How Desalination Could Create New Superpowers | Samo Burja

Today on Upstream, we're publishing a recent conversation with Samo Burja about a technology that could reshape our world and change how you think about human potential. That technology is desalination, which isn't just about providing drinking water. Up ahead, Samo breaks down what the greening of large deserts in countries like Saudi Arabia or Australia might cost, and how it could shift the geopolitical balance as we know it. This conversation showcases what Samo does best in his podcast Live Players, and newsletter, The Bismarck Brief, uncovering the key players, institutions, and technology that drive the world forward. Let's dive in.

Let's get into your piece on desalination. Maybe first, what inspired you to write this?

Basically, I think that people People have a somewhat inaccurate understanding of what are the actual bottlenecks for our society versus what are not. For the first approximation, we can, you know, grow all the food that's needed for all humans in the world. Technologically, there's no special secret to doing more. Now, we can always grow more food with less labor, lower environmental impact, et cetera. Well, we produce enough raw calories, yet people still talk about, you know, solving global hunger, which in a way is more of a problem of 50 years ago than it is of today. This is not to deny the famines happen, but they happen for, say, political reasons where, you know, you might have a redistribution of land, or you might have an interdiction of food supplies to a region, but they're, you know, all illusory. have been getting rarer and rarer since the Green Revolution. Similarly, there has been a quiet revolution in drinking water. We're all used to basically hearing about impending water wars and resource shortages, and the planet is warming up, so drinking water is going to be scarce. Therefore, we have to conserve drinking water and so on. Yet to the first approximation, Technologically, drinking water has been solved. In fact, it's been solved so well that the only reason we don't desalinate our water is because even cheaper than desalinating water is reclaiming wastewater. So I think that there's something sort of sacred about food. There's something sacred about forests. There's something sacred about water. I think that there's this view that catastrophic droughts And changes in climate means that there'll be places that just run out. And as far as I can tell, as long as you're a moderately developed country, you're not going to run out of water. In the modern world, there are tens of millions of people in various countries like Israel, Cyprus, Singapore, the Gulf monarchies that get their water from desalination. This is a result of a technological revolution. So just as calorie production in farming had the Green Revolution with the introduction of fertilizers, there was innovation driven by necessity, and it was basically driven by Israeli scientists in the '60s and '70s. They radically reduced the cost of desalination, and they used a technique called Reverse osmosis, where you basically force the salt water through a semi-permeable membrane, which then means that, you know, all the salt and so on gets stuck on one end, and on the other end you have desalinated water. It produces a certain amount of like concentrated waste, so basically brine water, and that, as far as I can tell, can just be disposed of quite safely back in the ocean. It might be a problem if you're desalinating a lake, but even in that situation, this isn't this terrible side effect. Note that, for example, the salinity of wastewater from a desalination plant, so the water that remains undrinkable, is actually lower than the salinity of the Dead Sea. So there has been a proposal to pipe the water into the Dead Sea actually reducing how saline it is, right? So it's really, it's a very handleable ecological project. Now, the truth is no one knows what would happen if you scaled this to providing drinking water for a billion people, but you kind of could. There's not that big a jump, and the industry and technology to make it is also well distributed at this point beyond Israel. It's again distributed in Singapore, in Spain. Spain has numerous desalination companies, but it's still notable to me that it took a country that was facing geographic and resource constraints that were pretty harsh to invest in developing a new technology. The reason why it's hard to just make money by making water is that water is ludicrously cheap on Almost everywhere. And the places where water remains scarce that are far away from the sea, so very inland, perhaps up in mountainous regions and so on, the cost of pumping it there is higher than just the cost of having fresh water made directly from the sea. So drinking water is sort of this resource where the The only time you are incentivized to develop the technology further is when you're facing a hard constraint, just as Israel did and just as Singapore does today. Singapore does not want to be dependent, for obvious reasons, on water imports. A lot of the technologies they develop and that they fund development of are attempts to use technology to make this very small city-state resilient. And then of course, there are many Pacific islands and so on where rainwater and collecting rainwater and treating wastewater, it becomes more of a hassle than just desalinating. I think that there has been research into making desalination more energy efficient, but the costs are relatively low already. And, you know, it doesn't quite make sense to power desalination with solar energy. Now, the reason is because the capital expenditure of building a desalination plant is the biggest cost. They don't actually need that much energy, right? You know, the cost of a plant can vary between $600,000 to $1.2 million per 100 cubic meters of capacity, like cubic meters of like water desalination capacity. For example, in Saudi Arabia, the Ras Al Khair plant's costs totaled up to over $7 billion, right? And these are costs that are primarily the costs of building the water intake and discharge tunnels. So just normal, straightforward construction, the high-pressure pumps for reverse osmosis, which is sort of like, you know, just a normal industrial product. And then various treatment equipment for both ingoing and outgoing water. You know, to explain the theoretical minimum of energy needed to totally desalinate 1 cubic meter of typical seawater with reverse osmosis is about 0.86 kilowatt hours per cubic meter. So, by the way, a cubic meter of water for intuition is like about 1,000 liters of water. Like quite a bit, quite a bit of water. And 0.86 kilowatt hours is the minimum energy cost. And that's the theoretical minimum. What's the actual practical minimum energy cost? In 2021, the Saudi Water Authority broke the record for using the least amount of energy for desalination, and they posted a result of 2.3 kilowatt hours. per cubic meter using reverse osmosis. Now, basically, for comparison, I gave the theoretical minimum for 1,000 liters of water, 0.86 kilowatt-hours. That's less than the amount of energy necessary to power a 100-watt light bulb for 10 hours. So we're really talking about not large amounts of energy at all. And because of that, the fact that solar makes daytime electricity very, very cheap doesn't actually reduce the costs. So the bulk of the cost that would make desalination plants cheaper would actually be ways of reducing the costs of digging and making tunnels and intake and outtake. You can probably on the margin improve the costs also with mechanical improvements. Better membranes are better membranes that wear out less and so on. But still, it seems that if you are a modestly well-off country, you can afford this. And like, which are the countries that afford this? Like, to give a relevant statistic on the water of the countries using it, right? You know, desalination is viable for countries like Israel and Saudi Arabia, Singapore. These are well-off countries. But they're also just places that are much, much poorer than Europe where desalination is viable, right? Like there is places such as Jordan and so on that have begun using desalination technology. The Arab Gulf states of Saudi Arabia, the United Arab Emirates, Qatar, Kuwait, Bahrain, and Oman together make up 60% of global desalination capacity. use over 400 desalination plants. These countries rely heavily on the capacity for drinking water. So approximately 42% of UAE's drinking water comes from desalination. For Kuwait, it's over 90%. For Oman, it's 86%. And for Saudi Arabia, it's 70%. So literally, we're talking about tens of millions of people. These are large populations, right? Saudi Arabia, Is again tens of millions of people. The Gulf monarchies together are an even bigger population, and you have much poorer countries around the world occasionally using desalination. I think that honestly, the main thing holding this technology back is that we don't have something to do with all that fresh water. I opened earlier talking about calories and how cheap calories are. To explain why scaling desalination to provide water to farm is unnecessary because we produce way too much food in the world already. We're currently already at a global caloric surplus. It's just a distribution question. So there's very little profit to be made from pushing forward desalination using the demand induced by farming. Because, you know, if a region dries up, people stop farming there. It actually doesn't matter. The most efficient farms in the world are in North America, in Europe, in Ukraine, in China, even Argentina. These are places that even if you reduced rainfall by 80% would still be very wet places with plenty of water. Now, don't get me wrong, it probably doesn't make sense to grow almonds which are very water-intense crop in California. That's a policy decision. That's not really an ecological decision, right? It's kind of a weird artifact of basically arbitrary farming subsidies. It's not a necessity for nutrition or food security or anything like that. So farming uses way more water than humans drink. So drinking water use is much smaller than farming. So then how in the world could you increase the demand for desalination by many orders of magnitude? So if the climate apocalypse and pollution environmentalist people are correct, then it's very natural to scale it. As we have water shortfalls, countries will build desalination plants to substitute water they previously got from rivers or from fossil water sources underground, or water that they could previously reclaim.

Do you think those people are correct? No.

I mean, they're, they're right that there will be a number of countries that will undergo actual climate-related stress. But I think it's going to be very gradual. And I think the biggest effect will be not so much in precipitation in those countries that won't necessarily become much drier. In fact, a lot of them might become wetter. It's just that they will experience more extreme weather. And actually, ironically, they might experience much more rainfall. You know, when the planet was much warmer, the Sahara was greener, not drier. There's a general way in which, you know, actually the planet becomes drier during an ice age because more of the humidity is locked up in ice, right? More of the fresh water is locked up in ice. And therefore you might have the biggest desert in the Sahara might be during an ice age. It depends on which ice age, which era, what the exact climate situation is. But I think equating global warming with water shortage is incorrect. Now, equating population explosion with water shortage, that can be correct. Because you might have competition between a rapidly growing population and the needs of, say, farming. And if you have a wave of farmers leave an area because they can no longer irrigate their crops, now maybe they're not starving, but maybe they become something like climate refugees. However, in a lot of the world, as we discussed in previous episodes, the population explosion is ending. Looking at the economics and the cost of desalination, It's shrinking, but even in places that are growing, let's look at India. The economics are actually affordable, right? Like, like, it's just, we're talking about cents, right? We're talking about very, very low costs per liter of water, right? Where it's sort of this, this kind of interesting situation where I think people assume it must be very expensive. But we're talking the production cost is 0.3 cents, right? Like, that's, that's like crazy.

We'll continue our interview in a moment after a word from our sponsors. Most of you probably know how much the Turpentine team loves Notion, the sponsor of today's episode. We use it every day for editorial planning, transcript markups, project managing, knowledge sharing, and it's where I drop random 3 AM ideas that turn into episodes. Now it has the power of AI built right inside of Notion and woven across our entire workspace. It's been a total game changer for our productivity and focus. We just used it the other day to architect the content schedule for next month across our network. Notion combines your notes, docs, and projects into one space that's simple and beautifully designed. The fully integrated Notion AI helps you work faster, Write better and think bigger, doing tasks that normally take you hours in just seconds. Try Notion for free when you go to notion.com/upstreampod. That's all lowercase letters: notion.com/upstreampod to try the powerful, easy-to-use Notion AI today. And when you use our link, you're supporting our show. That's notion.com/upstreampod.

It'll cost $32 billion to desalinate. 100 cubic kilometers of water. So $32 billion to desalinate at current technology with no technological innovation to desalinate 100 cubic kilometers of water. That number, for reference, is slightly more than the average annual amount of water that flows through the Nile River, the world's largest river. Like, that's kind of a crazy number. And look, Egypt's GDP, like, if we check up Egypt's GDP, its current GDP is like $400 billion. So, you know, $32 billion would be a big expenditure for a $400 billion GDP country until you remember that 100% of Egypt's GDP exists because of the Nile River. The Nile River is just what makes the country habitable. and worth living in. And it's the reason that there are 100 million people there. So there's a funny way where, you know, if Egypt wanted to clone itself, it could produce an artificial Nile River and probably solve its depopulation crisis. And the most expensive part would just be pumping the water uphill. But even that's not necessary. They have a large depression. There's an Egyptian depression that is below sea level. And there has been discussion of— it's called the Qattara Depression. The Qattara Depression is basically— the average depth of it is 60 meters below sea level. So it's stuck inland in the Sahara Desert. There have been proposals to flood it with seawater. I actually have a different proposal. I think what Egypt should do is build— well, let's say about $30 billion. dollars worth of desalination plants on the coast, desalinate water, and dig a long channel or tunnel, cover it so that you reduce the losses of precipitation, and flood the depression. Instead of flooding it with seawater, where the result would be over time rapid evaporation and eventually becomes a salt lake bed, it could become a freshwater lake. Immediately, the coastline of this lake becomes some of the best and most prime real estate in the world, or at least in Egypt. Modular economic development. The evaporation from the large body of water produces more rainfall. And you know, well, you've just doubled Egypt's habitable landmass. I think that's sort of the future of this technology. And I wish that, let's say Saudi Arabia or some of the Gulf monarchies— I mean, they're much smaller, they're more island-like. So let's focus on Saudi Arabia. If Saudi Arabia were to use $300 billion to desalinate the water equivalent of 10 Nile Rivers, it could transform a huge portion of their desert. into green, lush landscape. This might again sound relatively ambitious, but I think it would do more to change the opinion of people. It would do more for the brand of the country than The Line will. At the end of the day, even if you live in a shiny skyscraper that's air-conditioned, no one wants to live in a desert. Why aren't we more ambitious? Saudi Arabia has mountains. Why don't we plant both native tree species and completely new ones? Why not desalinate such vast amounts of water and pump them into lakes and pump them into inland rivers that you terraform landscapes into green, pleasant lands? Now, of course, these will still be very hot territories, but a green landscape. To be the strongest thing people associate with Saudi Arabia would do more to transform the image of the country in the minds of the people around the world than anywhere else. And you can call it the Mohammed bin Salman National Park, half of the size of the Yellowstone Park or something like this, maybe even plant redwoods just because you can. You have beautiful wild animals. You can camp there. You can hike there. You have air-conditioned hotels. You can kayak down artificial rivers that have been sculpted for an optimal rapids experience. It really starts to sound amazing. And it would also be one of the most effective flexes to the population of the country. If you create a giant park that's named after you, a natural park, think of it as a country-sized Central Park. Well, everyone in the country is going to remember your name forever. And it would be truly one of the wonders of the world. Let's remember, one of the wonders of the ancient world, the 7 original wonders of the ancient world, included the Colossus of Rhodes, the pyramids of Egypt, the Temple of Artemis, the Temple of Zeus, the Mausoleum of Halicarnassus, and it also included the Hanging Gardens, where they were a marvel of waterworks. Because you had these gardens on stepped ziggurats. And the wonder was that they could economically pump water up against gravity. So today, the King of Saudi Arabia pumped 10 Nile rivers worth of water from the ocean against gravity and desalinated them for a cost that would probably be lower than The Line. They could create a territory the size of Florida or half the size of California that was green, lush, beautiful, and could be sculpted as a sort of national park. And I think that would do more to attract residents and migrants to Egypt than even the fanciest architecture, because there are always people who love the harsh beauty of the desert. But I'm not even sure Arabs love the harsh beauty of the desert. I think most people Like greenery and beauty. So this kind of terraforming project undertaken in Saudi Arabia would be a national rebranding project. In a different country such as Australia, this project or these types of projects using desalination, where the cost of desalination would inevitably go even lower than the ludicrously low 32-cent number that I cited, this would put the country on an equal footing with the United States. If half of Australia could be greened for $30 billion a year, this would more than pay for itself. The mainland US and Australia are of comparable size. The main difference is the Mississippi River and the lush, temperate eastern coast of the US. So again, you could have an Australia Where 300 million people live. Now, where those people would come, whether we have the population growth, whether they all come from India, if, if I reference recent immigration discourse, that's a separate question. But it would definitely make Australia potentially a great power. So again, I think we're sleeping on the possibilities of using advanced technology to solve geopolitical problems. We're used to geopolitics being defined by the natural geography of the world, but we have achieved truly a level of technology where we can change the natural geography of the world. And that's viable.

And say more about this, about which countries are best suited to benefit from this in a geopolitical sense in terms of it changing their calculus. You mentioned Australia as one.

I mean, for Saudi Arabia, the branding works on 5 levels. One of them is You show you're climate resilient and that it's not going to be a collapse of your country because you're already changing what your country is, and that's not going to change no matter how bad global warming is. Secondly, it's a technological project. You show this ability. Third, it's like this immense demonstration of political legitimacy. You know, how many times have you heard that Israel has greened the desert, made the desert bloom?

Right.

That's true. Right. The landscape just looks better. And I think people on a basic level believe Israel is legitimate partially because the landscape looks nicer, because it looks greener. I think there's almost this superstition that if the land is thriving— now, of course, environmentalists will object because they always object. But people, when they see a big, beautiful natural park in Saudi Arabia, their first thought is not going to be, oh, this is like a terrible This is like a terrible shame. They reduced the size of their desert to merely cover 2/3 of the country rather than all of the country. They're going to think, oh wow, the Saudis are really good at taking care of the environment. Maybe we can learn something from them. Because in my own country, I see desertification happening and they're reversing desertification on a truly unprecedented scale. So that's why it works for Saudi Arabia, these 3 reasons. I mean, the fourth reason is I think it would develop an outdoor culture again. You could have like hunting expeditions in the forest, et cetera. Like the, the, the national traditions of falconry and all of that work just as well in a temperate landscape as they do in the desert and so on. But geopolitically, I think the biggest beneficiaries would be developed countries.

Hmm.

that for some reason or another wish to develop more land for settlement. Perhaps real estate costs are high. Perhaps there are important economic developments, minerals that they want to access, and it's difficult to get people to move there. So any country that is already fairly wealthy and that has a large desert territory could potentially just make it habitable, even if it's fairly deep inland, right? Because you do have saltwater lakes and so on. So in the Western US, which are these countries? Australia, the United States, China— actually, half of China is a desert. India, to some extent, could definitely afford to Reclaim more land. And they're one of the few countries that still have a growing population, though this will change very rapidly. Spain. Spain could actually— it could transform its semi-arid climate into a climate much more similar to that of France. Now, whether they would want to is a separate question, but that's a very important possibility. Turkey. A lot of southeastern Turkey is very arid and could be transformed with this technology, and the economic development this would bring, you know, basically it would result in greater Turkish national integration. Because if the economic disparities in regions are less, the regions are more integrated, then things like Kurdish separatism is less likely. There's also a way in which these mega projects Politically tend to bind countries together. So that's another domestic political advantage. If you have a national system of waterworks where for Madrid to stay lush and green, or for Catalonia to stay lush and green, you have to have cooperation of this massive water system across all of Spain. Well, that's then an argument against Spain breaking up into smaller countries. And this is actually an ancient argument. In Mesopotamia, Egypt, and China, the central administration of the water system meant that you couldn't just have a city on the river declare independence. You could literally cut off their water, right? So again, for Turkey and Spain today, this would be a centralizing measure. And ironically, it would probably be a centralizing measure too for China because its drier provinces tend to be the provinces where various ethnic minorities like the Uyghur live. In the United States though, it could result in the next frontier. The US had over 100 years ago this moment of the closing of the American frontier where it felt that the national character would be inevitably changed by there not being a frontier. Can you imagine Arizona, New Mexico, and Utah as green and with ample fresh water, with possibly new national parks just made in a terraformed landscape. So I'm including national parks because I think that part of people not just consuming the water indoors, but having outdoor areas that they admire and that they're beautiful. And again, you could say, oh, but this is wasting so much water. But what does it even mean to waste water? if you're desalinating it in huge quantities. A lot of federal land that is currently basically useless and underdeveloped would suddenly become extremely attractive. It would become some of the most attractive land in the world inside of the United States. At the very least, you could restore the Colorado River.

Are there any other technologies you're as excited about as this one?

Also, just to give a little bit of a shout out, the physicist and entrepreneur Casey Hadner has basically proposed a plan to unlock, like, you know, a lot of the potential of the western United States through such a measure. And I think it can easily lead to something like, you know, a California-sized population wanting to live there in addition to the current population that previously didn't, and would make these states some of the wealthiest and most important in the country. Well, for other technologies that I'm excited by, that are in a similar vein. I think that we greatly underestimate the possibilities offered by essentially solar mirrors. I think we could have advanced orbital mirrors provide sunlight to the far northern and far southern latitudes of the world. We're at a time in history where it's very easy to make indoors comfortable and warm and nice. But why do people not want to live in the north of Norway or in Iceland or in Greenland? Perhaps this will be freshly relevant if President Trump achieves his dream of adding Greenland as a territory or state of the US. People don't want to live there, or perhaps in Alaska— that's already a state— because the day is so short. I think we underestimate how important ample sunshine is to people. So literally having orbital mirrors that reflect a large amount of sunlight at particular times onto cities and other urban areas, I think that would radically increase the desirability of living there. Again, we don't want to be indoors all the time. We crave outside, we crave the fresh air. We don't care if the air is like super cold. It's actually even kind of pleasant if the air is very cold, but there's bright sun out. I think you just kind of feel good about it. It's almost refreshing or exhilarating, as long as you're dressed appropriately for the weather. What really gets to people is the source of massive winter depression, even in countries where the culture is quite adapted to this lifestyle, such as Finland, Sweden, and Norway. Massive boons to human wellbeing. And again, I think we're not used to thinking about re-engineering the amount of sunlight a city receives or how green the surrounding of a city is, right? And these are things that could be done with relatively minimal negative effect. I think that you would have a very negligible warming in the individual cities. But fun fact, Cities already are pretty warm. There's an urban heat island effect in play in every city, just because of all the machinery we are running and all the heating we produce. And also because the, there's a lot of concrete, which tends to absorb more sunlight than say a snowy field would, or a green forest, or, you know, a white, sandy white desert.

Is there desalination technology or related discourse or how it could change things or that we haven't yet covered that you want to make sure we do?

I don't know. I would just like it if people understood that water abundance is within reach and there's no need to fight water wars. Water wars are, you know, unless you're an extremely poor country, they're unnecessary. You could literally provide enough water to a billion Indians or a billion Chinese. The real fight there is arguably over the hydropotential. So the hydro energy potential where there's water coming down from the Himalayas. And this could be used in hydro plants to produce energy either in India or in China. And that sort of disagreement is driving a lot of the still remaining border tensions. And there's also the fact that no country ever wants to give up any territory if it can help it. I think that the technology is also, I mean, at least to me, exciting because I think that it truly allows us to live anywhere on the surface of the Earth, where before this was one of the last remaining hard physical limits.

Even Antarctica?

I mean, Antarctica was never short on water, but we can live in Antarctica. Nothing stops us. The main reason Antarctica is not developed is because of an international treaty which basically bans commercial development and mining. You know, this treaty will come up for renewal in 2048, and there are many countries that are positioning themselves for a scramble for Antarctica. China has started developing and supporting many research bases and military bases there. The U.S., of course, supports bases there. The Russians have been prospecting in the area. For example, they discovered the world's largest deposit of oil off the shore of Antarctica. The deposits are larger than that of Saudi Arabia and are probably technically recoverable even today, but they're in waters covered by the treaty. Now, the treaty froze the preexisting territorial claims. So there might be a second Falklands War, this time over the Arctic oil reserves, which are in an overlapping claim. between the United Kingdom extending down from the Falklands and Argentina.

How do you think it'll play out?

I think that by 2048, Argentina could be much more powerful if it continues developing well economically. You know, maybe Milei becomes president for life, and by 2048, the Argentinian economy is a first-world economy. They have great relations with the US. Meanwhile, Britain has imploded. They could win. Potentially such a war. And then there's a fun question where would China or other countries wish to see 2 US-aligned countries at it fighting each other? Of course they would. So there's also the fun story of foreign involvement. And by that point, if Argentina acquires the world's largest oil reserves, Erik, no matter what happens with the electric vehicle transition, the global demand for oil will stay massive. Not just to burn oil for energy and electricity directly, which is done in some places, of course, mostly more for natural gas, but plastics production. We're not going to be consuming less plastics by 2048, no matter how environmental we try to be. So much of the planet will be so much more developed that demand for oil has a floor below which it can't fall. It's a very, like, chemically complex substance. You know, when One of the key chemists first analyzed it, he felt it was too valuable to burn. So I would not be shocked if we find new uses for oil, which after all is this like interesting substance that has all these complex molecules in it, all these complex hydrocarbons, all useful precursors for all sorts of processes, all sorts of industrial chemistry.

Going back to the sort of global hunger part, when people say that China is going to have trouble feeding itself once sort of the US stops protecting the waterways, is the ocean sort of access to certain regions, is that just misunderstanding sort of technological advancements that we've made?

I think it's false. I think it's straightforwardly false because at the end of the day, If China was deeply worried about this, they would build a pipeline to Russia. And who in the world could stop them? No one really. It's so deep inland in Russian and Chinese territory that the West could not— no Western-backed countries could send divers to blow up a pipeline that's so far away from the sea. So here I'm referencing the Nord Stream pipeline, was blown up because it's accessible. This would be deep inland. It would require either a terrorist attack, which is like tricky to carry out if they cared about security, or it would require an airstrike, which again, we're already talking about such a serious war that it's almost incomprehensible, right? If you're bombing territory deep inside of China, well, they might bomb territory deep inside the US. In retaliation, right? So there's also like, there are also large shale reserves, which they could unlock if they privatized mineral rights for a variety of reasons. Shale oil, basically an oil well produces most of the output that it's going to produce in the first year or 2. So this means you actually have to have relatively Nimble teams developing these small shale oil fields over and over again. Technologically, it's a little bit tricky to get the deeper Chinese shale oil, but in a war situation, I think they would access it. And as a result, I don't think they would have a shortage of fertilizers. They wouldn't have energy shortage. Look, Erik, here's part of something that people miss. The European Union is transitioning to electric vehicles to save the environment. China is transitioning to electric vehicles for national security. They are not the same, right? One of these is in a way a much, much healthier project, and I'm referring to the Chinese one. They are trying to get off oil so that their national oil imports don't have to increase. We talked about this a little bit in the previous episode about how, and, you know, I recommend to listeners to check out that episode, which detailed China's energy needs. They actually produce quite a bit of oil already, even while not touching the shale oil and not importing anything from Russia by pipelines because they've not built any of those pipelines yet. And if you push forward the development of very cheap EVs, And just replace all urban cars with EVs, you suddenly have all this excess oil that you're just producing in China that can be used for fertilizer. And then if we're talking about that, if there was anything approaching famine, China is governed by Communist Party. They could actually simply consolidate the currently still very fragmented small Farms into larger, more efficient farms. Now they couldn't do it quite with central planning because that doesn't quite work. But if the Chinese government like picks ten big agri corporations, makes ten state agri corporations, and has them compete against each other, and just like takes a bunch of land from random small peasants and gives it to these big agri corps, I think they could massively increase food production while not massively. increasing oil production. So in a way, China is still lowballing its ability to feed itself because feeding itself is sort of like, you know, it's not that cost effective. They buy soy from the United States to feed to Chinese pigs because that pig feed is slightly cheaper than the alternative. And, you know, if the US doesn't give them soy, They're not going to automatically grow soy. They're going to go for less efficient, slightly costlier feeds for pigs. And price of pork might increase by 50%, but that's okay because the income of the average Chinese person has increased tenfold in a generation.

Right.

And this is not even positing any future growth. So I think they are not going to starve. They're not going to have food riots as long as you can blame foreigners for it. I think this is perhaps one of the biggest differences between the current era and the last era of industrial war. It's easy to think that, oh, it must be an information change. But unlike World War I, where food scarcity was a big problem for Germany and Austria, I think currently every major power, even Japan, even Japan with this extreme scarcity of land, can probably feed themselves just with the resources in their territories. And that's kind of crazy. That sort of, that is a very different dynamic of war. Laying siege to a whole country to starve it in a literal sense. just won't work. You can still starve the country industrial resources. Like, for example, I think the modern military stack still clearly runs on oil. No one's proposing EV tanks, okay, or EV personnel transports. But let's not ignore that there is something that is battery-powered about modern war. Drones are battery-powered, right? So It's not like oil is like absolutely necessary in huge quantities for everything. Though, of course, again, the drones, they don't burn oil, but they do consume oil in the sense of the plastic components of the drones. So again, you could choke off China industrially, prevent it from becoming a rich country if you're willing to engage in a multi-year war, but you couldn't starve it.

And, and the technology that enabled every country to feed itself is a combination of shale and sort of different energy sources, or what is the sort of the— what is the difference that enabled it?

It's just the green— it's just the green revolution, essentially. Agriculture, just agricultural advancements. And it is still relatively oil dependent. So I did caveat that with every industrial country. But basically, like, you know, every industrial country has— well, there's 2 things happening, right? Like, every industrial country has also massively automated farming. I think we, we don't think of it this way, but it used to take 60 to 80% of the population farming to feed yourself. Now it sort of takes 1 to 2% of the population.

Yeah.

So if you're experiencing some food insecurity or planning to, most of these countries actually have long-abandoned farmland. If you look at a map of Europe, of European forest coverage compared to 50 years ago, forest coverage has increased massively, mostly through abandoned farmland. If you look at the statistics for the agricultural output of those countries, most of them have increased the But so the first one is fertilizers, which are fossil fuel-based. Second one is mass automation in the form of farm machinery. And, you know, maybe an EV tractor is not quite viable right now, but, you know, the Cybertruck's not far off. I think 10 years from now, the EV tractor will be perfectly serviceable. You can have a battery-powered tractor with an electric motor and will do everything and be capable of doing everything that an internal combustion engine tractor is capable of. And the third reason is just incremental improvements in yield for every single species we farmed. Did you ever see that chart comparing chickens from the 1940s to chickens today?

No.

Okay, it's kind of an amazing graph. We can link it in the description. Here is a fun little chart. You know, since 1957, the size of the typical farm chicken has gone from the chicken weighing like a little over 900 grams to now weighing over 4,000 grams. So that's actually a fourfold increase in just chicken size. And then you have other advances such as rice, wheat, corn, where the plants are just much, much more productive. And of course you can argue, okay, a lot of this is dependent on pesticides and it's dependent on oil. So if you could deprive a country from oil, then its farm production will go down. But even so, some of these productivity gains do not rest on fertilizers, right? Some of these would just remain. And because of that, It seems fairly trivial to scale agriculture in most modern countries.

One thing I want to return to is when I went to college, I was told in my environmental science class that there was a sort of man-made climate change process happening and that it was potentially catastrophic if we did not act fast to reduce the amount of carbon that we emit into the atmosphere. What is sort of the right way of thinking about sort of the, you know, a couple decades later, 15 years later, to what extent was that correct? To what extent was that incorrect and how?

Well, we could do a full global warming episode and climate change episode, but to say a few things, I think that it is a good idea for humanity on the scale of centuries to reduce the atmospheric carbon to lower levels, but it is in fact not going to be disastrous in the next 50 years. And the reason is, well, if sea levels rise and you have a big city, hire the Dutch and build some dikes. If the temperature goes up and you have massive heat waves where thousands die or millions die, Well, you know what? This makes the AC unit a humanitarian necessity. Let's be pessimistic and say that it costs $2,000 to deliver an AC unit that anyone can plug in their window together with a solar cell rack that will be big enough to power it when the sun is out and when the temperature is high. Heck, let's make it $4,000. Okay, $4,000 means But there's a billion Africans. You can give each of them an AC unit for $4 trillion, right? That's a huge amount of money, but it's not that big actually. It's doable, right? Especially if it's very incremental, right? And the same is also true of India, right? Again, it's like, let's be hyper pessimistic. I think these are unrealistic costs. I think in reality you could get them down to like $400 to $500 actually, right? So AC units are climate resilient, especially if they're powered by solar energy. It's like one of the few very beautiful synergies because it's like, you know, when the sun is shining, it's super hot. There's a heat wave. That also means there's a lot of electricity from solar.

Okay.

Works out. Humans can take shelter inside. And then the final one, it's sort of, you know, extreme weather. I think extreme weather is a significant problem, but what does this amount to? This amounts to, well, you don't want your city's dikes to fail when a hurricane passes through the city. So again, it's an infrastructure problem, right? You don't want to be New Orleans, right? Or it might mean that your buildings have to be more resilient against extreme atmosphere. than they were before. Very high winds, for example. Again, this is doable. So in a very fundamental way, I don't think climate change as we are going to see it in the next 100 years will break anything about industrial civilization. Will it increase costs? Yes. But will it increase them prohibitively? No. Now, there are other challenges for industrial civilization. And they're much subtler, and we should discuss them. And of course, we should not be inviting environmental damage on a global scale. And I in fact support the extension of nuclear energy. I support the extension of solar. I support the mass use of batteries for everything we can possibly use batteries for. I support the development of geothermal. We can electrify our appliances thanks to the battery revolution, right? We can electrify all the uses for energy, or almost all of them, or 90% of them, that previously couldn't be electrified because first-generation electrification was we plugged everything into the national grid. And you know, your car can't be plugged into an outlet when it's driving around, right? But there are other things that can't be. Like, you know, like literally some, something trivial, like a chainsaw, right? A chainsaw can't be really plugged in to the outlet. So we had to burn fossil fuels for that. And I think aircraft are energy dense enough that we are going to have to burn some fossil fuels to keep aircraft around. Though, of course, we might choose to synthesize the fuel and like, we might literally mandate like Simply, we mandate, okay, aircraft have to be carbon neutral. Every flight you take has to have priced in an equal amount of carbon sequestration or carbon removal from the atmosphere. And, you know, we in fact can synthesize methane and we can synthesize just from pulling carbon from the air. If we had abundant electricity, we could synthesize hydrocarbons that planes could burn. So even that aspect is not insoluble, though of course that would really increase plane tickets. It would go back to being like a business class experience. Basically, you would pay as much for an economy class ticket as we today sort of pay for a business class ticket. But that's, again, that's like eminently survivable. And maybe it makes my train friends happier because the train tickets would have no reason to increase that way. So maybe the— Maybe by 2048 or 2050, when the California high-speed rail is finally built, you take a train instead of a plane. And that's kind of inconvenient, but you don't have to go through the TSA, so it doesn't feel that bad. You can work on the train. So really, I think that climate resilience actually amounts not to learning to live in harmony with nature, But solving something like 20 separate logistical challenges and then being rich enough that you can solve them, having the means to do it. Like you build a desalination plant now, it's much cheaper and easier to have. So it means that you have budget later on to say, build better storm drains for some of the Californian rivers that do become a flooding hazard. If you had extreme sudden rainfall, for example, or investing in water management full stop might be a great idea. You could literally divert, have overflow where rivers that might under extreme rainfall really flood dangerously. They might be easily diverted to reservoirs that keep the water for later to be used during a drought. Or for agriculture or for drinking water or whatever. So a lot of this is just building infrastructure in cities around the world. And I think in a way that's sort of, you know, it's, it's in a way also less sexy technological story than the technological story of, of batteries or desalination, because it means a bunch of local efforts everywhere. And if there is one thing, there is one transformative technology that I'm excited by that's not been discussed. I think it's the application of artificial intelligence to construction. Can you imagine automated mining equipment? Just not just self-driving trucks, but self-driving digging equipment. It's not that exotic, right? We basically, like, technologically, I'm not even sure there's a challenge. Like if there were someone that invested $5 to $10 billion to bootstrap an automated digging company, you could probably build the cheapest canals and the cheapest ditches and dig the cheapest giant holes ever dug in the history of mankind. There's just like not necessarily a massive market there, but that's like one of the few technological stories where you could make construction universally cheaper around the world, basically like excavation robotics, construction robotics. And those revolutions might be happening in the near future anyway. I think it's going to be a very natural next step to go from self-driving cars and self-driving trucks to self-driving excavators and self-driving farm equipment and all of that.

I want to wrap on that topic and I want to get you out of here in a few minutes, but I'm curious if you have any immediate reactions on the immigration side before you go, because, you know, it's been the first kind of fissure between—

It's been a hot-button topic on, on, on Twitter, right, over the last few days. People are starting now to talk about, you know, the paleo-MAGA right and the tech MAGA right. It's been like kind of a disagreement. Yeah, yeah, over the H-1B1 issue. You know, currently big tech has difficulty accessing H-1B ones. It's actually soaked up mostly by medium-sized companies. And there are like just a few companies that basically dole these out. So there's arguments for reforming how they're distributed. And I understand why there's a desire also to increase the supply. But, you know, let's put it this way. The von Neumann arguments are a little bit dishonest. They're not 300,000 von Neumanns. The Operation Paperclip arguments are also a little bit dishonest. That was a headhunting operation where you had like a big thick dossier on every single scientist and you would get them, you know, you get them a place to live, you get them amnesty for their crimes. Honestly, that would be a great idea. I would totally support a federal bureau called the 10,000.

Okay.

The 10,000 is a federal bureau that has the mission to find the 10,000 most talented people in the world and make sure every single one of them moves to America. And it has the following levers and incentives. It can give them a green card, no questions asked. It can get their kids into Harvard or Yale, which by the way, is something that was used on the Taliban commanders and apparently was one of the strongest incentives, getting their kids accepted into elite American schools. It can give them a tax break. Look, America taxes you for 10 years after you give up your citizenship. Maybe we should, to attract certain people, be like, okay, if you, you specifically come to our country, do your magic here, we're not going to tax you for the next 5 years. And then we can even go further. We can be like, oh, maybe there needs to be a witness protection program. Remember that guy, that warlord? That sort of rebelled against Putin and you know crashed crashed his flight. That's an asylum argument there. That's a you know even if he just tells us what he knows of the Russian military, that guy would be worth exfiltrating and you know giving a false name, etc. And let me say the final piece of this plan for the ten thousand whoever's in charge of it should be in a DARPA mindset. It should be like fifty people at most. going through these profiles and going through people in the world, and they should each have relative autonomy for who they nominate. So not a big sprawling bureaucracy. And also, when the quota is met, when the 10,000 are found, it wraps up. So that is what a modern-day Operation Paperclip would look like. Because often people who are super talented are deeply enmeshed in their own valuable networks in various parts of the world. Like, they actually have opportunities even outside of the US. The US is great and can certainly attract many of them, but many of them actually have a lot of things that they're up to in their part of the world because unsurprisingly they've climbed to the top. And the other side of this is big dumb bureaucracies are very bad at figuring out who's good. Any bureaucratic system of mass migration you set up will be gamed and will be gamed badly. It can probably run for a decade or 2. So I'm always in favor of exceptional talent and recruiting exceptional talent, but I think exceptional talent is super rare. And then in my mind, the question of whether we should double the number of H-1B visas is not a question of whether, whether America will get unique talent that it cannot get in the United States or cannot get through current immigration channels because it's a big, dumb, and somewhat broken immigration system that's actually really bad at recognizing talent. I think it becomes a question of, do we want to reduce the costs, the operating costs of large tech companies? That's an argument to be made. You could say, oh, if we increase H-1B1, you know, the number of them, then maybe GDP growth is better by a third of a percent every year. But, you know, that's a very different argument than saying that China is going to beat us to AI. Look, we can, we maybe, I don't know, maybe we have a different project. Maybe we have it, we call it the, the ultimate boatlift and we offer to relocate every single person from Taiwan and give them $10,000. It's like a one-time thing. And we, they have, we create the state of New Taiwan. We carve it out of New Mexico. We like terraform it. So it's nice and like, you know, wet with all the, with all the, with all the desalinated water. And you know, that's, that's okay. But like, seriously, no one is going to block the immigration of people working on the chip fabs in freaking Taiwan. Come on, please be serious. Like, like our industry, our semiconductor industry is already married to Taiwanese social networks. Right? Like the 2 biggest companies in the US, like AMD and Nvidia, both run by Taiwanese Americans who are each other's cousins. It's like, okay, they already know how to get here. Now, this isn't saying that we're not leaving talent like exceptional mathematicians, et cetera, on the table by limiting immigration from India. Sure. But Again, it's, it's the country does not yet have semiconductor expertise, right? So I think the real race with China, it's not a talent race. Like, who, who in reality thinks that the Chinese will be able to recruit the talent from the remaining 7 billion people that don't live in China? They suck at immigration. They suck at recruiting talent. They're never going to like, they're going to get a few handpicked Western researchers who are disaffected, and that's going to be it. Maybe a few Russians. The advantage China has in the semiconductor race is manufacturing. Okay. And, you know, the US has to get good.

That's— I love that idea. That's sort of the 10,000, almost like a Thiel Fellowship for immigration.

Thiel Fellowship. I mean, it could literally be a Thiel Fellowship. We can put Thiel in charge of the 10,000. Yeah, exactly.

That's a great place to wrap. Samo, this has been a great conversation on a wide-ranging set of topics. As always, until next time.

Until next time.

Upstream with Erik Torenberg is a show from Turpentine, the podcast network behind Moment of Zen and Cognitive Revolution. If you liked the episode, please leave a review in the Apple Store. Hey everyone, Erik here. At Turpentine, we're building the first media outlet for tech people by tech people. We're the network behind the show you're listening to right now. We have a slate of hit shows across a range of topics and industries, from our AI and investing cluster of podcasts to shows that drive the conversation in tech with the most interesting thinkers, founders, investors, and influencers, like Econ 102 with Noah Smith. We're launching new shows every week, and we're looking for industry-leading sponsors. If you think that might be you and your company, email me at erik@turpentine.co. That's E-R-I-K at turpentine.co, and let's partner together.