Fundamentally this distinction (re: 'engineerism') seems to embody the difference between *technology* and *product* - the capacity to deliver some kind of value proposition versus the mechanism that actually undertakes the delivery. And I love the coinage. You definitely see it amongst engineers, but I also think you see it in the policy world, too - we should have e.g. state-run grocery stores, or should e.g. invade Saddam Hussein, not because it nets us some outcome that people want, but because *doing so intrinsically is de-ontologically good*.
All such approaches are, to me, fundamentally variations of the same kind of micro-utopianism: that the point of life is to conduct oneself in accordance with the Way — and opponents are, in their opposition, primarily apostates.
Indeed - when I managed engineers, I would always have to remind them to start presentations with a clear statement of the problem to be solved before they launch into a description of their new technology.
For me, the advent of flying cars would lead to a sudden uptick in dematerialization because it would be another great reason to NEVER leave the house.
Lane markings offer average drivers a great deal of assistance of in staying out of each other's way. Did Hall describe a credible system of lane markings in the sky?
How could Americans ever have flying cars when they can barely maintain their actual cars? Imagine how many crashes there would be if lots of people had flying cars? It's already a problem with General Aviation.
This is really a matter of trust in the technology. For driving or for air cars, a computer can keep track of where every vehicle is at every moment and make decisions faster and more reliably than people can. In this scenario lane marking isn't needed.
But you really have to trust technology to want to go there. As it is, I'm freaked out by how reliant we are on the internet. I half expect a global internet shutdown that lasts days or weeks due to cyber terrorism or AI attack or cyber warfare. It will be horrific.
Great piece. I've spent my life and career kinda at the meeting point between physics and engineering and have considered myself one or the other or both at various times, and I've come to have entirely too many opinions on the good and bad ways each group thinks.
One thing that always struck me as a little funny about the "flying car" obsession is it's very much a linear extrapolation of American car culture. I'm not exactly a car hating urbanist (in fact I can be something of a car nerd) but I think it's pretty clear we invested in cars and their infrastructure somewhat past the point of good returns, well beyond that of most developed countries, and a a consequence basically took everything and just spread it out more. We have the same number of people in bigger cities with more infrastructure, and everyone traveling the same amount of time to get to the same places further apart from each other, while every household spends tens of thousands of dollars yearly buying and maintaining cars.
The "flying car" idea is just a super sized version of this. You take everything and spread it out even more, moving around gets even more expensive (don't skip out on that flying car mechanic bill!) and ultimately you go to the same places in the same amount of time.
"Where are flying cars" is a cliche that should remind us that engineers are intelligent but narrow single-variable optimisers that shouldn't be in charge of societal decisions, like the MENSA episode of The Simpsons.
The rise of software and software engineering has super charged this, because now you have a whole generation of engineers whose visions are entirely detached from physical reality and its constraints.
We have top-3 shortest commute times in the OECD and top-3 largest living spaces so I think our infrastructure choices have turned out well!
You hit some issues when superstar city centers physically can't fit more car infrastructure (NY, LA, SF could all use more public transit which is to say they could use more dense housing that can support public transit use).
But at the country level we've created a much more comfortable life for ourselves than say, Koreans and the Japanese or Germans and the Dutch.
I totally agree with your point on software engineering detaching from physical reality.
This trend probably kicked off thirty years ago back in the young Bill Gates era. Since then, software engineering has ballooned into something completely disconnected from the real world. I’d go as far as saying the entire Silicon Valley is built on a giant bubble of narrow, isolated disciplines, where far too many brilliant minds spend most of their time optimizing utterly trivial things.
As for whether engineers should make societal decisions—it depends on the context. If we're talking about classic garage entrepreneurs risking their own dime, fine. But today’s playbook involves NASDAQ filings, massive VC rounds, and crowdfunding. It effectively drags ordinary people’s retirement funds into these geek adventures, which crosses the line.
When it comes to flying cars: if they still rely on burning gas or depend on local dealerships and neighborhood tire shops for maintenance, it just proves one thing—the technology is nowhere near ready. But since we’re talking about flying cars, we really need to reset our minds back to the Wright brothers' era and re-examine the problem from first principles with a much simpler lens.
Love the critique of engineerism. You're too kind to not call it mad-scientism. Solar is ignorable by these types because it's not sexy and doesn't require massive funding for "cool" construction projects. Tech like solar that's "easy" and scale linearly "doesn't make sense" to the people who want to build a Dyson sphere, which Elon Mush has shoved back into the news cycle again. 🙄
From slightly different perspectives, Becker came to many of the same conclusions in "More Everything Forever" to debunk Hall's book.
Personally, I'd trust me with a flying car or a jetpack, but not "most people" who seem to struggle overmuch with two dimensions…
Yes, I suspect the obession with nuclear is, at the main point, an obsession with scale and bigness, and nuclear plants are "big" by nature.
Probably there is also the reason that some people are pro-solar, but with fetishes like "in the middle of Arizona we could have a square of some Km2 full of solar panels to power all of North America" - an obsession with BIG (even if is probably better for the safet of network to have several power sources in several regions)
Oh boy, I almost passed out when I saw the term "Engineerism"—I confess I used to be a rather mediocre, naive, weird, and overly ambitious engineer myself (well, a software engineer, not a "real" engineer though).
That said, a flying car is obviously a cool idea and undoubtedly valuable. I’m not saying EV-packed streets are bad, nor do I think 140-character tweets or short-video-addicting smartphones are particularly sinful—even though in previous Noah threads, half of us readers went around praising and bashing each of them. So I guess it’s fine that it’s the flying car's turn to get roasted.
To put it frankly: flying cars didn't happen simply because we, as engineers, failed—not because of anything else.
Market demand is certainly an issue, but let’s not forget: from Edison and Tesla to the Wright brothers, from automotive culture to transistors, software, and the iPhone—relentless, obsessed creation that builds a great product, hands it to users, and says, "This is what you want" has always been an aggressively successful path forward. Compared to the non-engineering, purely business-calculated product playbook—interviewing existing users, reading market research, and only building what’s already asked for—isn't this engineering culture somewhat nobler? I still think it is.
In my view, the fact that thermonuclear fusion (anticipated for almost a century), small modular reactors, and flying cars haven't materialized can easily be blamed on the market—saying all the money flowed into trivial social network messaging app and buying clothes online. But at its core, this is a failure of engineering capability.
Perhaps it’s partly a financial issue of under-investment, but it's far more a problem of under-investing in intellectual capital. There simply aren't enough truly smart, capable, and willing STEM minds dedicated to these problems.
There are plenty of flying car prototypes and multiple technical pathways. If given careful and sustained thought, none of the safety concerns are truly unsolvable or impossible to bypass. Yet, no viable commercial product exists today. Is it really because society doesn't want it? Is it possible that the practitioners just weren't smart enough or didn't push hard enough? Or perhaps the brilliant minds (and I say "perhaps" because they never actually entered the field to try, allowing them to boast freely) were too busy picking up easy money in quantitative trading and crypto, leaving promising technical routes stranded?
If you ask me, even the most arrogant tech titans, when facing the flying car problem, should at least admit that the sheer difficulty of the engineering problem defeated them before they start making excuses or talking nonsense.
To sum it up: this remains fundamentally an engineering and technology problem first, and an economics problem second. That is where I disagree with Noah.
In a world of energy abundance, all of the alleged disadvantages of flying cars that allegedly make them undesirable are solvable technical problems. They could be reliable enough to not require extensive preflight checks; they could be made compact enough and safe enough to park in an ordinary driveway; they could be networked and coordinated well enough to eliminate the need for flight plans and air traffic controllers. These are expensive, not impossible.
Even at scale, a world with flying cars doesn't have to become an expanded suburbanized dystopia. It could become a world where we reclaim much of the urban land we've surrendered to surface cars.
All this sits in contrast with e.g. cold fusion, which isn't even an engineering problem at all at this stage; it's a science problem. We don't know if it's even possible to fuse atoms at low temperatures at a useful rate.
(Hot fusion, on the other hand, *is* an engineering problem, and it's being solved at scale right now. We may be near another inflection point in the energy curve which will make all these other problems look tractable again.)
The problem with hot fusion is that, since the 1970s, scientists and engineers have been saying that in order to get to fusion power they need a lot of money to build equipment to study plasmas, and nobody's given it to them, so of course progress has been slow.
The possible silver lining of our new Gilded Age is that a certain subset of the ultrawealthy seem to have taken a page from their 19th-century counterparts and started pouring substantial amounts of resources into medium/long-term R&D.
This isn't the ideal funding model from a societal perspective, but from the perspective of scientists and engineers in the R&D phase, private patronage is in many respects *better* than public funding because there's more freedom to take risks and make mistakes.
Fusion is one of the beneficiaries of this kind of investment, and the result is that we're starting to see real progress after decades of near-stagnation.
(Elon Musk is a loathsome human being and I refuse to work for any of his companies, but I do have to reluctantly give him credit for exporting "move fast and break things" to the aerospace industry, displacing both the extractive legacy of Jack Welch and the bureaucratic paralysis of post-space-race NASA. Fusion cross-pollinates so heavily with aerospace because of the thermal/vacuum/plasma physics/radiation overlap that it can almost be treated as an extension of the same industry.)
The last few years, the US government has budgeted $1 billion per year for fusion research. That’s not peanuts.
The reality is, fusion power has been 20 years away since the 1950s because it’s a hard problem. I means genuinely hard problem, as in there may not be a practical solution to it.
It's also nowhere near the total investment in fusion over the last few years. The total sum, public and private, is now about 4.6 billion/year.
And, unsurprisingly, the timeline has gotten correspondingly shorter: Helion claims they will deliver power to the grid by 2028, and I think they're going to hit that target in a technical sense, although I'm not expecting a meaningful economic impact until 2030-2035.
Thanks, Evan! Talking about this brings back that raw optimism and excitement of a STEM-dreamer—the exact feeling you get in a classroom when you're lucky enough to have an extraordinary high school physics teacher.
I'm thinking this energy is somehow different from the dominant mood in today's Silicon Valley. It feels kinder, perhaps?
So what am I really trying to say here? On one hand, I agree with Noah's critique as an economist regarding today's somewhat chaotic and over-leveraged tech culture. But on the other hand, I can't help but feel: let engineers go back to being optimistic STEM-dreamers. Let's inspire more brilliant minds and souls to tackle concrete, physical problems.
Ultimately, this pivots the whole debate into a question of STEM education and intellectual capital allocation—bringing us right back to the same round table with economists and political scientists once again.
I work in new-space, where we're allowed to dream a little bit, especially at startups and in R&D. It's a breath of fresh air. There's nothing that kills creativity faster than worrying about short-term profitability.
(At the executive level, everything has to be dressed up in promises of long-term profit, which can give the industry a bit of a grifty feel. But (with the exception of the SpaceX IPO situation) everyone funding it is wealthy and has full access to the entire corpus of human knowledge, so if they actually think they're going to make money from funding our science experiments, I'm not going to argue. I actually think a lot of it is just modern-day patronage, dressed up in the language of capitalism to save face.)
Europe might be a better place for these types of middle distance trips. I live on the Kent coast so 200 - 300 miles puts a load of cities in the UK and W. Europe within 1-2 hours according to these figures which by train or car would take a long time and which there would be no point flying to.
Newcastle - which feels like forever away - would take 6 plus hours by car and 5-7 hours by train. By flying car: 2 hours. Paris would take 4-5 hours in total by local train and Eurostar but less than an hour by flying car. And there would be no risk of traffic jams or train delays. No having to get used to driving on the other side of the road and no having to drag your luggage around and make connections etc. Just get in the car outside your house and land outside or on the roof of your hotel.
And the views would be spectacular: not confined to ground level where you can't see much at all or so high up you can't see anything until you start your descent for landing (during which the views can be fantastic at the sort of heights at which I'm hoping cars would fly).
Not sure you'd need to own one. You'd hire them for the trip, like a taxi. Given the cost of taking the train, they could be quite competitive on price. And they would have to be far more expensive than the train to outweigh the advantages.
Something would need to be organised for passport control and to address the risk of smuggling. But I'm sure something could be arranged.
It would work if only a few people had them, but landing on a hotel roof is questionable unless the flying cars just drop you off and park themselves somewhere, because the roof area needed for parking for even a small fraction of the rooms in a hotel would be enormous. And the other downside is arriving late or going out at night wouldn’t be allowed, since one of the biggest drawbacks of VTOL type vehicles is the very loud noise they make when landing.
True, but if most of them were taxis, they could just take off again after letting the passengers off and fly off on the next job or to a parking and maintenance area outside town. But yes, noise and a sky full of flying objects could be a big barrier.
It would certainly take some getting used to. From where I live you can see France across the Channel so on a reasonably clear day you'd always be able to see land from your flying taxi, which would provide some reassurance. But on trips across wider bodies of water despair or panic might set in at some point!
What is the source of the assumption (in the GPT estimate) that flying cars require "20 minutes total for takeoff/landing/preflight"? That seems to be what's making them unattractive for short trips. If your flying car (or, I suspect, your flying Waymo) can take you on a 25-mile cross-town journey in 10 minutes instead of 25 minutes (or more realistically, instead of 45 minutes), suddenly it becomes much more attractive.
Yeah, maybe he’s assuming they’re nuclear powered ( a fun section of Hall’s book talks about tiny nuclear batteries for portable devices we could have had, but don’t), but in the comparison table Noah doesn’t add in recharging stops for 50 miles or above which would be needed for battery powered flying cars. You probably also need to account for pit stops, unless the flying cars run on heat generated from on board composting toilets. And nobody knows the traffic jam (dirty bomb like?) that would occur after a nuclear powered flying car fender bender or road rage incident.
Hilarious! Buzen is totally right, and you both are right.
Every time the topic of flying cars comes up (and by the way, I actually love the idea of flying cars), my brain automatically resets to the Wright brothers' era, while somehow simultaneously injecting extreme, Apollo-era tech-optimism.
So yeah, in my head, flying cars are obviously powered by small modular reactors, with nuclear fuel pellets no bigger than watch batteries. May God forgive my naivety!
I think they were very afraid of nuclear energy after 3 Mile Island, "The China Syndrome" and Chernobyl, and no one in a position of authority or the press was able to see the forest for the trees and publicize how many deaths per year were attributable to fossil fuel emissions that could have been saved by nuclear power.
Well yeah there was no person in authority that everyone trusted. Do we ever have that? It was clearly reported that coal every year caused many deaths from mining and from lung disease. Orders of magnitude more per kilowatt hour produced compared to nuclear.
It was reported. People in power knew these facts. The issue is one of trust.
Here is the question to ask: During those decades were the people in power trustworthy? Or were the people broadly right to distrust?
Regardless of the allure of the virtual world or the short-term economics, we should continue to do hard things in the physical world. You never know when you will need a large space-based presense to stop an asteroid, or an advanced fusion fleet to provide energy after a super-volcano blackout.
I’m grappling with your seeming disdain for nuclear energy. I don’t know the area required to electrify Philadelphia, but I imagine it would be very large, and you’d need to cover it with solar cells. In addition, you’d have to have quite the battery farm to store enough energy to power factories, a subway, and students' computers at Villanova.
Supplying Philadelphia's energy needs with solar seems technologically impossible in January, with 3 or 4 days of cloud cover when the sun is at its lowest. Whereas nuclear power could do so quite easily.
Noah, if you read the Axios Defense newsletter yesterday, you would have read a blurb about a microreactor being built at Ft Bragg and an SMR being built at Ft Campbell. A Rolls-Royce unit is building SMRs in Sweden. GE Vernova and Hitachi are building an SMR somewhere in the Midwest.
I don’t know if solar could power a hyperscaler, or how big an area would be needed, but it would seem that doing so would really enrage those who believe they are taking farmland to build a data center.
I am an all-of-the-above person, and I am surprised to see you are not.
China is an all-of-the-above kind of country. They ARE building nuclear. They're very good at building nuclear. They're just building much much much much much more solar than nuclear. Which reflects the cost differences between the two.
I'm just saying, fission power is a stagnant technology while solar+batteries is revolutionary. That doesn't mean I want to ban nuclear or something.
I guess we will see where SMRs and microreactors take us. Solar has a place, for sure. I have, however, seen more advances in nuclear technology than in solar farms.
I am still waiting for a battery farm that could run Minneapolis in January. Does that capacity exist? Is it coming? What is the cost? What are the dangers? I claim total ignorance on large-scale battery storage for a medium city, let alone LA.
I don't think Noah has disdain for fission energy. He acknowledges there are cases where it still makes sense. He makes a case that most of the time solar makes more sense.
Your example of Pennsylvania breaks down on examination in comparison with China. China is the world leader in long distance power transmission. That is how you get the energy from empty spaces with good solar potential to Pennsylvania.
Sure, it costs a lot. Sure it has its inefficiencies, but so does nuclear. His point is that China doesn't have to bother with environmentalists yet it is growing solar faster than it is growing nuclear.
While mentioning nuclear in China (62 GW) and solar (1200 GW), Noah doesn’t mention the biggest source coal (1270 GW) which is still growing at 17% a year. Those numbers are for capacity, and he doesn’t really mention how many batteries are being deployed, which is important to know for baseline power. To compare the solar capacity to nuclear capacity for baseline power, divide solar by four because you need to use more than half of your solar capacity just to charge batteries for nights, and then double it again to deal with winter and overcast periods.
That’s progress, but the article says solar and wind was only 24%. And it says the Chinese have the ambitious goal of increasing that to 30% in the next four years. Without storage, the curve for solar electricity generation flattens out at 50%, clear as day and night.
I complete agree that we shouldn't dismiss nuclear, but using it directly as a dedicated power source just for AI data centers feels like a misallocation of its unique capabilities.
Energy sources should be complementary, not competitive. There's no need to favor one while neglecting another. A practical, consensus-driven architecture looks something like this:
Primary Supply: Dramatically scale up solar, wind, and tidal power (we are nowhere near their true scaling potential yet).
Grid Backbone: Pair renewables with long-distance Ultra-High Voltage DC (UHVDC) transmission and a smart grid.
Peak Shaving & Storage: Deploy diverse, heavy-duty storage nodes—battery banks, pumped-storage hydro, and molten-salt thermal storage—to manage intermittency and smooth out peak loads.
Baseload & Voltage Stabilization: Use nuclear power as the strategic reserve and main anchor to guarantee grid stability and voltage control.
Fossil fuels still have a role to play during this evolutionary transition, though their overall share will gradually shrink—and that's a completely feasible path forward.
As for solar farms taking over valuable farmland, that’s just poor planning. The ideal home for utility-scale solar is in arid deserts and badlands. There’s a reason those areas aren't farmlands in the first place: too much sun, too little rain.
Your transmission lines stretching from Nevada to Philadelphia would certainly face some NIMBY issues. Transmission lines in Tornado Ally get torn up and freeze. Like I said, all of the above, but even that strategy has drawbacks.
The flying car and the car-boat are cut from the same cloth—solutions desperately searching for a problem. Engineerism may be real, but that's no reason to hate on the dreamers. After all, we'll need their wild "what ifs" to navigate a future pregnant with unknowns..
My neighbor back in the 1970s actually had a partially submersible car, which supposedly worked as a boat as well as a car. It looked like a Chevy Corvair with a small propeller on the back. I never saw it leave his driveway.
Sounds like that neighbor had a lot of fun! I’m guessing he loved both his car-boat and his dream.
Hey, it just hit me: as long as "Engineerism" isn't running amok—grabbing massive capital and scattering risks everywhere—and stays confined to a small scale, it's actually pretty charming, isn't it?
Yes planes are flying buses, and helicopters are flying cars! Or at least flying taxis. So they're expensive but there's enough use cases for them to support an industry even at that price: ambulances, pest control, ranching, rich commuters etc.
If a flying car means no pilot and cheaper then they could replace helicopters and then expand from those uses
Maybe they make helicopter trips cheaper, easier, safer, something like that. Maybe there's a tripling of the market relative to latent demand for helicopters, but it's not going to replace cars. Cars are cars and they're pretty good at it.
I have a business contact friend that booked a helicopter to get from Barcelona near the airport up to the Spanish Pyrenees on vacation. In the nearish future it'll be a flying car, i.e. helicopter 2.0 (basically a big drone with seats).
The business model for the flying car (or flying Waymo) is the weekly commute.
You have that large mansion in the Hamptons and your whole family just loves it? You are even allowed to work from home office one or two days a week. That means from Monday to Wednesday you buckle up for your weekly grind at the office (where you have miniscule studio apartment or even a capsule hotel), then it's back to your real home for family time and home office.
In my mind, the big advantage in flying cars is , not medium range travel but reducing congestion by taking advantage of the large 3 dimensional space over our heads. The limitation here is traffic coordination, which could be solved using computer networks to pilot the vehicles
You're moving the congestion from the road to the landing pad. You're going to be rate limited on how many flying cars you can land in a given area, and it's likely going to be a lot lower than private cars, let alone trains/buses/bicycles.
That already exists, but you can’t fly it yourself. Blade offers private air commutes between NYC and East Hampton, but it costs over $500 one way. If they used drones instead of pilots and the FAA allowed that, then it may become cheaper.
This book has been on my list for a while. Now I don’t need to read it.
Engineerism is a good diagnosis. The safety issue with flying cars has always been obvious to me. When it comes to personalized flying technology, drone delivery on demand has always seemed a lot more useful, and we indeed have it. In general, our need to move *ourselves* has decreased with virtualization, but we still need to move goods.
The discussion of energy usage is a mild criticism of Vaclav Smil’s work, which tends to reduce human progress to energy usage.
The engineerism critique is also a veiled critique of Musk. Data centers in space, the Boring Company, …
Fundamentally this distinction (re: 'engineerism') seems to embody the difference between *technology* and *product* - the capacity to deliver some kind of value proposition versus the mechanism that actually undertakes the delivery. And I love the coinage. You definitely see it amongst engineers, but I also think you see it in the policy world, too - we should have e.g. state-run grocery stores, or should e.g. invade Saddam Hussein, not because it nets us some outcome that people want, but because *doing so intrinsically is de-ontologically good*.
All such approaches are, to me, fundamentally variations of the same kind of micro-utopianism: that the point of life is to conduct oneself in accordance with the Way — and opponents are, in their opposition, primarily apostates.
Indeed - when I managed engineers, I would always have to remind them to start presentations with a clear statement of the problem to be solved before they launch into a description of their new technology.
Totally. “But don’t people really care about the FEATURES?!”
For me, the advent of flying cars would lead to a sudden uptick in dematerialization because it would be another great reason to NEVER leave the house.
Lane markings offer average drivers a great deal of assistance of in staying out of each other's way. Did Hall describe a credible system of lane markings in the sky?
How could Americans ever have flying cars when they can barely maintain their actual cars? Imagine how many crashes there would be if lots of people had flying cars? It's already a problem with General Aviation.
This is really a matter of trust in the technology. For driving or for air cars, a computer can keep track of where every vehicle is at every moment and make decisions faster and more reliably than people can. In this scenario lane marking isn't needed.
But you really have to trust technology to want to go there. As it is, I'm freaked out by how reliant we are on the internet. I half expect a global internet shutdown that lasts days or weeks due to cyber terrorism or AI attack or cyber warfare. It will be horrific.
Lane markings are the thing that allows you leave the house? You must live in some warm
Climate, lane marking get wiped out by snow.
In snow, lane markings are quickly replaced by tire tracks,
Only some of the time. When it snows hard enough, or snow is blowing, tracks quickly disappear.
Amazingly, people still get to work anyway.
Great piece. I've spent my life and career kinda at the meeting point between physics and engineering and have considered myself one or the other or both at various times, and I've come to have entirely too many opinions on the good and bad ways each group thinks.
One thing that always struck me as a little funny about the "flying car" obsession is it's very much a linear extrapolation of American car culture. I'm not exactly a car hating urbanist (in fact I can be something of a car nerd) but I think it's pretty clear we invested in cars and their infrastructure somewhat past the point of good returns, well beyond that of most developed countries, and a a consequence basically took everything and just spread it out more. We have the same number of people in bigger cities with more infrastructure, and everyone traveling the same amount of time to get to the same places further apart from each other, while every household spends tens of thousands of dollars yearly buying and maintaining cars.
The "flying car" idea is just a super sized version of this. You take everything and spread it out even more, moving around gets even more expensive (don't skip out on that flying car mechanic bill!) and ultimately you go to the same places in the same amount of time.
"Where are flying cars" is a cliche that should remind us that engineers are intelligent but narrow single-variable optimisers that shouldn't be in charge of societal decisions, like the MENSA episode of The Simpsons.
The rise of software and software engineering has super charged this, because now you have a whole generation of engineers whose visions are entirely detached from physical reality and its constraints.
We have top-3 shortest commute times in the OECD and top-3 largest living spaces so I think our infrastructure choices have turned out well!
You hit some issues when superstar city centers physically can't fit more car infrastructure (NY, LA, SF could all use more public transit which is to say they could use more dense housing that can support public transit use).
But at the country level we've created a much more comfortable life for ourselves than say, Koreans and the Japanese or Germans and the Dutch.
I totally agree with your point on software engineering detaching from physical reality.
This trend probably kicked off thirty years ago back in the young Bill Gates era. Since then, software engineering has ballooned into something completely disconnected from the real world. I’d go as far as saying the entire Silicon Valley is built on a giant bubble of narrow, isolated disciplines, where far too many brilliant minds spend most of their time optimizing utterly trivial things.
As for whether engineers should make societal decisions—it depends on the context. If we're talking about classic garage entrepreneurs risking their own dime, fine. But today’s playbook involves NASDAQ filings, massive VC rounds, and crowdfunding. It effectively drags ordinary people’s retirement funds into these geek adventures, which crosses the line.
When it comes to flying cars: if they still rely on burning gas or depend on local dealerships and neighborhood tire shops for maintenance, it just proves one thing—the technology is nowhere near ready. But since we’re talking about flying cars, we really need to reset our minds back to the Wright brothers' era and re-examine the problem from first principles with a much simpler lens.
Engineering is always about making a set of trade-offs. Engineers are optimizers, but not single-variable ones.
Love the critique of engineerism. You're too kind to not call it mad-scientism. Solar is ignorable by these types because it's not sexy and doesn't require massive funding for "cool" construction projects. Tech like solar that's "easy" and scale linearly "doesn't make sense" to the people who want to build a Dyson sphere, which Elon Mush has shoved back into the news cycle again. 🙄
From slightly different perspectives, Becker came to many of the same conclusions in "More Everything Forever" to debunk Hall's book.
Personally, I'd trust me with a flying car or a jetpack, but not "most people" who seem to struggle overmuch with two dimensions…
Yes, I suspect the obession with nuclear is, at the main point, an obsession with scale and bigness, and nuclear plants are "big" by nature.
Probably there is also the reason that some people are pro-solar, but with fetishes like "in the middle of Arizona we could have a square of some Km2 full of solar panels to power all of North America" - an obsession with BIG (even if is probably better for the safet of network to have several power sources in several regions)
Oh boy, I almost passed out when I saw the term "Engineerism"—I confess I used to be a rather mediocre, naive, weird, and overly ambitious engineer myself (well, a software engineer, not a "real" engineer though).
That said, a flying car is obviously a cool idea and undoubtedly valuable. I’m not saying EV-packed streets are bad, nor do I think 140-character tweets or short-video-addicting smartphones are particularly sinful—even though in previous Noah threads, half of us readers went around praising and bashing each of them. So I guess it’s fine that it’s the flying car's turn to get roasted.
To put it frankly: flying cars didn't happen simply because we, as engineers, failed—not because of anything else.
Market demand is certainly an issue, but let’s not forget: from Edison and Tesla to the Wright brothers, from automotive culture to transistors, software, and the iPhone—relentless, obsessed creation that builds a great product, hands it to users, and says, "This is what you want" has always been an aggressively successful path forward. Compared to the non-engineering, purely business-calculated product playbook—interviewing existing users, reading market research, and only building what’s already asked for—isn't this engineering culture somewhat nobler? I still think it is.
In my view, the fact that thermonuclear fusion (anticipated for almost a century), small modular reactors, and flying cars haven't materialized can easily be blamed on the market—saying all the money flowed into trivial social network messaging app and buying clothes online. But at its core, this is a failure of engineering capability.
Perhaps it’s partly a financial issue of under-investment, but it's far more a problem of under-investing in intellectual capital. There simply aren't enough truly smart, capable, and willing STEM minds dedicated to these problems.
There are plenty of flying car prototypes and multiple technical pathways. If given careful and sustained thought, none of the safety concerns are truly unsolvable or impossible to bypass. Yet, no viable commercial product exists today. Is it really because society doesn't want it? Is it possible that the practitioners just weren't smart enough or didn't push hard enough? Or perhaps the brilliant minds (and I say "perhaps" because they never actually entered the field to try, allowing them to boast freely) were too busy picking up easy money in quantitative trading and crypto, leaving promising technical routes stranded?
If you ask me, even the most arrogant tech titans, when facing the flying car problem, should at least admit that the sheer difficulty of the engineering problem defeated them before they start making excuses or talking nonsense.
To sum it up: this remains fundamentally an engineering and technology problem first, and an economics problem second. That is where I disagree with Noah.
Agreed, as another (not software) engineer.
In a world of energy abundance, all of the alleged disadvantages of flying cars that allegedly make them undesirable are solvable technical problems. They could be reliable enough to not require extensive preflight checks; they could be made compact enough and safe enough to park in an ordinary driveway; they could be networked and coordinated well enough to eliminate the need for flight plans and air traffic controllers. These are expensive, not impossible.
Even at scale, a world with flying cars doesn't have to become an expanded suburbanized dystopia. It could become a world where we reclaim much of the urban land we've surrendered to surface cars.
All this sits in contrast with e.g. cold fusion, which isn't even an engineering problem at all at this stage; it's a science problem. We don't know if it's even possible to fuse atoms at low temperatures at a useful rate.
(Hot fusion, on the other hand, *is* an engineering problem, and it's being solved at scale right now. We may be near another inflection point in the energy curve which will make all these other problems look tractable again.)
The problem with hot fusion is that, since the 1970s, scientists and engineers have been saying that in order to get to fusion power they need a lot of money to build equipment to study plasmas, and nobody's given it to them, so of course progress has been slow.
https://www.astralcodexten.com/p/your-book-review-the-future-of-fusion
Right now, the US government spends more money on peanut subsidies than fusion research. :P
Absolutely true.
The possible silver lining of our new Gilded Age is that a certain subset of the ultrawealthy seem to have taken a page from their 19th-century counterparts and started pouring substantial amounts of resources into medium/long-term R&D.
This isn't the ideal funding model from a societal perspective, but from the perspective of scientists and engineers in the R&D phase, private patronage is in many respects *better* than public funding because there's more freedom to take risks and make mistakes.
Fusion is one of the beneficiaries of this kind of investment, and the result is that we're starting to see real progress after decades of near-stagnation.
(Elon Musk is a loathsome human being and I refuse to work for any of his companies, but I do have to reluctantly give him credit for exporting "move fast and break things" to the aerospace industry, displacing both the extractive legacy of Jack Welch and the bureaucratic paralysis of post-space-race NASA. Fusion cross-pollinates so heavily with aerospace because of the thermal/vacuum/plasma physics/radiation overlap that it can almost be treated as an extension of the same industry.)
The last few years, the US government has budgeted $1 billion per year for fusion research. That’s not peanuts.
The reality is, fusion power has been 20 years away since the 1950s because it’s a hard problem. I means genuinely hard problem, as in there may not be a practical solution to it.
It's not nothing, but it's nowhere near the amount needed to make progress. https://i.imgur.com/3vYLQmm.png
It's also nowhere near the total investment in fusion over the last few years. The total sum, public and private, is now about 4.6 billion/year.
And, unsurprisingly, the timeline has gotten correspondingly shorter: Helion claims they will deliver power to the grid by 2028, and I think they're going to hit that target in a technical sense, although I'm not expecting a meaningful economic impact until 2030-2035.
Thanks, Evan! Talking about this brings back that raw optimism and excitement of a STEM-dreamer—the exact feeling you get in a classroom when you're lucky enough to have an extraordinary high school physics teacher.
I'm thinking this energy is somehow different from the dominant mood in today's Silicon Valley. It feels kinder, perhaps?
So what am I really trying to say here? On one hand, I agree with Noah's critique as an economist regarding today's somewhat chaotic and over-leveraged tech culture. But on the other hand, I can't help but feel: let engineers go back to being optimistic STEM-dreamers. Let's inspire more brilliant minds and souls to tackle concrete, physical problems.
Ultimately, this pivots the whole debate into a question of STEM education and intellectual capital allocation—bringing us right back to the same round table with economists and political scientists once again.
Absolutely!
I work in new-space, where we're allowed to dream a little bit, especially at startups and in R&D. It's a breath of fresh air. There's nothing that kills creativity faster than worrying about short-term profitability.
(At the executive level, everything has to be dressed up in promises of long-term profit, which can give the industry a bit of a grifty feel. But (with the exception of the SpaceX IPO situation) everyone funding it is wealthy and has full access to the entire corpus of human knowledge, so if they actually think they're going to make money from funding our science experiments, I'm not going to argue. I actually think a lot of it is just modern-day patronage, dressed up in the language of capitalism to save face.)
Europe might be a better place for these types of middle distance trips. I live on the Kent coast so 200 - 300 miles puts a load of cities in the UK and W. Europe within 1-2 hours according to these figures which by train or car would take a long time and which there would be no point flying to.
Newcastle - which feels like forever away - would take 6 plus hours by car and 5-7 hours by train. By flying car: 2 hours. Paris would take 4-5 hours in total by local train and Eurostar but less than an hour by flying car. And there would be no risk of traffic jams or train delays. No having to get used to driving on the other side of the road and no having to drag your luggage around and make connections etc. Just get in the car outside your house and land outside or on the roof of your hotel.
And the views would be spectacular: not confined to ground level where you can't see much at all or so high up you can't see anything until you start your descent for landing (during which the views can be fantastic at the sort of heights at which I'm hoping cars would fly).
Not sure you'd need to own one. You'd hire them for the trip, like a taxi. Given the cost of taking the train, they could be quite competitive on price. And they would have to be far more expensive than the train to outweigh the advantages.
Something would need to be organised for passport control and to address the risk of smuggling. But I'm sure something could be arranged.
I'm quite taken with the idea!
It would work if only a few people had them, but landing on a hotel roof is questionable unless the flying cars just drop you off and park themselves somewhere, because the roof area needed for parking for even a small fraction of the rooms in a hotel would be enormous. And the other downside is arriving late or going out at night wouldn’t be allowed, since one of the biggest drawbacks of VTOL type vehicles is the very loud noise they make when landing.
True, but if most of them were taxis, they could just take off again after letting the passengers off and fly off on the next job or to a parking and maintenance area outside town. But yes, noise and a sky full of flying objects could be a big barrier.
Imagining going across an open body of water in one of these....
It would certainly take some getting used to. From where I live you can see France across the Channel so on a reasonably clear day you'd always be able to see land from your flying taxi, which would provide some reassurance. But on trips across wider bodies of water despair or panic might set in at some point!
What is the source of the assumption (in the GPT estimate) that flying cars require "20 minutes total for takeoff/landing/preflight"? That seems to be what's making them unattractive for short trips. If your flying car (or, I suspect, your flying Waymo) can take you on a 25-mile cross-town journey in 10 minutes instead of 25 minutes (or more realistically, instead of 45 minutes), suddenly it becomes much more attractive.
Yeah, maybe he’s assuming they’re nuclear powered ( a fun section of Hall’s book talks about tiny nuclear batteries for portable devices we could have had, but don’t), but in the comparison table Noah doesn’t add in recharging stops for 50 miles or above which would be needed for battery powered flying cars. You probably also need to account for pit stops, unless the flying cars run on heat generated from on board composting toilets. And nobody knows the traffic jam (dirty bomb like?) that would occur after a nuclear powered flying car fender bender or road rage incident.
Hilarious! Buzen is totally right, and you both are right.
Every time the topic of flying cars comes up (and by the way, I actually love the idea of flying cars), my brain automatically resets to the Wright brothers' era, while somehow simultaneously injecting extreme, Apollo-era tech-optimism.
So yeah, in my head, flying cars are obviously powered by small modular reactors, with nuclear fuel pellets no bigger than watch batteries. May God forgive my naivety!
"Americans aren’t afraid of energy,"
I think they were very afraid of nuclear energy after 3 Mile Island, "The China Syndrome" and Chernobyl, and no one in a position of authority or the press was able to see the forest for the trees and publicize how many deaths per year were attributable to fossil fuel emissions that could have been saved by nuclear power.
Well yeah there was no person in authority that everyone trusted. Do we ever have that? It was clearly reported that coal every year caused many deaths from mining and from lung disease. Orders of magnitude more per kilowatt hour produced compared to nuclear.
It was reported. People in power knew these facts. The issue is one of trust.
Here is the question to ask: During those decades were the people in power trustworthy? Or were the people broadly right to distrust?
Forget the people in power - what about the environmental movement? All those deaths? Global warming? Where were they?
You should read "The Man Who Sold the Moon" if only because D.D.Harriman is who Elon Musk thinks he is.
Good piece, Noah. Lots of deeper musings.
Regardless of the allure of the virtual world or the short-term economics, we should continue to do hard things in the physical world. You never know when you will need a large space-based presense to stop an asteroid, or an advanced fusion fleet to provide energy after a super-volcano blackout.
I’m grappling with your seeming disdain for nuclear energy. I don’t know the area required to electrify Philadelphia, but I imagine it would be very large, and you’d need to cover it with solar cells. In addition, you’d have to have quite the battery farm to store enough energy to power factories, a subway, and students' computers at Villanova.
Supplying Philadelphia's energy needs with solar seems technologically impossible in January, with 3 or 4 days of cloud cover when the sun is at its lowest. Whereas nuclear power could do so quite easily.
Noah, if you read the Axios Defense newsletter yesterday, you would have read a blurb about a microreactor being built at Ft Bragg and an SMR being built at Ft Campbell. A Rolls-Royce unit is building SMRs in Sweden. GE Vernova and Hitachi are building an SMR somewhere in the Midwest.
I don’t know if solar could power a hyperscaler, or how big an area would be needed, but it would seem that doing so would really enrage those who believe they are taking farmland to build a data center.
I am an all-of-the-above person, and I am surprised to see you are not.
Of course I'm an all-of-the-above kind of person.
China is an all-of-the-above kind of country. They ARE building nuclear. They're very good at building nuclear. They're just building much much much much much more solar than nuclear. Which reflects the cost differences between the two.
I'm just saying, fission power is a stagnant technology while solar+batteries is revolutionary. That doesn't mean I want to ban nuclear or something.
I guess we will see where SMRs and microreactors take us. Solar has a place, for sure. I have, however, seen more advances in nuclear technology than in solar farms.
I am still waiting for a battery farm that could run Minneapolis in January. Does that capacity exist? Is it coming? What is the cost? What are the dangers? I claim total ignorance on large-scale battery storage for a medium city, let alone LA.
Yep, immaculate timing! I literally just typed out a whole mini-essay arguing for the exact same "all-of-the-above" approach right here: https://www.noahpinion.blog/p/why-you-wont-get-a-flying-car/comment/329135887
I don't think Noah has disdain for fission energy. He acknowledges there are cases where it still makes sense. He makes a case that most of the time solar makes more sense.
Your example of Pennsylvania breaks down on examination in comparison with China. China is the world leader in long distance power transmission. That is how you get the energy from empty spaces with good solar potential to Pennsylvania.
Sure, it costs a lot. Sure it has its inefficiencies, but so does nuclear. His point is that China doesn't have to bother with environmentalists yet it is growing solar faster than it is growing nuclear.
While mentioning nuclear in China (62 GW) and solar (1200 GW), Noah doesn’t mention the biggest source coal (1270 GW) which is still growing at 17% a year. Those numbers are for capacity, and he doesn’t really mention how many batteries are being deployed, which is important to know for baseline power. To compare the solar capacity to nuclear capacity for baseline power, divide solar by four because you need to use more than half of your solar capacity just to charge batteries for nights, and then double it again to deal with winter and overcast periods.
Coal just dropped below 50% in China!
https://www.nbcnews.com/world/asia/china-generates-less-half-electricity-coal-first-time-rcna589972
That’s progress, but the article says solar and wind was only 24%. And it says the Chinese have the ambitious goal of increasing that to 30% in the next four years. Without storage, the curve for solar electricity generation flattens out at 50%, clear as day and night.
I complete agree that we shouldn't dismiss nuclear, but using it directly as a dedicated power source just for AI data centers feels like a misallocation of its unique capabilities.
Energy sources should be complementary, not competitive. There's no need to favor one while neglecting another. A practical, consensus-driven architecture looks something like this:
Primary Supply: Dramatically scale up solar, wind, and tidal power (we are nowhere near their true scaling potential yet).
Grid Backbone: Pair renewables with long-distance Ultra-High Voltage DC (UHVDC) transmission and a smart grid.
Peak Shaving & Storage: Deploy diverse, heavy-duty storage nodes—battery banks, pumped-storage hydro, and molten-salt thermal storage—to manage intermittency and smooth out peak loads.
Baseload & Voltage Stabilization: Use nuclear power as the strategic reserve and main anchor to guarantee grid stability and voltage control.
Fossil fuels still have a role to play during this evolutionary transition, though their overall share will gradually shrink—and that's a completely feasible path forward.
As for solar farms taking over valuable farmland, that’s just poor planning. The ideal home for utility-scale solar is in arid deserts and badlands. There’s a reason those areas aren't farmlands in the first place: too much sun, too little rain.
Your transmission lines stretching from Nevada to Philadelphia would certainly face some NIMBY issues. Transmission lines in Tornado Ally get torn up and freeze. Like I said, all of the above, but even that strategy has drawbacks.
The flying car and the car-boat are cut from the same cloth—solutions desperately searching for a problem. Engineerism may be real, but that's no reason to hate on the dreamers. After all, we'll need their wild "what ifs" to navigate a future pregnant with unknowns..
My neighbor back in the 1970s actually had a partially submersible car, which supposedly worked as a boat as well as a car. It looked like a Chevy Corvair with a small propeller on the back. I never saw it leave his driveway.
Sounds like that neighbor had a lot of fun! I’m guessing he loved both his car-boat and his dream.
Hey, it just hit me: as long as "Engineerism" isn't running amok—grabbing massive capital and scattering risks everywhere—and stays confined to a small scale, it's actually pretty charming, isn't it?
Yes planes are flying buses, and helicopters are flying cars! Or at least flying taxis. So they're expensive but there's enough use cases for them to support an industry even at that price: ambulances, pest control, ranching, rich commuters etc.
If a flying car means no pilot and cheaper then they could replace helicopters and then expand from those uses
This is it, flying cars = Helicopter 2.0
Maybe they make helicopter trips cheaper, easier, safer, something like that. Maybe there's a tripling of the market relative to latent demand for helicopters, but it's not going to replace cars. Cars are cars and they're pretty good at it.
I have a business contact friend that booked a helicopter to get from Barcelona near the airport up to the Spanish Pyrenees on vacation. In the nearish future it'll be a flying car, i.e. helicopter 2.0 (basically a big drone with seats).
The business model for the flying car (or flying Waymo) is the weekly commute.
You have that large mansion in the Hamptons and your whole family just loves it? You are even allowed to work from home office one or two days a week. That means from Monday to Wednesday you buckle up for your weekly grind at the office (where you have miniscule studio apartment or even a capsule hotel), then it's back to your real home for family time and home office.
Yep. That's a very niche use case! Weekly commutes are rare and only rich people have mansions in the Hamptons. Thus, it's a small market!
In my mind, the big advantage in flying cars is , not medium range travel but reducing congestion by taking advantage of the large 3 dimensional space over our heads. The limitation here is traffic coordination, which could be solved using computer networks to pilot the vehicles
You're moving the congestion from the road to the landing pad. You're going to be rate limited on how many flying cars you can land in a given area, and it's likely going to be a lot lower than private cars, let alone trains/buses/bicycles.
That already exists, but you can’t fly it yourself. Blade offers private air commutes between NYC and East Hampton, but it costs over $500 one way. If they used drones instead of pilots and the FAA allowed that, then it may become cheaper.
https://www.blade.com/p/hamptons-same-day
This book has been on my list for a while. Now I don’t need to read it.
Engineerism is a good diagnosis. The safety issue with flying cars has always been obvious to me. When it comes to personalized flying technology, drone delivery on demand has always seemed a lot more useful, and we indeed have it. In general, our need to move *ourselves* has decreased with virtualization, but we still need to move goods.
The discussion of energy usage is a mild criticism of Vaclav Smil’s work, which tends to reduce human progress to energy usage.
The engineerism critique is also a veiled critique of Musk. Data centers in space, the Boring Company, …