of reason to be very optimistic.

Final question: What is the next thing you will think about in the shower?

I don't know. I think a lot about why problems that seem really obvious, that have in some ways solutions that seem really obvious, don't get solved. I think if we could figure that out, we could maybe unlock a lot of innovation. I don't know the answer.

Blake Shawl, thank you very much.

Thanks, Tyler.

Thanks for listening to Conversations with Tyler. You can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. If you like this podcast, please consider giving us a rating and leaving a review. This helps other listeners find the show. On Twitter, I'm @TylerCowen and the show is @CowenConvos. Until next time, please keep listening and learning.