Latest / 80,000 Hours Podcast / Rebuilding after apocalypse: What 13 experts say about bouncing back

Transcript

Rebuilding after apocalypse: What 13 experts say about bouncing back · 80,000 Hours Podcast

• Lewis Dartnell: Just imagine you wake up tomorrow morning and civilisation’s collapsed and then disappeared. And you have to ask yourself, “What do I actually know how to make or do? How could I go about rebooting civilisation?” For many good reasons we would have to reboot civilisation — after some kind of global catastrophe, some kind of doomsday event or apocalypse — along a different trajectory, along different developmental lines than we did the first time around in our own history. What would you most want to whisper in someone’s ear — like 2,000 years ago, or if someone’s having to go through this process again — that once you’ve told someone, it kind of makes immediate sense? If you just tell someone the most useful thing to do or to try, you don’t have to stumble across that invention again serendipitously like we did in our own history — you can leapfrog straight to it, cut out hundreds of years of fumbling around in the dark. If you took the risk of catastrophic civilisational collapse seriously — and I think there’s good reasons to take that seriously — there are pragmatic, hands-on things we could be doing about that right now to dramatically increase the chance of a rapid bounceback, of a rapid reboot. Luisa Rodriguez: Hey listeners, Luisa here. You were just listening to Lewis Dartnell, who Rob and I interviewed back in 2022 about his book The Knowledge — a really deep dive into a thought experiment about how humanity might fare if the unthinkable happened and civilisation collapsed. Today we’ve got another compilation for you of our favourite bits from past shows. This one some of you might know I’ve done a bit of research on myself — as actually my first appearance on the podcast, I was a guest, where I talked with Rob about why global catastrophes seem unlikely to kill us all. So, you know: pretty lighthearted stuff! But I think you’ll be pleasantly surprised by how this episode is not only fun and fascinating from a scientific and sociological perspective, but also filled with many hopeful messages about human ingenuity and practical solutions to seemingly impossible challenges. Of course, it’s also dark, dark stuff to reckon with — so maybe keep that in mind when deciding whether to keep listening. In this episode you’ll hear a bit from me, plus 12 other past guests who take us through a range of civilisation-threatening scenarios, and exactly what we’d need to do to make ourselves more resilient if the worst should happen, like: Zach Weinersmith on why we shouldn’t bet on space settlement to save us from problems on Earth Toby Ord on the many risks to our atmosphere from supervolcanoes, asteroids, nuclear winter, and extreme climate change David Denkenberger on his team’s work to ensure we have resilient food systems in place to prevent widespread panic and famine Lewis Dartnell on how we could rediscover essential information in the aftermath of collapse Kevin Esvelt on the scariest pandemic scenarios that could quietly and quickly bring down civilisation Andy Weber on what people in the Pentagon think about nuclear winter Plus loads more! All right, I hope you enjoy! Zach Weinersmith on how settling space won’t help with threats to civilisation anytime soon (unless AI gets crazy good) Luisa Rodriguez: There are some arguments for settling space that you don’t think are very good. One is that space will save humanity from near-term calamity by providing a new home. Why do you think that is a bad argument? Zach Weinersmith: Yeah, I’m excited to talk to y’all about this one, because it’s in your wheelhouse. We’re talking about existential risk, right? So imagine you have the ability to shift funds around however you like to try to make it through the next century. The question then is: is space settlement on your docket? What’s your allocation? And I would say it’s probably zero. So pick your calamity, right? If your calamity is some sort of worldwide disaster like nuclear war or an asteroid, we calculate the amount of people you need to have a permanent Mars settlement is going to be enormous: probably on the order of a million people, maybe multiple orders of magnitude larger, to have a settlement that could survive the death of Earth — or loss of contact, let’s say. If you’re talking about climate calamity, massive global warming and the Southern Hemisphere is just uninhabitable, the thing to know is that Mars is still a million times worse. Mars is worse than a Superfund site. It’s a disaster. It’s worse than trying to live in Antarctica. And so unless that climate change is even beyond a worst-case IPCC scenario, it’s still a bad idea. And even then, if the goal was just to have a human pocket that survives, you should dig a hole or put a city underwater. It’s going to be much easier than doing Mars. So I don’t think the x-risk stuff applies. Luisa Rodriguez: If progress continues as it’s been, or maybe speeds up at the rate that you’d expect it to, based on historical trends, we shouldn’t actually expect to have a million people on Mars anytime in the next decades, maybe much longer. One reason I wonder if it might happen faster than that is if AI becomes as smart or smarter than humans and is therefore able to accelerate progress on some of these technological challenges. So right now, there are companies spending enormous amounts of compute trying to train really smart AI systems. But at some point, if they become really smart and we have trained systems that can do a bunch of science — which I think is the goal of many or some of these companies — then we can kind of redirect that compute to running many, many of these systems, potentially millions of systems. And if we threw just even a tiny fraction of that at space settlement as a scientific and social problem, you might think that not all, but some of these problems might get solved on much shorter timescales. There’s still the problem of, some of these things you can only get data on them at kind of biological time scales — humans need as much time as they need to develop, trees need as much time to grow as they need — but still, you might think that we can accelerate a lot of it, and AI might help us kind of optimise these experiments to get the kinds of data we need in the most efficient way. Does that sound at all plausible to you? Zach Weinersmith: Yeah. So I’m excited about this AI stuff. It’s tough because on the one hand, you want to, as we try to in this book, stick as closely to what’s known. We don’t like to do a lot of speculating. But you’re right: if the truth is that everything’s going to change completely in 20 years, that would suggest that some of the stuff we say is wrong. So a couple of thoughts. One is, yeah, you can imagine a world where, for example, we have in-silico drug development, and actually it turns out there’s a great way to halt bone loss and muscle loss. And hey, it’s safe for pregnant women and babies. That would be great. I mean, it’s possible. We don’t even know if such a drug exists, but it could be, and that would definitely change that aspect of things. Although, gosh, it’s hard to imagine a future where even if you say, in silico, this is totally safe for pregnant women, that it’s still ethical to do, you know? But I don’t know, maybe it’s a brave new world. I don’t know. The next thing is, if we’re talking about x-risk, then I am not sure how much that technological improvement reduces it. For example, people will often say to me, “There’s some really cool fusion stuff going on right now. It’s these neat new superconducting tapes and things, and AI is helping with plasma confinement. So maybe in 10 years we can send a plasma rocket to space and it’ll be able to either bring a lot more mass or get there a lot faster, or maybe both. Is that going to change your equation?” Absolutely. But on the other hand, one of the things we haven’t talked about yet is that one way you’re increasing existential risk is just by putting massive amounts of big heavy stuff in space operated by many players. So the math I like to give on that is: if you have an object moving at three kilometres per second, when it impacts it has an explosive yield equivalent to it being made entirely of TNT. Objects in low Earth orbit go more than twice as fast, and the yield is the velocity squared: it scales with the square of the velocity. So if you imagine a world where every country has access to millions of tonnes of space objects — like many corporations do, individual rich people do — that’s just a world of enhanced existential risk. And I don’t know that there’s a way to avoid that unless you also want to postulate like tractor beams and weird, really spacey stuff. So if you strap a plasma rocket to that, it gets there faster, but you haven’t obviously reduced your danger. So, at least if the question is colonising another planet to reduce existential risk, it’s not obvious to me that adding sci-fi elements, even if they’re real, makes you safer. The last thing I want to say is: when we look at numbers, we try to look at numbers of how many humans you’d need for one of these operations. And there’s a lot of debate, there’s questions like genetic inbreeding stuff, but the bigger question is: can you go economically independent? The lowest number we found was from what I would say is a very optimistic analysis done by Dr Casey Handmer. And the balance he came up with, which assumed big advances in AI robotics, I think the lowest number he was willing to do was 100,000: 100,000 humans to have a permanent settlement. So my response to that is, one, at least short term, there’s no way — you know, depending on what “short term” means. But I would also say, to me, if you’re saying we can do Mars just as soon as every person has the equivalent of 100 robot butlers, we are talking about Mars not as a kind of destiny thing, not as a frontier thing, not as a human enhancement thing: we’re talking about it as the aesthetic choice of an advanced humanity. Right? It’s Star Trek. And so to me, that’s the deep answer here. I mean, we’re worried about the existential risk aspect, but you know, if I’m in a world where we all get 100 butlers, I’m probably staying home. I’m glad there are people who are ready to go to Mars. I will be walking my orchard, being fanned and fed cupcakes. And the last point I want to make — and this ties to the x-risk thing — if that sounds very tempting to you, and it does to me, then you need to make sure humanity makes it through the next century, or the next 20 or 30 years. And as we agreed earlier, if your goal is to reduce existential risk, it’s really not obvious that space is the way to go. So you’re absolutely right: I want to absolutely admit AI has gotten so weird in the last few years, I don’t feel like I have any kind of prediction capacity for what five years looks like, at least in terms of LLMs and stuff. But if we say that if you have it for Mars, you have it back on Earth, I think it at least changes the nature of the question. Luisa Rodriguez on what the world might look like after a global catastrophe Rob Wiblin: So the big thing we’re here to talk about is this series of articles you’ve written on the probability of humanity ultimately recovering and building back from a serious collapse of civilisation. I guess not many people have, as far as I know, considered that in a sober way. It seems like an area where there’s a lot of loose talk and maybe not a lot of hard talk. How did you first get into this topic? Luisa Rodriguez: Yeah, exactly. I got into the topic after thinking about nuclear war, where I’d done this big project thinking about whether a nuclear war would cause a nuclear winter and whether a nuclear winter could cause human extinction. And I ended up feeling like it causing human extinction was pretty implausible — it’s just really hard to tell a story of how that actually happens. It still seemed bad to me if nuclear winter caused civilisational collapse, but then even when I reflected on that, collapse isn’t necessarily as bad as human extinction, at least on its face. And then I realised I actually just had no idea if it was as bad as human extinction. I didn’t know if it would necessarily lead to it. I didn’t know if it was completely implausible that it would lead to it. And then it just started to feel like this gaping hole in a lot of these global catastrophic risks — that get you part of the way to extinction, lots of people die, and it’s horrible. But I couldn’t find any really plausible stories, at least right away, for thinking that it was as bad. Rob Wiblin: What sort of different disasters did you analyse? Luisa Rodriguez: So I found that it did kind of matter, the type of disaster that you’re talking about, to one, how likely it is that civilisation would even collapse, but also, two, how likely it is that civilisation would recover from collapse. So I tried to take a few examples — just to keep the scope narrow enough to even think about — that got at some of the key parameters of a catastrophe. I didn’t quite hit all of the ones I wanted, but the main three I considered were: a scenario where 50% of the population dies. You can imagine it as a pandemic. So you have high population loss — not extremely high, but much higher than, for example, COVID — but no loss of physical infrastructure. Rob Wiblin: The houses are still here. Luisa Rodriguez: Yeah, yeah. And basically the climate is the same and there are no lasting environmental effects too. So that’s one case where you’re just kind of toggling up and down on population death. Another that I thought about was adding in infrastructure damage, and because nuclear war seems like a plausible catastrophic risk, I went ahead and just tried to think about nuclear winter in particular. So for that one, I thought the parameters would be something like, maybe 90% population death, through both the initial war, and then famine in the very immediate aftermath of the nuclear winter. And then I also thought there would be something like five to 10 years of environmental effects, but then those would go away. So that was another one. And then the last one I was trying to come up with basically a set of catastrophes that might make the story look worse. And the thing that I’ve actually written up and posted, I ended up thinking about nuclear war and biological weapons. And so in that scenario, I ended up assuming that population death would be much higher — so 99.99% — which I think leaves you with 800,000. Rob Wiblin: 1 in 10,000? So 800,000 people left globally? Luisa Rodriguez: Yep. And then you still had the infrastructure damage and you still had the environmental effects that were temporary. So that was one of the worst cases I could think of. And then one that I’ve thought about since writing publicly about the topic, so maybe a fourth one, is what happens if you have a nuclear war that kills lots of people, but then you also have climate change or climate effects that do last for a really, really long time — on the order of not a decade, but thousands of years. So those are the main parameters I see being important: how many people die, whether infrastructure is left standing, whether there are environmental effects, and how long all of these effects last. Regardless of the scenario, a thing that kept coming up as important was the fact that catastrophes will kind of inevitably be non-uniform in their effects. So I guess if you have a catastrophe that’s so big that it’s actually uniform in its effects and it’s really severe, it’s going to kill everyone. So that would be a catastrophe like an asteroid that actually really impacted the Earth. Rob Wiblin: It’s like even bigger than the one that killed the dinosaurs or something like that? Luisa Rodriguez: Yeah. So that is one where you can imagine having consistent mortality rates everywhere, consistent climate effects everywhere. But for the catastrophes that interest us — so the ones that don’t actually kill everyone at once, but leave some survivors — the reason that happens is because the catastrophe is going to have non-uniform effects. And I think for lots of catastrophes, they could just be very non-uniform, where you might get some entire continents that are much, much less affected than others — both in terms of population death and in terms of climate effects. So an example in nuclear winter that’s kind of well known is you have some continents where agriculture becomes near impossible, and then you have others where it’s maybe even a bit better, because it’s colder and otherwise it was too hot for agriculture. Mostly you wouldn’t get better. It just wouldn’t be quite as devastating as other places. Rob Wiblin: So this non-uniformity is super central, if you’re thinking about if it will kill 100% of people. Because even if there’s just like 1% of people living in some place that’s largely unaffected, then that basically answers the question for you. Luisa Rodriguez: Exactly. And I think sometimes it’s slight non-uniformity that makes a bit of difference — maybe isn’t decisive. So like with nuclear winter, you’ll have some areas that are cold and some areas that are slightly less cold. But some catastrophes would cause extremely non-uniform effects, where even if you had really, really enormous population losses, and actually the collapse of society and political systems — all of these systems that we think of as critical — collapsed on one continent, I think you might see society continue on others. And I think that’s something that I didn’t intuitively have in my head when I first started thinking about this, is just some places might really be kind of unfazed. Rob Wiblin: Right. Thinking about New Zealand again, I suppose also Tasmania. Are there any other things that we should have in mind? I guess it’s like Pacific islands. Luisa Rodriguez: Pacific islands. Yeah, depending on the thing again, in some scenarios you really want coastlines because fishing is a key source of food. So Chile looks kind of good in a couple of scenarios. I think those are the main ones. I think a really interesting emergent finding is that the number of survivors really does interact in a kind of funny way with other things. Supplies left is a good example: if you have lots of survivors, the supplies go very quickly. But on the other hand, if you have lots of people dying, which would be terrible in some respects, and makes it a bit harder to rebuild industry and harder to make sure that you… Well, ideally that you’d keep some of the necessary knowledge and skills — you’d want to at least eventually rebuild industry. When you are down to tens of thousands or hundreds of thousands of people, it’s less guaranteed that you’ll keep all of those skills. So you just get these kinds of tradeoffs. And I think, mostly, I felt like it worked in favour of survival. Rob Wiblin: This kind of relates to this concept that you’ve called the “grace period”: this temporary time after some disaster where a bunch of people are now dead, but you still have this overhang of supplies from the pre-apocalypse world. And other things like cars are around, other infrastructure, so this potentially helps you to stick around and then rebuild. Do you want to elaborate on that? Luisa Rodriguez: Yeah, sure. So again, it kind of depends on the catastrophe, but at least in some places, in most catastrophe scenarios, you have maybe all of the infrastructure that you did otherwise: you have a power grid that still exists (even if it’s not working), you have grocery stores with food in them, you even have petrol stations that still have petrol in them and that you can siphon out pretty easily. So you have things that mean that you can kind of survive in a reasonably easy and accessible way. And just how long that period lasts, again, depends on how many people there are. But I think you’d be surprised how much stuff you can still access, with the limitations being the power grid will stop working. But even water will still run for at least a while, and even maybe for a long while in some places, where the water supply isn’t run by electricity. Rob Wiblin: Just gravity. Is there any way of giving a sense of how long it is, or what kind of stuff might we run out of first? Luisa Rodriguez: So if you have 50% population loss, I think that, theoretically, if you actually allocated all of the supermarket food and grain stocks and water — kind of literally rationed it — then it would only last days, like under a week. If you actually just divide the US, China, some of the really big food suppliers, over everyone. We used to have a much greater volume of food stocks than we do now, especially during the Cold War, kind of unsurprisingly. But currently, lots of countries have about six months of grain reserves, and then lots have basically none. Rob Wiblin: So that means that if you have a situation where 99% of people suddenly die, then now you can last years with that kind of food. Luisa Rodriguez: Yeah. It’s years if you make a few assumptions about how you prioritise the order in which you eat things. David Denkenberger on the catastrophes that could cause global starvation David Denkenberger: There are a number of catastrophes that could disrupt agriculture globally, and the most extreme ones could basically collapse agriculture. These are the ones that could block the sun: they include a large asteroid that caused the extinction of the dinosaurs, a supervolcanic eruption like many people think almost caused the extinction of humans about 70,000 years ago, and a nuclear winter and this would be caused by what’s called full-scale nuclear war involving thousands of nuclear weapons between the US and Russia. This causes burning of cities and smoke goes into the atmosphere and it can remain there for a decade. In these scenarios, since almost all of our food comes from plants that need the sun to grow, it’s generally assumed that most people would die. We only have a few months of food storage. People who have looked at this problem before have suggested, let’s just store up more food. And that would be technically possible, but as a way of visualising it, you can think of a 150-litre or a 40-gallon oil drum: if that’s full of dry food, that can feed a person for a year. But to have enough food to feed 7 billion people for five years, you would need to pile those drums from the Earth to the Moon and back 40 times. As you can imagine, that would be very expensive, many trillions of dollars, and you can’t do it really fast. It would take a while to store up that food, but you would want to do it as fast as practical so that you’re ready to weather disasters that could happen at any time. But if you store it up fast, then you inflate the food price and then millions of people more would die of malnutrition than already occurs. I wasn’t satisfied with that not very good solution. I was thinking, well, is there another way of producing food when the sun is blocked? In 2011, I was reading this paper called Fungi and Sustainability, and the premise was that after the dinosaur-killing asteroid, there would not have been sunlight and there were lots of dead trees and so mushrooms could grow really well. But its conclusion was that maybe when humans go extinct, the world will be ruled by mushrooms again. I thought, why don’t we just eat the mushrooms and not go extinct? I branched out from mushrooms, thinking of all the ways that we could convert either dead vegetation into food or even fossil fuels into food. Rob Wiblin: Yeah. I just love this idea. Whether it’s practical or not, I’m just so fond of it because you’ve taken a problem that most people have just thought of as inevitable and basically accepted that if there’s a nuclear war, then maybe billions of people would die of starvation. And I guess bringing an engineering mindset that you’ve been trained in, you’ve just seen this as a practical challenge to overcome and pretty quickly come up with a bunch of seemingly plausible ways that we could actually just feed everyone even if the sun were completely blocked out, which is amazing. David Denkenberger: Yeah, that’s right. I do think that that engineering mindset is really important — in that certainly, it’s important to know what the problems are and ideally prevent them, but realistically, they still could happen. I think that there should be more work done on this resilience end of the spectrum. Rob Wiblin: Who’s working on this problem? I guess there’s some people who would try to stockpile food and presumably some countries have food stored away. How much is that and what else are people trying to do? David Denkenberger: There are certain countries that have done a lot of food stockpiling. One of them is Switzerland. I believe they have around a one-year food supply. I think China does as well, though it’s not public knowledge so you don’t know exactly. But on average, it’s more like a few months across the world. You’ll often hear numbers of we have two months of grain supply. That’s actually about right, but for the wrong reasons. It’s at current grain consumption, but of course we feed half of our grain to animals. If we did have one of these catastrophes, hopefully we would be not feeding as much edible food to animals, so that helps. But there’s also the fact that most of our stored food is grain because it’s really easy to store dried, and half of our calories comes from food that is not grain. We have less food storage that is not grain. It works out to just a few months of food storage, and it depends on what time of the year, but that’s the minimum. Rob Wiblin: Maybe let’s just dive into thinking about the nature of the problem that you’re solving and the concrete solutions that you’ve come up with for it. Not all listeners I imagine will be convinced that a mass starvation of humanity is a terribly likely risk. What are the major possible causes of a global food shortage and how likely might they be? David Denkenberger: I mentioned asteroid and supervolcano, but as the existential risk community has pointed out, generally the natural catastrophes can’t be that bad, because they can’t be happening every century or we would have been doomed long ago — though of course there is the possibility that we’ve been extremely lucky, which people have tried to adjust for. I think still, even adjusting for that, the natural catastrophes are lower probability. Rob Wiblin: Do we have any estimate of the probability of an asteroid or supervolcano that would cause a risk to human food supply? David Denkenberger: Yeah. If you’re talking about blocking the sun, then it really is around one in 10 million years or 100 million years for an asteroid. And the volcano is more likely, more like one in 100,000 or a million years. Rob Wiblin: How robust are those estimates? Because I know we’ve had a couple of volcanoes erupt and changed the biosphere during the million years or so that humanity has been around. David Denkenberger: The latest estimate I saw on the supervolcanoes was actually coming up with a larger number that they happen more frequently. Apparently it’s difficult to identify further back than a few million years. There’s some uncertainty there. There’s also uncertainty whether the supervolcanic eruption would actually block the sun well enough and long enough to cause mass starvation. It’s the same with nuclear war as well. I’ve written a paper that tries to go through all the uncertainties of, well, how many nuclear weapons would be detonated? Which cities would be hit? What fraction of the combustible material actually burns quickly? How much of that turns into smoke? How much of that smoke makes it into the upper atmosphere? How much do the particles block the sun? How much does it impact agriculture? As you can imagine, there’s uncertainty in all of those steps, so it adds up. Rob Wiblin: What about the more moderate cases where there’s a more normal war or a regional nuclear war. How much would that interfere with the food supply? David Denkenberger: That’s the class of catastrophes that I’ve been labeling the order of magnitude 10% global food production shortfall. Roughly 3% loss to 30% percent loss. In those scenarios, there are a number of catastrophes that could cause that. As you mentioned, a regional nuclear war between India and Pakistan might only involve 100 total nuclear weapons, and actually much smaller nuclear weapons than the US and Russia have, but still they would be targeted at highly populated cities. Some work on this has estimated something like a 10% or 20% global food production shortfall because of it. You can think of smaller versions of the natural catastrophes like asteroid and supervolcano, but again, they’re not too likely. Another one that’s received some interest recently is called the coincident extreme weather or multiple breadbasket failure. Here the scenario is you have droughts or floods on multiple continents at once. There was a UK government study on this that estimated right now, it might be around 1% chance per year, but with the slow climate change, that extreme weather probability, so difference between climate and weather, actually gets more likely. They were getting more like 80% chance of this century that something like that would happen. Rob Wiblin: Wow. OK. David Denkenberger: There are other scenarios as well, like the superweed — which is not the saviour of medicinal marijuana users, but is instead a weed that outcompetes crops. If this were a natural thing, we could probably contain it or at least slow it down so that it wouldn’t happen too fast. If we have more time in any of these, it makes it less extreme, but it could be an actual coordinated terrorist attack. There have been some examples of people trying to use I think more a crop disease, but that’s another category, something that would kill crops directly could be a coordinated attack as well. Rob Wiblin: A superweed sounds a bit outlandish, because how would a weed spread all around the world really quickly? Wouldn’t a super pathogen like a virus or bacteria be more able to spread to lots of crops quickly enough that it’s hard to respond? David Denkenberger: Yeah, right. It’s hard to have something that can live in many different climates, yet I’m particularly worried about crop disease. If something could target the grass family, then you affect not just grass that feeds a lot of our animals, but also wheat, corn, maize, rice, and sugar cane. If you add up all of the human calories, it’s something like two-thirds of our calories comes from the grass family. That really could be catastrophic. Rob Wiblin: You mentioned in the book that a coordinated terrorist attack using superweeds to destroy a lot of crops across a very large region could be a global threat that not that many people are thinking about. Do you have any evidence that a group might actually plan such an attack or has been considering planning such an attack and do you know if anyone else is working on trying to figure out if this is a real risk and what might be done? David Denkenberger: I have not heard of evidence for this particular type of risk, though a related one is crop diseases. These have been used in biological warfare programmes, so there is concern that this could be a potentially larger attack. As for the probabilities, I haven’t seen anything quantitative. I just try to quantify what I can and then say, we’re not including all these other things, so really the risk is higher. Rob Wiblin: Are there any just non-nuclear wars that you think would create scenarios that are plausible? Who could end up fighting without nuclear weapons? If you had a major war, wouldn’t that massively interfere with trade, and so some countries that are currently importing most of their food could well end up starving pretty fast? David Denkenberger: That’s right. The other scenario that could really interfere with trade is a pandemic. If it were a severe one, then it may be rational to close borders to reduce transmission, but you’re right, importing-food countries would be in big trouble. Or it could be a smaller pandemic and just people overreact. It might not be rational, but they still might close the borders. There are a couple of other scenarios that are similar to this idea of a 10% global shortfall. One of them is called abrupt climate change. This is like Europe going back into the Ice Age, which some people have talked about because of breakdown of the thermohaline circulation, circulation of the ocean that’s driven by salt. That could be a 10% shortfall. Another one that people have talked about, and some existential risk researchers are looking at, is extreme climate change. It’s global but it still happens slowly, like over a century. That’s a little different than the problems I’m mainly focusing on, but I think some of the ideas could be relevant to that. Lewis Dartnell on how we could rediscover essential information if the worst happened Rob Wiblin: So we’ve got this big common interest in understanding how humanity would recover from a major catastrophe that knocked us back a long way. Collecting a bunch of information that would be useful for survivors of a disaster was the topic of your book, The Knowledge: How to Rebuild Civilization in the Aftermath of a Cataclysm. To get straight to the most important issues here: what’s one thing you think humanity will find especially difficult — oh, sorry — would find especially difficult about recovering? Lewis Dartnell: Do you know something that I don’t know? Rob Wiblin: Yeah, sorry. Knock wood. Would find especially difficult about recovering to our level of technology, say after a pandemic that killed 90% of the population, or something around that level? Lewis Dartnell: Let’s just imagine you wake up tomorrow morning and civilisation’s collapsed and then disappeared. And you have to ask yourself, “What do I actually know how to make or do? How could I go about rebooting civilisation?” — in the way you’d reboot a computer after it’s crashed? And as you’re asking about, for many good reasons we would have to reboot civilisation — after some kind of global catastrophe, some kind of doomsday event or apocalypse — along a different trajectory, along different developmental lines than we did the first time around in our own history. What I try to get at in The Knowledge was using this thought experiment — this playful hypothesis, this scenario of the post-apocalyptic world — to answer as a genuine question, but by doing so, looking into our own history and how we got to where we are today, and why the world that we live in looks the way it does. One of the ideas I played with in The Knowledge was what would you most want to whisper in someone’s ear — like 2,000 years ago, or if someone’s having to go through this process again — that once you’ve told someone, it kind of makes immediate sense. Or you give them a very simple set of instructions for how they can make something or build something or demonstrate something for themselves. And for me, the one that stood out by far the most significantly was this idea of germ theory and how that links to the microscope. Imagine the centuries and centuries and centuries of human suffering through history, because we didn’t have the right idea about why people got sick, and why they died, and why plagues seem to spread very quickly through cities and from person to person. So if you told people that the reason people get sick isn’t because of bad air — mal aria, from the Italian — and it’s not because some fractious God has smited you. It’s because there are things which are invisibly small, they’re so tiny you can’t see them with your naked eye, but they’re there and they get into your body and they multiply and you pass them onto one another. But tell you what, this is how you make glass from scratch. And I give the recipe in the book. And actually, one of my favorite maker projects when I was researching for The Knowledge was making some Robinson Crusoe glass from scratch. I went to a beach and I got sand and seashells, chalk, and soda ash — sodium carbonate — and made some glass from scratch in the course of a weekend. Which you could then fashion into a lens to manipulate and control light, and then build a microscope from it. And there’s nothing stopping the ancient Romans over 2,000 years ago building a microscope, if only they’d known what to do. If you just tell someone the most useful thing to do or to try, you don’t have to stumble across that invention again serendipitously like we did in our own history — you can leapfrog straight to it, cut out hundreds of years of fumbling around in the dark. And perhaps the best way of doing that would be to build these repositories of human knowledge, something like a manual for civilisation, or a total book: a book that contains the sum total of human knowledge, but is also organised in a way that is useful and progressive, holding your hand and leading you through the steps of the ladder. Unlike something like Wikipedia, which is an absolute mess of information just dumped in there. You might then have these repositories of the sum total of human knowledge dotted around the globe. Maybe you have some big conspicuous markers that point post-apocalyptic survivors to where their local library for rebooting is. I appreciate that this is starting to sound a lot like sci-fi. And this sort of idea has been explored really well in some cracking books. But I think that’s an intriguing idea. It is something that if you took the risk of catastrophic civilisation collapse seriously — and I think there’s good reasons to take that seriously — there are pragmatic, hands-on things we could be doing about that right now to dramatically increase the chance of a rapid bounceback, of a rapid reboot. Andy Weber on how people in US defence circles think about nuclear winter Rob Wiblin: I’ve had this kind of longstanding confusion about the issue of nuclear winter. My impression is that there’s some natural scientists who come at this question from far outside the national security community who think that it’s a real threat and a real likelihood — though admittedly, it seems like it’s a pretty small community of people who study climate and study physics and so on who have looked into it in any great detail. But I get the sense that there’s more scepticism, or just kind of disinterest maybe, from people who are approaching this from the national security or nuclear weapons point of view, that it’s not talked about that much. And relative to the stakes that seem to be on the table — where, you know, there’s a question of if we use nuclear weapons, would it lead to billions more people dying because of this nuclear winter, it would seem like it’d be a really high priority. But it doesn’t get a lot of takeup, as far as I can tell, from the Pentagon. To what extent do people in US defence circles take nuclear winter seriously? Do they think about it at all? Andy Weber: We do, and we do take it seriously, but you’re right, there has not been as much investment in recent years in the types of research that led to the conclusions in the 1980s regarding nuclear winter. And our modelling capabilities have gotten so much better, so we should invest again in those types of studies. But there is no question that a massive use of nuclear weapons would cause a nuclear winter. Rob Wiblin: As I understand it, the crux of the disagreement is how long it would last, which comes down to how high the particulates from the burning cities and forests go. Because if they go high enough, then they tend to just stay up in the upper atmosphere for a very long time. But if they’re in the lower atmosphere, then they kind of rain out and fall down. Have I got that right? Andy Weber: All of that can be modeled. And the conclusions of the work that was done previously have not changed. So this is a real concern, and this is why this is just one of many reasons — I mean, let’s talk about the lives that would be lost with the use of any nuclear weapon, people who would be killed immediately. So we need to make sure that nuclear weapons are never, ever used. Rob Wiblin: Yeah. So you think there’s not that much dispute. So obviously it would have a large effect on the weather, but you think we’re talking like many years and a lot of people starving, that that is the most likely outcome of a major nuclear exchange. Andy Weber: Oh, absolutely. Absolutely. It would be a near species-ending event. Rob Wiblin: And do other people who’ve worked in the Pentagon or people who think about nuclear policy, is that their view as well? Andy Weber: I think it’s generally accepted. Yes. It’s not something we dwell on or talk about very much. And like I said, there has not been much work funded in this area in recent years, but I think it’s general conventional wisdom that a large exchange of nuclear weapons would cause a nuclear winter. Rob Wiblin: It’s good to know that people believe it. It seems better to maybe err on the side of caution with that one. Toby Ord on risks to our atmosphere and whether climate change could really threaten civilisation Toby Ord: So the mechanism from supervolcanoes and from asteroids and from nuclear war, the main mechanism for causing existential risk is via a nuclear winter or volcanic winter or asteroid winter — where particles get up into the stratosphere so high they can’t be rained out, and then they cause global cooling: cooling, darkening, and drying. But it’s the cooling that’s the main one, because it shortens the growing season for crops. So that’s the main concern. And interestingly, for all three cases, it is a form of climate change and it is mediated by atmospheric science, which is the subject that studies this. So if you look at the size of these asteroids, 10 kilometres is very big. It’s the size of a mountain. But it’s very small compared to the Earth. And the kind of image you might have in your mind of the asteroid ploughing into the Earth, it is more of a pinprick than two things of similar size. Rob Wiblin: But so much dust. Toby Ord: Yeah. But it is the dust. I guess these types of people who, you know, which would make them a climate denier, they could deny that there would be these dust effects and things. And in the case of nuclear winter, there was a lot of denying of this. There’s a lot of pushback that Carl Sagan and others received on the theory. Partly because it was a politicised issue, somewhat like we’re seeing with climate change. So one could push back on supervolcanoes and asteroids for the same reason, but you don’t see that so much because it’s not politicised. Arden Koehler: It’s interesting that so many different risks share this same mechanism. It suggests that one of our biggest vulnerabilities is our atmosphere, or our access to sunlight. Toby Ord: Yeah. That’s right. And there’s a useful way of thinking about this, which is that once there’s some kind of event, is the event so big that it just would obviously destroy the Earth? For example, if an entire planet crashed into the Earth or something, it’d be pretty obvious how it gets big enough. But in other cases, there’s this question about how does it scale up to be something that could threaten us all? How does it get everywhere, like to all the humans and so on? In the case of all of these things, what happens is that the atmosphere is what takes 1,000 kilometres of rock or square cubic kilometres of rock or what have you, and distributes it in such a way to create this opaque layer around the Earth. And without the atmosphere doing that, it would be more of a regional catastrophe. Then the atmosphere is also important in climate change, and also temperature changes are also important there, and the effects of temperature change potentially on crops. So there’s actually quite similar things about some of these natural catastrophes and even some anthropogenic ones that are quite interesting. Rob Wiblin: Did your opinions change at all on how resilient we would be to these changes? I suppose at the moment it seems like human ingenuity is winning out. The climate’s heating, but we’re getting so much better at farming all the time that the amount of food output just keeps rising at a pretty good clip. So is it possible that we will just be able to adapt to this because it’s happening over decades? Toby Ord: I think so. It would still be much worse than if it wasn’t happening. Just to be clear on that for the audience. Rob Wiblin: We’re talking here about like would we all die, or would it cause the collapse of civilisation, which is a high bar. Toby Ord: That’s right, it’s an extremely high bar. And while there are a lot of things which could very clearly cause a very large amount of human misery and damage, it’s quite unclear how it could cause the extinction of humanity or some kind of irrevocable collapse of civilisation. I just don’t know of any effect that could plausibly cause that. There has been some analysis of if you had very large amounts of warming, such as 10 degrees of warming, would it start to make areas of the world uninhabitable? And it looks like the answer is yes. At least being outside. Air conditioning could still work. It’d still be much more habitable, say, than Mars. People are perhaps thinking of setting up settlements. But also that argument, if you run it through, it really just suggests that the habitable part of the world would be smaller. So coastal areas are much less affected. High plateaus such as Tibet wouldn’t be moved to super hot temperatures. So there would still be many places one could be. It would be a smaller world. And it seems hard for me to think that, given it wouldn’t be that much smaller, why then civilisation would be impossible, or a flourishing future would be impossible in such a world. That just doesn’t seem to have much to back it up at all. Arden Koehler: Even if it was a third of the size. Toby Ord: I mean, if we heard that someone had found a planet in the habitable zone around a nearby star, but it had a lot of ocean and only had a third of the land mass of the Earth, we wouldn’t think, “No need to worry about ever meeting anyone from that planet, because it’s impossible to create a civilisation on such a planet.” Or say it was only the Americas, and you didn’t have Africa or Eurasia or Australia: “Obviously, you never could have had civilisation there, or you could never sustain it.” That would seem kind of like a pretty crazy view. So I don’t really buy the idea that large enough parts of the Earth could be made uninhabitable either. Arden Koehler: Well at degrees of warming like 10 or whatever. But if we get up really high? Toby Ord: Yeah, I looked at these models up to about 20 degrees of warming, and it still seems like there would be substantial habitable areas. But it’s something where it’d be very bad, just to be clear to the audience. Rob Wiblin: Most people are dying. Toby Ord: Yeah, it’d be very bad. But it’s hard to see any particular mechanism that’s being floated as to how it would happen on model. But my concern is more than just the prior probability. Before you even got into these models or got into the science of it, if we make an unprecedented change to the Earth’s climate — perhaps at a truly unprecedented rate over the last 4 billion years, and also to a level which has only a couple of times been reached or something and never been reached with the current configuration of continents or with a species like us and so on — that it does seem like there’s just some plausible chance that this is the end. It’s not that if you imagine kind of appearing before Saint Peter at the pearly gates and he said, “Hey, yes it was climate change” and you’re like, “How could we have possibly known that making these radical changes to the Earth’s climate that haven’t been seen for millions of years could do us in?” I think we’d be looking pretty foolish, even if we said, “But our scientists looked at these different pathways and none of them could lead to it.” And you’d think, “Well, it could have been one that you hadn’t thought of.” In the case of nuclear war, for example, nuclear winter hadn’t been thought of until 1982 and 1983. So that’s a case where we had nuclear weapons from 1945 and there was a lot of conversation about how they could cause the end of the world perhaps, but they hadn’t stumbled upon a mechanism that actually really was one that really could pose a threat. But I don’t think it was misguided to think that perhaps it could cause the end of humanity at those early times, even when they hadn’t stumbled across the correct mechanism yet. Arden Koehler: Because it was just an unprecedented event? Toby Ord: Yeah, and there hadn’t been that many people searching for such mechanisms, and they ended up kind of getting there from thinking about other planets. Planetary exploration made people think about how very different atmospheres worked and to get some kind of data on what it’s like to have radically different atmospheres or dust storms throughout the whole Martian atmosphere and things like that. And that made them think about this. But you could easily imagine them just never having noticed that mechanism, actually, since the Cold War ended shortly after that. And so I think that this is just the kind of thing that on priors, it’s such a big change. But I want to stress that my best-guess number for the chance of existential risk due to climate change is about one in 1,000 over the century. And that’s mainly coming from this kind of, “I don’t know the mechanism, but that our models aren’t sufficiently good.” Rob Wiblin: Some real x-factor. Toby Ord: Yeah. I think that it’s often suggested that climate change might not be so much an existential risk, but that it’s something that would increase other existential risks. So in this case, my terminology would be a risk factor. I think that this is probably right. If we could just somehow have the next century but make it so that climate change wasn’t an issue, all of the dedicated altruists who are working on fighting climate change could then work on other things, and global international tensions on this would go down — so nations could spend their “altruistic international cooperation” kind of budget on something else. So I do think that that could actually be quite helpful. As to how big it is as a risk factor, my guess would be somewhere between roughly about 0.1% and 1%. So maybe a bit bigger as a risk factor, but probably not a whole order of magnitude bigger. Rob Wiblin: So you think it’s quite a bit less important than a great power war? Toby Ord: My guess is that it is less important from the perspective of existential risk reduction. Arden Koehler: It sounded like some of the main mechanisms you were thinking about by which this could be a risk factor is basically that it distracts people. So the budgets of these governments and of organisations and people’s personal careers will be spent on it instead of on other things that you think might be more important ultimately? Toby Ord: Yeah. I think distracts is kind of right, but it has the wrong emphasis or something — because you think distraction can’t be that bad. Maybe a better way to think about it is this is a stressor on national and international relations and so forth. Rob Wiblin: And our capacity to solve problems. So our capacity gets used up trying to solve this thing and then we don’t have headspace to think about something else. Toby Ord: Yeah, that’s right. Arden Koehler: What about if some moderately high level of warming comes about, such that maybe this actually just ultimately falls into the bucket of reducing our capacity to solve problems, but it seems like if health systems and economic systems suffer a lot, it could leave us more vulnerable to things like pandemics, naturally occurring and engineered? Does that seem plausible? Toby Ord: Yeah, I think it’s quite plausible that it could leave us more vulnerable to pandemics. Also the fact that effectively a larger part of the Earth would be in a tropical environment. I think that this is something that is certainly recognised that there could be more endemic disease and maybe more pandemics as well. Arden Koehler: One thing that what you’re saying suggests is that maybe some of the most serious ways in which climate change or something else could be a risk factor is by impacting the other bigger risks. So you know, even if you think there’s a plausible mechanism for increasing some other existential risk that we can think of, it really matters how big that other existential risk is for how much that translates into being a risk factor. Toby Ord: Yeah, and so I think it may even be the case that, say the median level of climate change, like the stress that that creates on international institutions and governments and so forth, that that’s large enough to change the risk of, say, the biggest risks such as AI or engineered pandemics, to increase them by a 10th or something like that compared to if we definitely could just not have to worry about all of these challenges of climate change. That could be a mechanism that produces a significant amount of risk as a risk factor. But it’d be interesting to see some robust conversation about that. This is just me kind of sketching out some combinations of numbers, where I find it a bit hard to see how it would really work. But the people at CSER, the Centre for the Study of Existential Risk in Cambridge in the UK, are quite concerned about this and they think that climate change is a much bigger existential risk than I do. And they think this is largely through risk factors. Largely also through things to do with the collapse of civilisation. But I think that better characterisation of the chance of extreme warming and better understanding about how bad it would be and could we survive the extreme warming, and could there really be cases or blue sky thinking about… We understand a bit about various mechanisms which would cause the kind of central case of damage that we’re thinking about when economists model a damage function of warming and they’re thinking about extra disease burden, extra kind of adaptation, crop failure and so on. But rather to think, are there any things that, like in the case of nuclear winter, some really quite different mechanism which could cause a different kind of threat that only happens when it gets to a very high level? Blue sky thinking about that could be extremely valuable and could then help us much better understand how much of a risk climate change poses. Mark Lynas on how likely it is that climate change leads to civilisational collapse Rob Wiblin: How likely do you think it is that climate change will lead to the breakdown of industrial civilisation as we know it today? And secondarily, how likely might it be to effectively cause human extinction? I guess probably those numbers are going to be pretty different, they’re two fairly different standards. Mark Lynas: Well, for me the question isn’t quite in the way you framed it; the question is: At what degrees of warming do these outcomes become more likely? I remember when Extinction Rebellion, some spokespeople were in the media saying, “Billions are going to die in the next 10 years.” And I think it was Andrew Neil actually might be, who in the interview said, “Well, how? What’s going to happen that’ll kill all these people?” And they weren’t able to answer. So it’s one thing to have an apocalyptic fear and another to actually try and think through what the mechanisms are which would actually kill people, if that’s what you’re concerned about. What kills people? Thirst kills people. War kills people. Anger kills people. People die because of lack of shelter. So it’s different degrees of civilisational collapse. I mean, war is a civilisational collapse in a way, particularly if it’s essentially worldwide. To me, I think the most concerning scenario is one where you can’t produce enough food to sustain the world’s population, which is a lot higher. We learned that by 2050, the latest UN figures are 9.7 billion by 2050. But you have to have pretty much doubled the world’s food supply at the same time, as you’ve got less and less of the world’s land being able to produce because of the combined impacts of heat and drought. And obviously, plants can’t grow without water. Lots of plants can’t grow if it’s too hot. I mean, you can push the thermal tolerance threshold of crops with genetic engineering to some extent, but there comes a point where enzymes get denatured and your plant will just die, or certainly a long time before that yields will fall off, and so on. So for me, that’s the main question. And one of the most important studies I think that’s ever been performed on this was a study in the PNAS Journal, which looked at what they called “synchronous collapse” in breadbaskets around the world. At the moment, the world still produces enough food every single year very reliably. We’ve never had a major food shortage which has been as a result of harvest failure. So if the US Corn Belt was knocked out one year, that would have a huge impact on food prices, and have a huge impact on food security, in fact, as a direct result of that. But imagine if it really wasn’t just the US Corn Belt. It was Australia, it was Brazil, and Argentina, it was breadbaskets of Eastern Europe, and the former USSR, all of that added together — then you enter a situation which humanity has never experienced before, and which looks very much like famine. So that’s how I answer the question: how are people going to die? People are going to die from starvation if there isn’t enough food globally, and that’s one of the mechanisms. And when that happens, you can actually link with some degree of confidence to levels of climate warming. Arden Koehler: You said the way that you prefer to think about this is in terms of which temperature increases give us a good chance of these impacts. Do you have a guess at what degree of warming we would need to reach for the full-scale collapse of society, perhaps due to very, very widespread famine to have say a 10% chance of happening? Mark Lynas: You want to put me on the spot. I would say it has a 30% to 40% chance of happening at three degrees, and a 60% chance of happening at four degrees, and 90% at five degrees, and 97% at six degrees. Arden Koehler: I appreciate you being willing to put numbers on this, because I feel that’s always really hard, but it’s really helpful. Mark Lynas: Maybe 10% at two degrees. Arden Koehler: OK, so we’re getting close to 10% already. Mark Lynas: Yeah, but I think these early stages of warming, I mean, we’re at one degree now and obviously it hasn’t happened, so that’s a 0% chance of civilisational collapse unless it starts happening tomorrow. But the biggest impacts now have been on our natural ecosystems. Remember, we lose 99% of the world’s coral reefs before two degrees, so that’s an entire biome wiped out. And the same happens with rainforests and other important ecosystems, the whole Arctic tundra, and so on, fairly early on. And so humans are actually, and this is a bit of a truism, we’re a remarkably adaptable species. We’ve got all sorts of tools and technologies at our disposal. We can even deal with food production by using microorganisms and industrial baths. You have to have feedstocks like hydrogen and CO2 to put into those, but then you’ve completely delinked food production from climate. So this is why I wouldn’t put 100% risk on civilisational collapse, because there’s always the possibility that we can adapt our way throughout almost any imaginable scenario with the exception of turning the planet into Venus. But even then I suppose we can go off and terraform Mars, and maybe a few hundred people can do that for a couple of decades until it doesn’t work out. I don’t know. I’m indulging in spurious speculation now. But where does the synchronous breadbasket harvest failure come in? About four degrees. Where does this issue of biological and uninhabitability of large areas of very densely populated parts of the world? That’s about four degrees as well. So that’s why the numbers really go up at that level of warming. Arden Koehler: Long-time listeners will know we’re especially interested in extinction risks at 80,000 Hours. Can you answer for humanity going extinct? What temperature are we 10% likely to go extinct at? Mark Lynas: Well, to extinct humanity, you have to eliminate the last breeding pair… Arden Koehler: It’s a high bar. Mark Lynas: It’s a pretty high bar when you think people can live in artificial environments giving them sufficient food for quite a long time. Because of our technologies, we can live in minus 40, we can live in plus 40 Celsius. But I suppose our resilience comes, in the same way, as our Achilles heel — in that we are interdependent in incredibly complex societies which you can’t even understand. I mean, they happen dynamically, but no one really knows how the economy works. And so this fantasy that survivalists have of being able to look after themselves and maybe their family in a protected fortress-type environment, if you stockpile canned foods and the zombies don’t get you. You can do that for a decade or two, but — Arden Koehler: Then extinction would just happen later if that was all it depended on. Mark Lynas: Yeah. But I do think that humans are one of the last species to go extinct on the planet, not one of the first, because it’s just so obvious. A) There’s a lot of us, so to kill us all is going to take some doing. B) We’re incredibly adaptable and can live in all sorts of different environments. We’ve got all of the technologies and we’ve also got advanced knowledge of these impacts, i.e., because we’re discussing them now, so we can avoid most extinction scenarios: we don’t even have to be that smart to do so. So I’m not too worried about the near- to medium-term prospects of human extinction. I think there wouldn’t be much of any life left on the planet if humans were to fail to survive as well. Arden Koehler: What about the long-term risk of extinction? So let’s say if we get up to six degrees, do you have a guess at how probable it is that we go extinct? Mark Lynas: I think some humans would survive even at six degrees. You need to be in a situation where none of the planet’s surfaces are able to produce crops. Even at six degrees, you could produce crops up in the Arctic, or on the Antarctic Peninsula, or Alaska, or somewhere. Lewis Dartnell on how we could recover without much coal or oil Rob Wiblin: You think the first energy source that people should be reaching for, other than fossil fuels, is using hydropower and wind energy. Basically using these natural forces to turn a turbine, which you can then use either for mechanical work, like people have done throughout history for thousands of years, or to produce electricity. Because if you can spin a turbine, then you’ve got an electricity source with the right magnet setup. Do you think it would be possible to pre-prepare some kind of schematics for simpler wind generators or hydro generators that could be built and maintained with more medieval levels of technology, but which would be actually pretty good at generating energy? Lewis Dartnell: Hydropower and wind power are ancient technologies. We exploited them first for grinding flour and timber mills. And the idea I played with in terms of this sort of steampunk reboot process in The Knowledge was if you combine that mediaeval technology (i.e. easily re-achievable technology) with modern knowledge — which is actually quite simple, if only you know the secret to it, of the electromagnetism and magnets and copper wires — you could create a windmill that looks mediaeval, but is spinning a generator to create electricity for you. You’d have this sort of steampunk, electro-windmill type mashup. And so what you would want if you were to do this genuinely, is you could relatively easily prepare schematics, construction diagrams, blueprints of how to create a turbine — which is not quite as efficient as the 100-metre-tall ones you see dotted across the countryside, but is still a lot more superior to sort of Dutch 16th century design — and give all the wiring diagrams for the generator, and tell people how to make an electromagnet and make the thing work. Luisa Rodriguez: A minute ago you were saying that there’s a reason to think that some of the solar and other renewable energy sources might not be enough to do some of the critical energy-intensive things we’d want to be doing. And in a 2015 article you wrote — “Out of the Ashes,” which we’ll link to — you talk about some specific challenges humanity would face, having used up a lot of the readily available coal and oil. My colleague, Will MacAskill, has written a forthcoming book called What We Owe the Future, which you and I actually both contributed to a little bit. And it argues, among other things, that this is an important reason to get off fossil fuels ASAP, so that there’s some left over for our descendants in a post-disaster scenario. Does that sound right to you? Lewis Dartnell: Again, it comes down to which axioms you want to play around with: What was the exact scenario that necessitated everyone starting again from scratch? What state do you find the world in? And you can play around with some of those parameters, like: Is there still going to be crude oil underground — yes or no? Even if there’s not easily accessible crude oil, there’s still lakes of the stuff held in petrochemical refining stations around the world. So would you be able to get access to those? Or have they all leaked away, or have they burned, or how long has it been since the collapse before you are trying to recover? I think oil is a simpler answer, because I suspect we would struggle to get access to lots of oil starting again. Geologically, there’s not a great deal of oil left, because we’ve got very good at extracting it up until now. Whereas coal is a very different matter: there are megatons of coal. There is plenty of coal left underground, and you would only need to open-cast mine it. It’s relatively easy to get to. But for that Aeon article, I was playing with the idea of, let’s imagine that doesn’t exist. Let’s imagine that the collapse happens 100 years, 200 years in the future. Or for whatever reason, you are trying to recover society in a part of the world where you are not within walking distance of an open-cast coal mine. What alternatives might you go through and might you be able to use? And I talked about charcoal and how you could use that to smelt metals. And indeed, it’s not just a thought experiment, because a large fraction of the steel that’s being smelted in Brazil I think is done using charcoal. They have a lot of natural wood resources, and they use them sustainably, and they make metal from their forests. So you could go through that process if you needed to. And of course, charcoal is what we used before coal in the first place. You would just be stepping back to a slightly simpler technology and not having to really reinvent anything again. Rob Wiblin: I suppose I came away from the book and that article also having the same intuition that energy was going to be the big challenge for people after the apocalypse. But I felt cautiously optimistic that they’d be able to cobble together a bunch of different solutions that you were suggesting and kind of make things work. I mean, in order to get really high temperatures to do industrial processes, you’re saying we could use charcoal kind of like we used to. Currently that’s not practical to do on a large scale, because we need that land for food because there’s 8 billion people with mouths to feed. One benefit that we’d have after the apocalypse is there’s a lot less need for food, and so there’ll be more land available to grow trees, to make charcoal, to run this kind of stuff. And also, I suppose I just have this economics-y intuition, as someone who studied economics, that the prices will move, charcoal becomes more expensive. We’ll economise on our use of these industrial processes that require really high temperatures. And it’ll be horrible, but people will find a way to kind of make things livable. Lewis Dartnell: Yeah. I mean, that’s it. By its very nature, apocalypse kind of presupposes a mass depopulation event. So there’ll be a lot of free land available that is currently growing grain to feed mouths. And you won’t even need to plant forests to chop them down for charcoal — it will re-wild in a decade or two anyway, the forests will grow back over the farms. So you would have access to lots of wood, I think, at least in the early days as you recover. And then your own population of your rebooting society will start growing again, and you’ll start hitting exactly the same base of land use threshold that we started encountering in Britain as early as the Elizabethan Age. We had chopped down all the forests near the cities and towns, and so the price of fuel, the price of wood, was starting to go up — which was starting that process towards looking for alternatives, i.e. coal. Kevin Esvelt on people who want to bring down civilisation — and how AI could help them succeed Luisa Rodriguez: I imagine some of our listeners will be confused about the idea that people might be trying to kill everyone. When I first heard this argument, I found it very counterintuitive and just really hard to wrap my head around. Can you help make it a bit more intuitive? Why would any individual or group want to actually do this? Kill billions of people, maybe everyone, including themselves? Kevin Esvelt: Well, I think there’s a big difference between people who want to kill everyone and people who just want to bring down civilisation. And there’s one very famous individual who thought this way: Ted Kaczynski, the Unabomber. Luisa Rodriguez: Do you want to say more about that case? Kevin Esvelt: So I hate to recommend a manifesto written by a mass murderer, but it was pretty darn prescient considering that he wrote it in the early 1980s. He viewed the market system and technology as creating socioeconomic, sociotechnical incentives that would eventually cause us to use what he called the “immense power” of biotechnology to change who we fundamentally are, to make ourselves less than human in order to compete more effectively in the marketplace — and that we would thereby make ourselves increasingly miserable and an increasing travesty relative to what humanity should have been. This is what got him against technology. And this is a man who went to Harvard, who became a mathematics professor at Berkeley, and then threw it all away to live in a cabin in the woods and develop his philosophy and try to thwart progress by murdering people with incredibly sophisticated mail bombs that completely threw off the FBI for over a decade. Luisa Rodriguez: I do feel like a real concrete example of a person who thinks this way was necessary for me to actually get behind the idea that it’s one thing to have these kinds of beliefs as an ideology, and it’s another thing to want to act on them, and to actually act on them. I think it just felt like, sure, there are environmental activists who think humans are destroying the beauty of the Earth, and maybe they wish humans weren’t even here — but that’s very different from wanting to take the humans that are here, give them a horrible plague, and kill them all. That’s just such enormous suffering. I found it just really unbelievable. So yeah, I feel like the concrete cases made a huge difference. Another one I’ve heard of is the omnicidal cult Aum Shinrikyo, which I never pronounce correctly. Kevin Esvelt: I don’t either. Luisa Rodriguez: Do you mind giving the basics of that story as well? Kevin Esvelt: So Aum was a religious movement that arose in the 1980s in Japan — which is not the kind of society that we normally think of, at first glance, as spawning extremely radical religious movements that launch weapons of mass destruction programmes and then try to use them. I don’t know how typical religious movements are, but it clearly developed in directions moving towards a cult, but one that had a lot of members, a tremendously large income stream from thousands and thousands of members who tithed. But when it was developing, they eventually moved in a messianic direction — the founder, Asahara, did — and it moved in an apocalyptic direction. Not that everyone should die, including members of the cult, but very much that most people were going to die: the apocalypse was inevitable, and it would be in some ways more humane to bring about the end times of the current world so that the enlightened could build a better one. That was a major part of it. And they stooped to targeted murder of rivals and inconveniences relatively quickly, as they were explosively growing. Because first they had to hide an accidental death, which they did, and then it became, “Well, there’s this one person who is very much in the way.” And so they developed this sophisticated form of assassination, which then they botched, and had to kill more people and then cover that up. And eventually it became an adversarial thing where the apocalypse will happen, but it won’t happen soon enough, and the suffering will continue until we bring it about. So they launched weapons of mass destruction development programmes. They bought a uranium mine, they started developing chemical weapons. They started looking for biological weapons — and while there weren’t very many that they had access to at the time, they were able to produce botulinum toxin and they tried to make enough anthrax. And at least as a passing thought, the leader of their bioweapons programme, when they went to Africa, he was hoping that they would find someone who was infected with Ebola so that he could purify the virus and spread it around, so that it would hopefully transmit and kill as many people as possible. Luisa Rodriguez: That’s horrific. Kevin Esvelt: And is the most important part. So Ted Kaczynski, you can say, well, he was a mathematics professor. He clearly appreciated the immense power of biotechnology, even in the early 1980s. But he’s a mathematics professor: would he really have been able to do something about it? Well, again, this man is a genius and he was willing to throw away everything in his life and go and live in a cabin in the woods to pursue his philosophy. Would someone like that be willing to dedicate the time to picking up the skills which someone of that capability clearly could? I think so, if it was obvious that he could have access to something pandemic-like that could threaten the stability of civilisation. He didn’t have that, so he didn’t. Luisa Rodriguez: At the time. Right. Kevin Esvelt: We should return to the cult, though. Because their lead bioweaponeer was one of the original disciples, and he rose high in the ranks — in part because he was a graduate-trained virologist out of the University of Kyoto. Luisa Rodriguez: Yikes. Kevin Esvelt: Anyone with that level of technical training today has access to reverse genetics protocols that would let them make many of the smaller viruses. If you want access to 1918 influenza, you can just order it from a company that does not screen its orders, and you can follow the reverse genetics protocol that is freely available online. It’s open access. I would be incredibly disappointed in the University of Kyoto if a graduate-trained virologist from there, specialising in genetic engineering, could not obtain pretty much any influenza virus they want — and frankly, many of the others as well. There is just about no chance that they would be unable to successfully perform that protocol. And it does not require that much equipment either. And anecdotes are not data, but my second-year graduate student, who had never done virology before — had done mammalian tissue culture, but only for a couple of years — needed an influenza replicon for her research. And I said, “This is a good test case. How about I don’t help you, and you just try to figure it out from the protocols online? Do the design yourself. Go ahead and do it.” She did it. And I decided to check with some of my other students: “Do you think you can figure out how to design the reverse genetics plasmids for 1918?” They all could. Luisa Rodriguez: Can you say a bit more about what exactly that entails, just so I have a sense of actually how hard it is? Kevin Esvelt: Oh, you shouldn’t ask me. You should ask GPT-4. Luisa Rodriguez: Oh, god. OK, GPT-4 could probably tell me. That is terrifying. Could you expand on your concerns around AI models like GPT-4? Kevin Esvelt: I have two concerns with natural language processing models, large language models. Number one is they could expand access to existing nasty pandemic-class agents. Right now, you need some degree of lab skills in order to turn a publicly available genome into an infectious sample of virus. But we asked students in one of my classes — Safeguarding the Future, non-scientists — to leverage chatbots to figure out how to cause a pandemic. And in one hour, the three groups of students, plus the chatbots, came up with four of the nastiest viruses known — that would be not particularly likely, any of them, but among the most likely that we know of to cause pandemics. It told them that scientists can access these viruses by reverse genetics, producing infectious samples from synthetic DNA constructs, that not all companies do screen DNA to make sure that you’re not ordering something nasty, and that all of the companies that do all have their names conveniently listed on a website, so you can be sure that you’re ordering from one that does not. And then perhaps even more concerning, when the students asked, “What if I am a biochemist and I don’t know how to do reverse genetics? What do I do?,” it said that you can work with a core facility or a contract research organisation that will perform reverse genetics for you. You can send them your DNA constructs which you designed — and the LLM again will help you with the design — and they will send you back infectious samples. And it even will go into how to test whether the CRO is actually going to sequence your samples to make sure it is what you say it is. So the upshot is the LLM taught non-scientists, in an hour, which viruses are most dangerous, how to design DNA sufficient to produce them, who to order that DNA from and who to send it to, and how to do so in ways that could allow them to obtain infectious samples without being detected. That dramatically expands the number of folks who could plausibly gain access to potential pandemic agents. And it’s why we need to close some of those loopholes. We need universal DNA synthesis screening and we need to ensure that those contract research orgs really do sequence all of their customer samples — and not in a way such that someone who has penetrated their network can ensure that the sequencing file is replaced by a false one as soon as it appears, which is again something that the LLMs will talk to you about. So expanding access is one risk, but the other is just that we anticipate that scientists will learn to program biology in ways that, used maliciously, could create worse agents than natural ones. Eventually, I anticipate AI to get as good as human scientists at doing such things. If they are willing to tell the world how one might do that, then they will — and people will ask, and folks who are willing to misuse them will gain access. Toby Ord on whether rogue AI really could wipe us all out Rob Wiblin: One way that AI ends up really standing out in your analysis… You divide these three steps that you need in order for everyone to lose control of the future: you’ve got to have something that arises, and then spreads out over the whole world so it can affect everyone, then also that it finishes the job. So you have this problem with a disease: how does it reach the 1% of hardest to reach people in really remote places and nuclear submarines and all that? I guess with AI, even if it’s quite unlikely to arise, it’s a lot easier to see how these later steps happen, because it has this motivation to spread. It actually has intent in a way that a disease doesn’t. Toby Ord: Yeah. Or it’s at least easier to see at some very high level because of the intent. It’s actually trying to wrest control of the future. A couple of things that I just want to explain, though. A lot of people don’t quite see how the scaling up would work because they’re thinking about robots, though it’s probably less so among your audience. But ultimately, without any kind of robotic manipulators, AI could. If you think about Stalin or Hitler, we have people who scaled up from being one person, to being in control of a significant fraction of the world’s military power. Most of the orders of magnitude, up to the whole power, they scaled. They did so through manipulation of other humans. You can imagine AI systems on the internet spreading to millions of unsecured computers and trying to manipulate millions of people who are on the internet into doing their physical bidding by paying them or threatening them or enticing them; promising them things in the future. Rob Wiblin: It seems like even humans who aren’t super sophisticated sometimes seem to pull this off, influencing very large numbers of people to do strange things. Toby Ord: Yeah. I paint a picture of this in the book of perhaps how this could happen. It’s not meant to be the only picture, but it’s meant to show how, by building up a whole lot of abilities that humans have already achieved — things that are clearly within human level of intelligence; basically, building a large-scale criminal underworld — that AI going through that route could scale its way up to the power of a nation state or something. Then it’d be almost impossible to eradicate as well. That’s a bit of an explanation of the scaling step. As to the final step, I’m not claiming that AI is an extinction risk. I think that it’s not clear that even an AI that went badly wrong would want to kill everyone. I think that humans are the most interesting thing that it would have access to, possibly the most interesting thing in the affectable part of the universe. But that doesn’t make a substantial change. I don’t think they’ll be saying, “OK, humans, fill the universe with love and flourishing and all the things you want.” Our future would be radically curtailed if we were just there as something for them. Rob Wiblin: At the behest of them. Toby Ord: Yeah, at the behest of AI. And whatever goal function it was programmed with, it would be attempting to achieve that using us as an interesting piece of evidence to help it better achieve that goal, rather than listening to us and doing what we want. It would count as an existential catastrophe, though not necessarily claiming it would be an extinction catastrophe. [Update from 2025:] Rob Wiblin: The Precipice was this book that you wrote in 2018/2019, and came out in 2020. It explored the science behind all of the different major threats to humanity’s future: pandemics, asteroids, AI of course, nuclear war, that sort of stuff. Of course there’s been lots of developments since then. I think last year you wanted to look back and say, over the five years since you wrote it, what have been the major changes in the picture? Is humanity in a better situation? Is it in a worse situation? What have been the major changes? And in particular on AI, where so much has been going on? Toby Ord: Obviously lots of changes in pandemics. We had COVID hit us, mRNA vaccines, so on and so forth. And nuclear war: the prospects of that felt like a distant memory back in 2019, and now it’s become more of a realistic possibility. But AI is where the most changes have happened. Rob Wiblin: Back in 2019, I think in The Precipice, you estimated that the chance of humanity losing most of its potential future value due to AI in this century was around 1 in 10. Would you say that number has drifted up or drifted down? Toby Ord: I’m not sure. For a lot of the other risks, it was easier to see whether it’s gone up or down. For this one, I think that we’ve really been gifted a relatively good situation in terms of the way the technology has panned out, in that it’s this technology that imitated human values and human reasoning and so on by training on this huge corpus of human data. I think that has just been tremendously helpful. And the fact that it’s not an agent by default. Real gifts. It wasn’t that we steered towards that because we knew that would help with safety. It’s just that that turned out to be the easiest way. And I don’t think we quite recognise that enough: that the biggest effects on whether we’re safe or not have just come from somewhat random aspects about this technology landscape, rather than deliberate attempts to steer it. But I am very concerned about the racing, and I’m concerned that we’ve seen evidence that the players who are trying to make these systems are ultimately going to cut corners in order to win these races. Joan Rohlfing on why we need to worry about more than just nuclear winter Rob Wiblin: The mass use of nuclear weapons would be the greatest atrocity in human history and the greatest setback that humanity arguably has probably ever suffered — so by any normal definition, the scale is enormous. But people I know who are particularly concerned with extinction and existential risk have always been very interested to understand what would be the flow-through consequences from that, in order to see what would be humanity’s chances of bouncing back over a period of decades or centuries after the use of nuclear weapons. They’re very interested in questions like: How bad would the nuclear winter be? How long would it last? How many people would die? And things like that. You maybe think that people who aren’t familiar with the reading on this are slightly prioritising the wrong questions here, or perhaps misconstruing the issues, so I’m very interested to hear what you have to say about that. Joan Rohlfing: When I think about the long-term impacts of nuclear use, what I would really urge the audience to think about is not just an extinction event, but what happens when we have a major event to the resilience and ability of civilisation to flourish? So I would reframe it slightly to say that in addition to civilisation being extinguished, we have to worry about the disruption of the long-term potential of civilisation. I’m reminded of the excellent graph that Luke Muehlhauser put together that shows so vividly the massive post–Industrial Revolution gains across a variety of metrics — like life expectancy, GDP per capita, percentage of people not in poverty, energy capture, et cetera. It just takes off with the Industrial Revolution. I can imagine a major nuclear event pushing us back to a point where we’ve lost all those gains — and it’s not clear whether we recover them on the same trajectory. We just don’t know. We don’t know how we recover, how quickly we recover, whether we can recover. So it’s not just an extinction event that I would argue we need to worry about. I would also say one of the things I’ve noticed is that many in the EA community, I think, are overly focused on nuclear winter as the only trigger of a possible extinction event. What’s clear to me is that we need to worry about a lot more than just nuclear winter. Not to minimise it: nuclear winter would be a horrible thing if it happened. But NTI has been doing a landscape analysis to better understand the gaps in our knowledge of nuclear effects of a major nuclear exchange. What has become clear is that while we have a pretty good grip on the prompt effects of nuclear weapons, and some excellent research done on the likelihood of nuclear winter being triggered — I know there are uncertainties there, but some really excellent work informs our understanding of that — there is almost no work that has been done on the dynamic cascading effects of a series of impacts from the loss of critical civilisational systems, critical infrastructure. What happens when you’re losing power grids, banking systems, the ability to produce water? Rob Wiblin: Trade. Joan Rohlfing: Yeah. And so imagine these things happening all together, and people can’t get food. I mean, it’s inconceivable to me that our governments function the way we would hope and expect them to show up with services, when people are just fending for themselves, to provide for their basic daily needs. So I think there is a lot of interesting work that needs to happen to understand the knock-on effects: in a highly interdependent world, when you start losing key systems, how do things fall apart over time? Rob Wiblin: An audience member wrote in: “How much of our effort should be spent on preventing a war, and how much should we spend on building resilience to mitigate the harm and speed recovery?” They might be thinking of the kind of project like ALLFED — the Alliance to Feed the Earth in Disasters — which is trying to find ways of potentially producing enough food to sustain everyone through a nuclear winter. Joan Rohlfing: My own view is both are important, but I would put much more weight on prevention for a couple of reasons. One is I think that’s our highest leverage: if we can prevent the catastrophe from ever happening, then we don’t need to worry about the resilience. I also think, when I think about resilience on this scale — we’re talking about across the wide range of systems that would be required — it is a really, really hard problem. If we think reducing nuclear risks is hard, imagine fundamentally redesigning, decentralising all of our critical civilisational systems: this is what would be required to build sufficient resilience to not worry about critical systems failures. So I’d want to think about what that looks like for the power system, for the banking system, for international trade, for food production — there are just so many different layers of systems that would require massive rethinking of how those systems are structured, and massive reinvestment in rebuilding them, redesigning them, changing governance structures. I do think that’s important, and we should understand and give thought to those questions, but that feels lots more daunting and longer term and still with no guarantee of success. By the way, I think it’s also the reason to go back to an earlier observation about us not really fully understanding nuclear effects on all of these systems. It’s why investment and research and better understanding of nuclear effects is so important — because it would help point us to what we need to do to strengthen resilience. We wouldn’t even really have a clue how to go about resilience right now, because we don’t know where the biggest vulnerabilities are. Annie Jacobsen on the rings of annihilation and electromagnetic pulses from nuclear blasts Annie Jacobsen: Let’s talk about those nuclear effects, let’s talk about what nuclear weapons do to people and to things. These are effects that I source specifically from Defense Department documents. The Defense Department has been keeping track of all of this since the Hiroshima and Nagasaki bombs. And we exploded hundreds of atmospheric nuclear weapons during the Cold War, the effects of which were measured for these documents. I myself interviewed a number of these nuclear weapons engineers who armed, wired, and fired these weapons. And I thank them in my acknowledgments for teaching me so much about this horror. The idea of what a nuclear weapon does to people begins with this thermonuclear flash of light. And I describe the effects of the one-megaton attack on Washington, DC — which sets off this horrible scenario — as “19 football fields of fire.” That is the centre fireball. It will obliterate everything in a one-mile square radius. Everything. Nothing will remain: no cellular life, nothing. All humans will be turned into combusting carbon. Then you have to deal with the fact that there will be this dense wall of air pushing out from the centre point that is the blast, three miles out in every direction, knocking down everything. Engineered buildings change shapes; bridges, roads will melt into pools of essentially asphalt lava. That thermonuclear flash has set everything in the line of sight on fire. And for the book, I interviewed one of the world’s experts on nuclear firestorms. Her name is Lynn Eden. She’s a professor emeritus at Stanford University. And she talks about how the energy from the mass fire that will ensue is 15 to 50 times greater than the energy from the initial blast. Luisa Rodriguez: So the fireball that you’ve described is what happens kind of right at the centre of the diameter of the nuclear detonation. But there are other rings further out that also have really horrific consequences for all of the kind of life in those areas too. Can you talk about what happens in the second ring? Annie Jacobsen: The short version goes like this. Ring one: total annihilation; no cellular life remains. Ring two, another three-mile diameter out: everything is ablaze. Ring three, another three or five miles out on every side: third-degree burns among almost everyone. You are talking about people who may have gone down into the secret tunnels beneath Washington, DC, escaped from the Capitol and such: people are now broiling to death, people are dying from carbon monoxide poisoning, people who followed instructions and went into their basement are dying of suffocation. Everywhere there is death, everywhere there is fire. People are dying left and right of blood loss. Imagine the projectiles flying through the air after several-hundred-mile-an-hour winds overtake Washington, DC. Here’s an image which you will not be able to get out of your mind: the mushroom cloud — that iconic mushroom stem and cap that represents a nuclear blast — that mushroom cap and stem, when exploded, when a nuclear weapon has been exploded on a city, that stem and cap is made up of people. What is left over of people and of human civilisation. Luisa Rodriguez: Oh my god. And do you mean that in the sense that it’s basically the incinerated particles of people and buildings and everything and that’s being kind of lofted up as…? Annie Jacobsen: That is precisely what it is. I love your reaction, because it was exactly my reaction when I first learned that. You think to yourself, that is gruesome beyond gruesome. Then you ask yourself, is that really true? And then you think, well, of course it’s true. What else would it be? Luisa Rodriguez: Further out, people survive. But one might imagine — and I used to think — that if you’re not in the direct vicinity of a detonation, things might be kind of OK. But you paint a picture of what things are like for the survivors in other areas, and it’s actually really terrifying. It really changed my perception of what would happen in Wisconsin — or Texas, where I grew up. I don’t think that my particular neighbourhood in San Antonio, Texas is that likely to be the target, but everywhere in the US will be drastically impacted in ways that I really had not put together. So I found that extremely, extremely sobering. Annie Jacobsen: When you look at the target list for the United States — and again, these are sourced from very obfuscated maps that have been released over the decades from the various organisations; not just the Defense Department, but FEMA, the Department of Energy — you realise very quickly that everything is a target. If you’re talking about 1,000 nuclear warheads from Russia coming at the United States, in the words of Professor Brian Toon — who is the world’s expert on nuclear winter, an original author of that paper back in the 1980s, who I interviewed for the book — he explained to me that if you are in any kind of a city, not just a major city — I’m talking about small cities, any place with a military base, in any place with an industrial capacity — you are a target. There is a nuclear weapon pointed at you. And remember, this is happening over minutes, so it’s not a long amount of time that the survivors get to realise how horrific this all is. I did an interview with America’s first cyber chief, General Touhill. He was appointed by President Obama. He’s able to tell us a little bit about this idea of what happens to the United States of America — the chaos that ensues, the mayhem, the madness in the seconds and minutes after the initial two nuclear strikes from North Korea. Because given the age of social media, someone is videotaping this from even 25 miles out, or on an aeroplane far enough away that it wasn’t taken down by the EMP. And these images begin to circulate. This is not just plausible; this is probable. And then the world knows what has happened, and there begins a scramble for cover, a scramble for escape, a scramble for dot, dot, dot. And I describe that in the book. Luisa Rodriguez: Yeah. And yet, as you’ve already said, FEMA has no plan to help people, partly because it’s not like FEMA employees will be out there with the ability to help. An EMP plus the fallout, literally and figuratively, of nuclear detonations will mean that really no one has anything remotely like normal emergency infrastructure, normal anything. Actually, we haven’t touched on EMPs yet. Can you talk us through what would happen if a nuclear weapon were detonated in space in order to create a super EMP, or electromagnetic pulse? Annie Jacobsen: A super EMP is a three-phased electromagnetic shockwave so powerful that industrial-strength surge suppressors and lightning arrestors — all of which have been designed to block this kind of high-voltage spike — all become rendered useless at once. A super EMP happens by detonating a nuclear weapon — and it can be a small nuclear warhead — in space. Three hundred miles above the United States would do unbelievably catastrophic damage to the entire country. I went to the world experts to drill down on this: as I mentioned, General Touhill, Obama’s first cyber chief, and also Richard Garwin. Not only did he design the first thermonuclear weapon, Ivy Mike, but he wrote the first classified paper on EMP back in the 1950s. It’s still classified. They both confirmed with me that the collateral effects of this EMP strike are what would be so deadly. In simple terms, what essentially happens is the whole electrical grid would shut down. It would be a catastrophic failure: a failure of the systems of systems ultimately centred on this one concept I think that listeners might be able to wrap their head around, which is there’s a system in the United States that pretty much runs everything, and it’s called SCADA. That stands for “supervisory control and data acquisition.” It sounds like a mouthful, but what it really is is a system of systems of computers and people that control all the systems of the industrial underpinnings of the United States. So imagine losing control of railroad routers, of the liftgates on dams, of gas and oil refinery transmission centres. You’re going to lose assembly lines, you’re going to lose air traffic control, port facilities, fibre optics, GPS, hazardous materials. I go on and on in the book about what happens. It all happens in a second. Never mind that at least, at minimum, 10% of all cars in the United States will stop running simultaneously. Luisa Rodriguez: The passage that struck me most, I’ll just quote: There will be no more fresh water, no more toilets to flush, no sanitation, no streetlights, no tunnel lights, no lights at all — only candles until there are none left to burn. No gas pumps, no fuel, no ATMs, no cash withdrawals, no cell phones, no landlines, no emergency communication except for high frequency radios, no hospital equipment. It’s really the end of… everything. Annie Jacobsen: It really is. And you know, when I was reporting this book, really drilling down on the scientific facts about EMP, because the issue has tragically been politicised: some go so far as to make fun of people who refer to the EMP threat. This, despite the fact that the world experts pretty much unilaterally agree. And I’m talking about the Richard Garwins and the General Touhills of the world: when we’re talking about super EMP, they agree on the catastrophic possibilities here. Imagine me, writing this book, drilling down on that, wondering how I might be able to bridge the information gap between people’s preconception and what experts say is reality, coupled with the fact that almost no one had ever even heard of this threat. And then shortly before the book publishes in March, you learn from the US intelligence community of fears that Russia is considering putting a nuclear warhead in orbit. This is precisely what an EMP would do. David Denkenberger on disruptions to electricity and communications David Denkenberger: I think there’s another whole class of catastrophes that could disrupt electricity or infrastructure. One of them would be solar storms. Another is a detonation of a nuclear weapon at high altitude, causing an electromagnetic pulse, which could destroy electronics. Another one would be a coordinated cyber attack, perhaps enabled by narrow AI. A fourth one would actually be an extreme pandemic, where people are too scared to show up to work at critical industries like electric power plants. Rob Wiblin: Do you have any thoughts on whether we should be more or less worried about those things versus the risk of a nuclear war? David Denkenberger: I think as you’ve pointed out on other shows, the natural risks are probably lower probability. But the interesting thing about solar storms is that it’s a new threat in a way, because a solar storm 1,000 years ago was not going to hurt us — but now that we have this electric system that could be damaged, that’s a newer thing, so I think there’s some threat there. I think [a solar storm] is unlikely to be global, but I think even a regional disruption of industry or electricity is a significant shock — and would also likely be accompanied by a food production shock. Rob Wiblin: Is there much to say about how swiftly we might be able to respond to something like an electromagnetic pulse that damages the electricity grid? Do we have a good idea about how much damage that would do and how hard it would be to fix? Or are we left to some extent to speculate about that? David Denkenberger: There’s some debate, and former guest David Roodman has done some analysis on this. We’re actually working on a global analysis, looking at the impacts of a solar storm. But I think that in the case of the solar storm, there might be some things we could do ahead of time, if we detect it, because we might have a few days and be able to turn things off. The EMP is tough. Rob Wiblin: It’s by nature a weapon that would be designed to be a bolt from the blue. That would be the aim. David Denkenberger: Right. Rob Wiblin: Are we sure that if you detonate a nuclear weapon at high altitude, it has this effect? Has it ever actually been tested? David Denkenberger: It has. And in fact, we accidentally discovered how destructive it could be in the first test. Because it interacts with the atmosphere, it had a much greater radius of impact, and disrupted electricity I think on Hawaii, very far away from the original test. Rob Wiblin: Wow, interesting. So we discovered this effect basically through testing. Have we ever gotten to the point where we’re going to break electricity grids, or test it at that level? David Denkenberger: I think there was a test in the Soviet Union that was closer to people and there was disruption. And some people have said there may be these super EMP weapons developed at some point, and they could be even more destructive. But even just regular nuclear weapons are quite destructive. Modern society is extremely dependent on electricity, so there are these interdependent webs of causality here. If you lose electricity, you lose typically a lot of fossil fuel production, you lose communications. So basically, you’d have a collapse of industrial civilisation. Now, what we want to avoid is the full collapse of civilisation, which includes cooperation outside of small groups. But if we lose this industrial production, then there are some immediate needs. ALLFED has a catastrophe planning expert who used to work in the Royal Air Force. He has this rule of thumb that you’ll die in three minutes without air, three hours without shelter, three days without water, and three weeks without food. So fortunately, we’re still going to have buildings, but we’d need to heat those buildings. And then we also need to provide water very fast. So we are looking at how many people live close to water, such that they would not have to move locations and still gather water. But then eventually it’s going to impact food, because right now our system is very dependent on artificial fertilisers, pesticides, tractors, irrigation, et cetera. At least the sun would still be shining, but it would require dramatic scaling of hand or animal tools to farm, which we’re still working on. But we do have some estimate of the direct impact on agriculture of losing these industrial inputs: it’s something like cutting production in half. Now it could be, especially if there’s a trade disruption — and trade would be more difficult of course in these scenarios — that just hand farming is not going to produce enough food. Certainly, especially regionally, then it could be that you want to do some of these more resilient-type foods. Rob Wiblin: I guess an EMP, it’s at least somewhat local. Well, you could potentially have people setting them off all over the place, but potentially it could be a local thing. I guess solar storms as well, they tend to target particular parts of the Earth that happen to be hit by them. In that case, I wonder whether you just want to move all of the people to other places where the equipment is still running and the electricity grid is still up. Maybe that’s the best way of producing lots of food, is just to get people to places where it’s going to be easier to work. Does that sound plausible? David Denkenberger: Yeah, that’s one option, assuming that people are — Rob Wiblin: Transport’s open. David Denkenberger: Well, yeah. And that there’s enough cooperation that they could handle hundreds of millions of migrants. Rob Wiblin: Yeah, interesting. OK, so we’ve talked a bit about electricity. You also raised the issue of communication going down. How long do you think it would stay down, and how severely would it be down, and how big a problem might that be? David Denkenberger: Again, it depends on if it’s regional or global. And I think even though it’s unlikely, it’s possible that there could be multiple EMPs around the world. You look at a full-scale nuclear war between Russia and NATO, then you have most of Europe. It could even spread to other nuclear powers. I mean, that’s the majority of the world’s population and infrastructure. Rob Wiblin: Yeah, so it’s a lot of people covered potentially. Are there sorts of radios that work for this, if you really needed to? Or maybe the mail? I guess mail might be down. David Denkenberger: Yeah, or mail would be slow, on horses and such. So one thing we’ve looked at is shortwave radio, or sometimes known as ham radio, and they use a frequency that you don’t need to produce that much power. You can have a $20,000 system that can actually communicate across an ocean. So it could only be a few million dollars to have a backup communication system, and this would be extremely valuable — especially if the catastrophe were abrupt, that we didn’t have the internet to learn about what to do in this circumstance. So that we could get messages out of what to do, how to meet basic needs — at least in the first few weeks, and then buy us some more time. Rob Wiblin: Yeah. It seems like most countries do some stuff to prepare themselves against disasters. It seems like this would be well within the budget to have ham radios in major cities for this purpose. David Denkenberger: There are some, and I think the issue is that countries will have systems for themselves, and not every country has it. So what we’re interested in is doing a global system backup. We’ve also done a cost-effectiveness analysis for these interventions, and it’s actually only for the long-term future perspective, but I think there are, again, some mechanisms that this type of a loss of industrial civilisation could cascade downward to loss of conventional civilisation. I mean, the last time we were hunter-gatherers, there were only a few million people on Earth, so 99.9% mortality — and again, are we sure we’re going to recover from that? And in a way, I think it would be less expensive to get ready than for the resilient foods for nuclear winter, because we don’t have these expensive factories we need to pilot — we won’t be able to make factories. So the experiments will probably cost less, but we do need this radio backup system. But again, a few million dollars or so gets us pretty far. So we’re talking in the range of something like $100 million. Luisa Rodriguez on how we might lose critical knowledge Rob Wiblin: I suppose in the situations where there’s far fewer people — like 99% of people dead or something like that — how much will we struggle to access knowledge about, say, medicine? Or other topics where it’ll be really potentially useful to remember stuff from the pre-apocalypse? Luisa Rodriguez: Luckily, there is kind of an inverse correlation between how valuable knowledge is and how few people have it. This isn’t perfectly true, but mostly, even though there are many fewer brain surgeons than general practitioners or something, brain surgery is just less important than all the knowledge that general practitioners have. And that just does a lot of work. So we’ll lose lots of sophisticated knowledge of some types of medicine, but it’s hard to imagine why we’d lose germ theory, and even some basic things that will make maternal health better during childbirth, for example. And even that will be a huge improvement on where our ancestors were at similar population levels. Rob Wiblin: How many people know things like how to keep a car running, or how to run a power station, or how to run the electrical grid, or get that back up and running? Are there things that kind of stand out as maybe there’s not enough people in that group? Luisa Rodriguez: Yeah. Those were good candidates for what I called “critical skill sets,” and it’s just kind of hard to think about. One reason it’s hard is no single person knows how to run a power grid — it’s really distributed knowledge. And leadership at a plant might have more knowledge of the bigger picture, but it’s still distributed — not just in the many individuals who know the different steps, but also it’s going to be in manuals, and some of those will survive. So it’s hard to think of really critical knowledge that’s super concentrated in a couple of people. And I tried to do some back-of-the-envelope calculations to be like, what is concentrated and what could we lose? And I would love for someone else to try doing this research in a way that produces more interesting results than what I did. I ended up doing things like: we think advanced chemistry would be really nice to have, so I was like, “How many PhD chemists are there?” and try to think of it this way. And when you think of it that way, and you just make some naive assumptions about where they live, you still have lots of PhD chemists, even in a world where 99% of people have died. Rob Wiblin: I guess it might be getting down if you’re at the 99.99%? Luisa Rodriguez: That’s where you start to hypothesise. I just don’t know enough about how critical infrastructure works to be like, “What’s the job? What’s the one job that’s scarce, but super critical?” But you can think about it just theoretically, like, maybe there are some jobs like that, and you do get to the 99.99% population loss level before you start thinking that there are jobs like that that you’ll definitely lose. That opened this thread of, how does knowledge like that persist over time? Does it degrade? Does it get taught? When it gets taught, do we lose some important bits? Maybe there’s some evidence that when you get really small populations, when you try to pass down skills between generations, in a way you’re making copies of information — and when you try to teach it, if the number of people learning it is small enough, the copies will get lower and lower fidelity. Rob Wiblin: Because there’s not enough people to cross-reference and correct mistakes, maybe? Luisa Rodriguez: Yeah, I think that’s part of it. And another one is the more people you have, the more people innovate, and then you get improvements that make up for losses in the basic skill. But yeah, I think the way you can think of it is if someone’s teaching you to shoot an arrow, lots of people will be worse at it than the master, and you need enough people learning it to have some people exceed [the master]. Rob Wiblin: So that in the next generation, there’s someone as good as the best person in the previous time. Luisa Rodriguez: Yeah, exactly. And so, some examples of knowledge degrading over time that were interesting, I think one was Polynesian Islanders lost the ability to build boats and then got stuck. Rob Wiblin: That’s a big problem. Luisa Rodriguez: Yes. And apparently this is the way in which at least one group of Tasmanians lost the ability to build fire. Kevin Esvelt on the pandemic scenarios that could bring down civilisation Luisa Rodriguez: So you’ve written a paper describing two scenarios so catastrophic you think they could cause society to collapse. There’s what you call the “wildfire” pandemic and the “stealth” pandemic. And I want to talk about both of them, but I want to start with the stealth pandemic scenario. Can you say what happens in the stealth pandemic scenario? Kevin Esvelt: Imagine a fast-spreading respiratory HIV. It sweeps around the world. Almost nobody has symptoms. Nobody notices until years later, when the first people who are infected begin to succumb. They might die, something else debilitating might happen to them, but by that point, just about everyone on the planet would have been infected already. And then it would be a race. Can we come up with some way of defusing the thing? Can we come up with the equivalent of HIV antiretrovirals before it’s too late? Luisa Rodriguez: How hard is it then to convince whoever you need to convince — politicians, other academics, society — to actually take action? Given that no one’s going to be having any symptoms, everyone’s going to feel fine, and you’re going to be like, “But we think there’s a weird thing, and we think everyone should invest loads of money and effort to try to figure out what’s going on with this weird thing.” Is that going to happen? Kevin Esvelt: Well, I guess one question is: Who would need to support us for you to believe it? For the most sceptical member of your family, the most conspiracy-theory-minded person, to believe it? And those are two very different levels of burden of proof. Luisa Rodriguez: They are. Kevin Esvelt: But the horrible thing is we don’t need everyone to believe us; we need enough essential workers to believe us such that we can protect enough people to keep civilisation running, is the horrible answer. Obviously we want to save as many people as we can. We need to provide tools that will allow them to protect themselves if they believe — even if many other people living around them don’t believe, even in the same family. That’s hard, but it’s not impossible. And as to who do you need to believe? Well, you probably need at least a plurality of scientists, and ideally you would get near-unanimity in the scientific community. Now, scientists are like any other group of people: it’s very difficult to get 90% of scientists to agree on anything at all. We can always argue over something. But if you can get about 90% of life scientists to look at the genome and say, “Oh my god,” that’s probably good enough. And maybe you need to run experimental tests that it’s behaving the way you expect in the cell types that you expect. Maybe you would need to track down people who are infected and get their permission to run tests on them, verify that it’s doing the things that you would predict based on all of your analyses. You certainly, at a minimum, need to convince your defence establishment. Luisa Rodriguez: It’s interesting that so much of the problem is the sociological side. It sounds like we’re making good progress on the science side, but then how we get the people to do the things sounds extremely challenging. But it also sounds totally right that if you’ve been in conversation with the defence community for five years, you’ve shown them what your system is like and they’re bought in. And you’re like, “At some point we’re going to come to you and we’re going to say, ‘There’s this pathogen that we found and it looks really bad,’ we’re going to need you to have a plan already” then they might have a plan. And that plan should make things go much, much better than if they were just, one, in some disbelief, and two, just totally surprised. Kevin Esvelt: But that does bring us to the last point about the stealth scenario: the scientific community is not necessarily going to be willing to just go into their labs like normal, even given this motivation — because they believe that it’s out there, and they would be at risk and putting their families at risk. So this underscores the importance of ensuring that there is good enough protective equipment and healthy buildings initiatives to block transmission in buildings, perhaps starting in research labs. One of the things my lab did during COVID is, very early on, I figured this is obviously airborne. And therefore we knew pretty early on that there weren’t as many superspreading events in aeroplanes as you would expect; therefore, one complete air exchange every three minutes is your target. What would it take to get our laboratory’s rate up to that? And we just ordered 20 consumer-grade HEPA air purifiers and installed them in the lab and ran them full blast and had it as safe as planes, and then wore masks. And sure enough, we had zero infection events. Luisa Rodriguez: Wow. Kevin Esvelt: But you’re going to need to do something like that, ideally in advance. And again, that was with COVID which wasn’t all that serious to young and healthy people — although we were pretty scared at the beginning, because everyone was, and we didn’t know for sure. But if it’s something that is much worse than that, you’re going to need enough reliable protective equipment for people to go out there. And this is also true: the better you are at persuading people of the risk, if they’re essential workers, then you need protective equipment to persuade them that they can still go out there and keep everyone alive. Because I feel like that is one area where COVID taught us that some people are more essential than others. But we defined “essential” at a level that basically let society continue on more or less as it had — with some restrictions on the sides, a few inefficiencies, but otherwise basically that. We need to be a little bit more serious about it. We need enough reliable, no-fit-testing-required protective equipment like current PAPRs, but better and cheaper. Either we need enough for everyone at the outset — which would be the great way: that’s the way that any nation can just be like, “Yes, we are totally ready for whatever comes. If it’s stealth, we may not be able to persuade everyone to wear it, but it will be available for everyone” — but if you aren’t willing to put in that kind of investment, you really need to know who needs it. And this is more important for the other scenario. Luisa Rodriguez: OK, let’s move on to the wildfire pandemic scenario. Can you describe what happens there? Kevin Esvelt: Wildfire is fairly simple. There is a pandemic so contagious that we can’t stop it. And although COVID showed us that most of society can in fact stay home and avoid getting infected, in extremis, there’s quite a lot of people who can’t. The people who need to ensure the continued distribution of food, water, power, and law enforcement: those folks still need to be out there. Some of them need to interact with other people. Any pandemic agent that is contagious enough to spread through those people and take them out will disrupt essential services, and society will collapse. Many of these essential workers are some of the most vulnerable members of society, and yet they’re the ones who literally keep everything running. They are the ones who keep the lights on, the food on the table, the water in the taps, and order in the streets. And you can imagine that you don’t necessarily need the police per se (if you’re willing to call down martial law and have the military do the same thing), but you do need someone to be handling that. So the defence against wildfire is very straightforward: you need enough units of pandemic-proof personal protective equipment that don’t require fit testing to be sent to everyone who is going to need it — in the sense of the people who really do need to go out there and do their jobs or everything falls apart. All those people need protective equipment. If they have it, and all of the people who make the protective equipment have it, then we can weather the initial surge while everyone else locks down. Then we need to have enough protective equipment for the next group of essential workers. The primary are the ones who directly deliver those key services. Secondary essential workers are those that repair the equipment that the primary ones rely on, or produce the kinds of supplies that the primary workers need. That is, the secondaries aren’t needed immediately when the pandemic hits and everyone is terrified — because, again, this is very different from COVID: this is, if you get it, you are very likely to die. Just a wildly different setting; people would take it way more seriously. And the risk, in fact, is that too many people are no longer willing to go out at all, under any circumstances — because they quite reasonably believe if they get infected, they will likely bring it home to their families and then their families will die too. So we just need enough for all the primaries, and then we need to ensure that we produce it fast enough to deliver it to all the secondaries in time for when their services are required. And the really harsh truth is, if you’re trying to ensure that civilisation survives, you don’t necessarily need medical workers. You don’t need doctors and nurses and physicians’ assistants and all of the support that they require. Nor do you need elder care, nor do you need social workers, nor do you need any of those things. Because if those people aren’t there, lots of people are going to die, and that would be horrific and tragic. But from a very cold-eyed, cold-hearted perspective, that’s better than almost everyone dying — which is what happens if you lose the truly essential services. So a sane government will invest in enough units for primary and secondary essential workers, and all the lifesaving workers — but if they don’t, we at least need to know who those primaries and secondaries are, and at an absolute minimum, ensure that we can get enough units to them quickly. Now, the United States has this Strategic National Stockpile for essential medical goods and medical countermeasures and all kinds of disaster-preparedness type stuff. Bluntly, I don’t want the Strategic National Stockpile to have the pandemic-proof PPE. I mean, I would love for it to have it stockpiled, but they’re not very good at getting it out of the stockpile and into people’s hands — whereas we know for a fact that the private sector can do that reliably, probably on a next-day basis, but certainly within three days. Luisa Rodriguez: Amazon Prime. Kevin Esvelt: Amazon Prime. So we know we can deliver whatever to whoever very, very quickly using some services in society, and it is not obvious to me that the Strategic National Stockpile can do that. Now, perhaps they should just talk to Amazon and say, “Here is where it’s going to be stored. You need to get it into your distribution network for delivery within five days, absolute maximum.” Fine. But the point is, we need to know who are the essential workers. We need enough “P4E” units — pandemic-proof personal protective equipment, P4E for short — for all of those at a minimum, and preferably for all lifesaving workers, and have lists of their addresses. And of course, “essential workers” includes all the folks who are going to be doing the deliveries, because everyone is going to need food. Are people going to need to go and pick up their food, or can we do online delivery? But if we have that — if we have the lists of who needs it, and we have the equipment, and we have stockpiled enough materials so that we can last until we have equipment for the additional groups of workers who are needed to repair the essential equipment and provide the supplies and stockpiles of the new stuff where we don’t have enough in reserve — then we’ll be fine. Wildfire is an obviously solvable problem using our current capabilities and current technologies. How much does it cost? Well, it depends on how low you can get the price of P4E. I’m pretty confident we can get it down below $250. But the trick is we also would ideally want it to be comfortable. And of course, it needs to be reliable and it needs to be convincing. People need to believe that it will work. Because if people don’t believe it works, it doesn’t matter whether or not it actually works: they’re not going to go out there and keep everybody alive. Or probably not. We can’t assume that they would. We do have some historical data to rely on here: we can look at the SARS-1 outbreak, about 10% lethality. And there was a lot of pressure on nurses and doctors, especially the ones who had families and young children, to not go to work, and you definitely saw a bias towards the young and childless as the ones in the wards. So at 10%, that’s what you see. How high does it need to be to be a wildfire? Well, I don’t know, but it would need to debilitate enough of the essential workers that services would collapse. That’s the only way that you actually lose civilisation. Luisa Rodriguez: How hard is it going to be to convince people not to give lifesaving healthcare workers P4E so that they can go on doing lifesaving work? That sounds like the kind of thing people are going to find really, really objectionable. Kevin Esvelt: I totally agree. It’s just that people are bad at making tradeoffs. And in any scenario where you can afford to, you absolutely should give them P4E. There’s just no question, so I am not for a moment going to argue it. I’m just going to point out that suppose that we fail to bring down the cost of P4E, and you just need to buy it now. Call it $1,000 a unit: how many units are we going to buy? The United States, in terms of primary essential workers and then the very near-term secondaries, probably can get away with 20 million or so. So there’s $20 billion. Last year, for context, Congress gave the Department of Defense $30 billion more than they asked for — so we could have just used that bonus that Congress handed the DoD to completely immunise the United States against wildfire pandemics. That’s why I mean this is a totally solvable problem. And then figuring out who needs it is not super challenging. We did it on the fly for COVID, just under much reduced stringency in terms of who was needed to be essential. We can figure all this out. It’s not that much effort. It’s not that expensive. But if you wanted to get all the healthcare workers now, that’s like 16% of the workforce on top. So there you’re looking at probably more like 50 million [workers] instead of 20 million, if you want to get all of the medical workforce and the folks who support them. That’s just a lot more people. And that’s if you include social workers, elder care, all of that jazz. So should we do that? Yes, absolutely. It’s just that’s an extra $30 billion. Should we spend that? Yes, yes, we should. But if we’re not, then it’s really important that when the time comes, we recognise the fact that you ship the units to the most essential people — and those may not be medical workers. Luisa Rodriguez: It sounds like you don’t have some exact percent lethality that you need to have to definitely be in a wildfire scenario — where people are unwilling to go to work, causing something like civilisational collapse. I’ve heard something like the reason we haven’t seen pandemics that have both high transmissibility and high lethality before in a way that causes this kind of particularly horrible situation is because those things come with evolutionary tradeoffs. Is that right? Kevin Esvelt: That’s probably right for some pathogens and not for others. Certainly the Black Death had both. Certainly smallpox has both — or at least the variola major strain is 30% lethal and R0 between 3.5 and 6. So is that in wildfire territory? That’s the only one, though, that we label as being probably transmissible enough — because the Black Death, even if it weren’t susceptible to antibiotics, was just not transmissible enough in the modern world. So the only one we know about is smallpox that we think would possibly be wildfire level today, at 30% lethality. And it’s worth noting that the Soviets almost certainly enhanced it, to the point where when there was an accidental outbreak in the Aralsk region, out of 10 known victims, the three who were unvaccinated all died and it was transmitted efficiently by vaccinated people, which wild-type smallpox does not do. So clearly it is possible. And again, this caused an outbreak. They managed to contain it: they shut down all the trains, they got it under control. Once we get good enough at programming biology, such that these other capabilities can apply, we just don’t know what is going to become possible. I would not assume that whatever natural tradeoff exists between contagiousness and virulence is necessarily going to always apply. And what’s more, even if there is a pathogen that is not evolutionarily stable — that is, mutants will accumulate that will, say, reduce the lethality over time — that doesn’t mean it can’t crash civilisation first. Luisa Rodriguez: Just to make sure I understand the evolutionary tradeoff, is it basically at some high enough level of lethality, it can’t actually spread very far, because it’s killing people before it spreads? Kevin Esvelt: Yes, it seems to be linked to whether or not it kills you before you have a chance to transmit it. If the transmission window ends and then it kills you, then there’s no potential limit. Some known pathogens often kill you after the transmission window is mostly closed, so those ones don’t seem to be particularly subject to the tradeoff — and you can certainly imagine selecting viruses for that particular trait. Andy Weber on the tech that could prevent catastrophic pandemics Rob Wiblin: Is there any case for not worrying that much about bioweapons? They sound pretty bad, but is there anything you can say that maybe makes the picture seem a bit less scary? Andy Weber: Maybe it doesn’t come through, but I’m an optimist by nature, sort of a pragmatic optimist. And I think we can take the whole class of biological weapons off the table. We’re doing work on this at the Council on Strategic Risks. We have a programme called Making Bioweapons Obsolete. At the end of the day, biological weapons are just infectious disease, right? So you can have a system of early warning detection and then rapid medical countermeasures that would give you such good defences against biological weapons that you could deter your adversaries from pursuing them, because they would realise they wouldn’t be effective. And we’re not there yet, but the science and the tools that are now available can enable this possibility of making bioweapons obsolete. And so we think that’s the right vision. Rob Wiblin: OK. Yeah, fantastic. I’m super curious, can you maybe just go through some of the technologies that you’re most excited about, that could potentially make bioweapons obsolete? Andy Weber: Absolutely. So it starts with early warning. So think about some of the revolution in diagnostics and testing that’s been happening because of COVID. So imagine in-home testing: every morning when you’re brushing your teeth, you breathe into a tube and you know if you’re infected with a virus and it tells you which one. That’s sort of the early detection that’s now possible. You might have it hooked up to your smartphone and it would report into a public health surveillance system. And so you would know if it’s just an isolated individual or if it’s spreading into the community. And then sequencing is becoming so cheap and fast that within 24 hours, you can determine if a person has a respiratory illness, for example. You can test through sequencing for 300 different pathogens, as well as virus X, the unknown, the novel. And those are the sorts of capabilities that we need in order to have that weather map, that prediction of infectious disease outbreaks that allows us to nip them in the bud. Rob Wiblin: I know that sequencing technology has gotten a lot cheaper, and I’ve heard people talk about at every hospital or medical facility, you could have one of these little nanopore sequencers, and then you can take lots of samples and see if you’re getting DNA results or sequence results that you don’t recognise. And that would set off alarm bells because it’s like, “Oh, this is something new. We need to look into it.” How would you do this at home? You’re saying you would blow into something at home and this will be cheap enough and it would be able to tell all the viruses that you had going around in your respiratory tract? That’s amazing. Andy Weber: Right. So those capabilities that you’re talking about, that a hospital laboratory might have today, they’re getting smaller, cheaper, and easier to use. So it’s just a matter of time before they’re available for use at home. So using sequences, we can have what I call rapid medical countermeasures — therapeutics and vaccines that are based on the sequence of the pathogen and can be developed and then manufactured quickly. Think sort of nucleic acid 3D printers that could use this digital information — because biology is now digital — and develop and produce these vaccines in a distributed way. So as soon as even a novel or previously unknown pathogen were released, we would in days — or a month at the most — have a countermeasure we can apply to preventing it. So these are the capabilities that while even 10 or 15 years ago may have seemed like science fiction, today they’re upon us. But we need to invest in this overall system of defences. Rob Wiblin: And I guess a benefit of mRNA vaccines, as I understand it, is that you can very quickly go from the RNA or DNA sequence of the pathogen to, I guess you take that and you do some work on the computer quite quickly, and then you’re like, “All right, we already have mRNA vaccines, here’s the code that we’re going to put into the mRNA vaccine. Here’s the code that we’re going to put inside these little fat droplets that then is going to cause your own body to produce these proteins that will then inspire the appropriate immune response.” And so I guess you can go much faster, potentially, from discovering a new thing to having a vaccine against it. Andy Weber: Yeah. And linking back to the Soviet biological weapons programme, there’s a reason the Department of Defense invested in a capability like the mRNA vaccines. It’s simple: we knew the Soviet Union had been applying bioengineering to its weapons programme. So that meant we couldn’t just worry about a list of known pathogens. It meant we had to worry about a created pathogen, perhaps a chimera — a hybrid pathogen that had the worst properties of several viruses combined. So that meant that we might face a total unknown. And so we needed a capability to sequence it, to characterise it, and then develop the countermeasure. And it’s exactly that reason that the Department of Defense invested in what we call these ‘rapid response’ or ‘platform’ technologies. Rob Wiblin: Yeah. That’s fantastic. And if we want to have this ability to quickly disable any new bioweapon, we need to not only be able to invent the vaccine — it sounds like possibly within days — but then also very quickly manufacture a lot of it, and then deliver it into arms on a very accelerated timeline. Andy Weber: That’s absolutely right. And technology is helping us with that too. You no longer have to build a massive vaccine plant to produce vaccines. Again, think about 3D printers. It’s called cell-free manufacturing. There’s a company I’m aware of that is building something about the size of a glove box that can produce, for example, millions of doses of mRNA vaccine in days. Because as you said, it’s actually your body that does the production of the mRNA vaccine; it’s just the message to your cells to produce it that you’re injecting. And then for example, there’s microneedle patches, vaccine patches that could be mailed to people, right? So that would make the logistics of vaccinating people much, much easier if you can just apply a patch to your arm: new methods to eliminate the need for a cold chain, for low temperature freezers. All of these are technical challenges that we need to work on, but none of them are insurmountable. Compare the investment we’re talking about with the costs of this pandemic and it’s a no-brainer. It really is an ounce of prevention. So we’re now using the term ‘the biodefence industrial base’ in the United States. We need that warm base, but it’s multi-use right? During a non-pandemic time, it can be used to produce all sorts of things. But then it needs to have the agility to shift quickly to snuff out a pandemic before it becomes one. Rob Wiblin: The political challenge here is that people have been saying for decades that we need to stockpile more masks, we need to have more vaccine manufacturing capacity on standby. We need to be doing all of these things to prepare for a pandemic when it arrives. But the challenge then is how do you get politicians and the public to vote to spend this money, given that if you haven’t had a pandemic lately, it’s going to seem like it’s being wasted? The challenge here kind of seems like more a political economy question or a public choice question than one of figuring out whether it’s sensible to do, because it just so clearly is. Andy Weber: Yeah. The simple fact is that it needs to be a priority. And we’re not talking about huge numbers. I think the United States, if we invested through the Department of Defense and Department of Health and Human Services about $20 billion a year in this — which when you think about it, the defence budget is currently $750 billion a year, it’s actually a pittance in DoD terms — I think we could solve this problem over about a decade. So it’s an investment that is worth making for our defence, but also for our health. Imagine a world where you don’t get colds. That’s what we’re talking about too. So it would improve the day-to-day lives of people. And that’s the beauty of this: it’s a win-win-win investment. It’s an investment we can’t afford not to make. It’s cheap, but we haven’t done it. What it comes down to is we need to broaden the definition of national security and defence. It’s not just about kinetic warfare. It’s much broader than that. Keeping the American people and our allies and partners around the world safe from infectious disease is part of that. Christian Ruhl on why we need the equivalents of seatbelts and airbags to prevent nuclear war from threatening civilisation Luisa Rodriguez: My impression is that most of the work in the nuclear space is on nonproliferation and nuclear disarmament and deterrence — so mostly things that aim to reduce the odds that a nuclear war ever starts. But you’re particularly interested in nuclear-related interventions that are helpful in scenarios where nuclear bombs have already been detonated. You call these “right-of-boom” interventions. Can you explain the distinction between right-of-boom interventions and left-of-boom ones? Christian Ruhl: Yeah. So the big motivation here is: what is the high-impact thing to do here? And I understand that “high-impact” is a very unfortunate phrase to use when we’re talking about bombs going off. I’m going to make an analogy to car crashes. So for car crashes, we can maybe think about dividing interventions into left-of-crash and right-of-crash interventions. Left-of-crash includes things like rules of the road, stop signs, traffic lights, driver’s licences, and so on. And we also have right-of-crash interventions in case the first layer of defence fails — so we’re talking seatbelts, airbags, features of a car to make it safer, ambulances, hospitals. Why do we have this? Because we know that there are many reasons that cars crash and accidents happen. So we can take that back to nuclear war. We really, really want to make sure that nuclear war never breaks out. But we also know — from all of the examples of the Cold War, all these close calls — that it very well could, as long as there are nuclear weapons in the world. So if it does, we want to have some ways of preventing that from turning into a civilisation-threatening, cataclysmic kind of war that you’ve thought about in your own work. And those kinds of interventions — war limitation, intrawar escalation management, civil defence — those are kind of the seatbelts and airbags of the nuclear world. So to borrow a phrase from one of my colleagues, right-of-boom is a class of interventions for when “shit hits the fan.” Basically, the logic here is we should have a layered defence against catastrophic risks. So imagine you live in a world, again, with cars — but no seatbelts, no airbags, or any other safety features. That’s the world we live in right now when it comes to nuclear war. And fundamentally, that’s why I think we should be dedicating more resources to right-of-boom interventions. So we know a few things about nuclear war. First of all, not all nuclear wars are created equal. There’s a qualitative difference between a single weapon going off, and the superpowers unleashing their full arsenals. One of those is, as you said, a truly horrific humanitarian disaster, but it’s mostly local. And the other one is this unprecedented global cataclysm that might well threaten modern civilisation itself. The largest nuclear wars are disproportionately worse than smaller nuclear wars, which means that much of the total expected cost there lies with those largest wars. It’s a familiar feature in catastrophic risk; I think we see something very similar when looking at pandemics and biosecurity. So that’s the general idea, but it’s also a subtler argument about philanthropic strategy here, and about making allocations in philanthropy under high uncertainty. So fundamentally, this is about taking not just one step back but like 10 steps back, and thinking about the structure of the problem at a really high level, to kind of figure out the most effective ways to do good at the margins. Luisa Rodriguez: OK, so another one of these right-of-boom interventions you think is promising is civil defence. What does that mean? What does it look like? Christian Ruhl: So think back to March 2020, and the confusion and fear that many of us felt about COVID, not knowing how do I keep myself and my family safe? Imagine instead of that, we had learned in school, “We know these outbreaks happen, so if something like this happens, here are some measures that can help.” And imagine the government had taken steps to research pandemic-proof PPE and sent high-quality respirators to every household just in case — not that much money, and protecting the civilian population in case of a war or in case of a different catastrophe. That’s what we’re talking about when we talk about civil defence. Today it’s often called “emergency management.” So with nuclear war, we’re talking about measures like shelters, evacuation, public education on what to do, PPE again, stockpiles of critical supplies, food. It was always a topic that came up in the Cold War, but it never really got off — but mostly for political reasons, not because it wouldn’t have worked. Luisa Rodriguez: It does seem like we did more of that in the Cold War. I feel like there’s this cultural meme of “duck and cover.” And it’s kind of funny that we don’t do that, even though it sounds like at least some of it would be reasonable to do. According to some people, we face similarly scary nuclear threats today relative to the ones we were facing then. Arguably that’s not true, but it still seems like probably the risks are high enough that we should be doing something there, if something there would work. Christian Ruhl: And we have some evidence that some of these measures probably would work. In Nagasaki, there were about 400 people who were very close to ground zero, and took shelter inside of these hillside caves. And these 400 people survived. And not only did they survive, except for the people who were immediately by the entrances to the caves, they survived uninjured. We know there are ways to protect people. It is really scary to think about, but it’s potentially worth thinking about. Luisa Rodriguez: Moving to another one, we’ve already touched on this a little bit, but we have hotlines and also just specifically war termination. What do those look like? Christian Ruhl: So one of the things you want to be able to do, again, if a nuclear war breaks out, is make sure you have a way to communicate with the adversary to say, “Let’s stop this. Let’s find a way to have peace.” Unfortunately, as far as I can tell from public sources and from talking to experts about this, the nuclear hotlines we have seem likely to fail in the event of war. And obviously, you can’t end the war if you can’t communicate with your adversary. This hotline was implemented in 1963. Really early versions were pretty fragile and insecure. There was one kind of funny case where a farmer in Finland accidentally ploughed through a cable in 1965. So they’ve made updates every once in a while. They switched to satellite in 1971, added fax 1984, and switched to email in 2008. But one of the last public examinations we have about hotline resilience is from the ’90s, and they write: “The DCL is not designed to survive or function in a war environment. Its principal component subsystems are essentially unprotected against blasts or other nuclear effects or electronic countermeasures such as jamming.” So again: a huge problem. Seems obvious that we should try to fix it. I think like a study group in the public to say, you know, we don’t have access to the classified information about what these systems actually look like, but we want to make sure this will actually work. Luisa Rodriguez: Yeah. Again, I feel pretty shocked. I did not know that the hotlines would fail after a nuclear blast. That’s insane. It feels hard enough for leaders trying to deescalate a nuclear war to do that with verbal communication — and without, when you’re just guessing about what another country is thinking, that just feels impossible. So that’s absolutely horrifying. Christian Ruhl: Yeah. What are you going to do, explode bombs in Morse code or something? I don’t know. Luisa Rodriguez: It’s bizarre. Truly bizarre. Do we have technology that would survive nuclear detonations and we could just, in theory, implement that, but haven’t? Or do we have to develop something somehow that’s more robust? Christian Ruhl: I think it gets really complicated. And a lot of the stuff about this I do think is probably classified. So it’s totally possible that I’m wrong about a lot of this. But we know that nuclear weapons states care a lot about the communications links of their own systems, so the NC2 are making sure that that keeps working if a war breaks out. It’s kind of this perennial focus of war planning is ensuring the resiliency of NC2 systems. And it would seem like the very same care and whatever measures are taken to, say, protect electronics from EMP and stuff like that might very well be used for state-to-state communications. Or maybe you make it redundant with multiple satellites, in case it’s like war in space. There are many things we can think of. Again, totally possible that I’m just wrong about this, but from what we can tell publicly, it would fail. I think one of the things that people don’t appreciate is that, even if nuclear war breaks out, and even if humanity survives, and even if civilisation recovers, that recovery might not be with good values. So we can imagine things like unconstrained competition around AI between multiple authoritarian states. After a nuclear war over Taiwan, we might worry about the risk of locking in certain kinds of political systems, and about totalitarian political systems perhaps being more likely to survive. Luisa Rodriguez: Sure, and just for context: the reason we might worry about this is because for various reasons, AI might make it possible to lock in political systems or value systems by stopping different modes of cultural evolution — for example, by enabling identical digital copies of totalitarian leaders so that there never has to be a regime change, just for one example. Christian Ruhl: So let’s think about [with which] concrete levers we can actually maybe affect what kinds of political systems might survive after nuclear war. One of these in the US is the continuity of government plans, or COG plans. So there’s like Raven Rock “designated survivor” kind of stuff, making sure that the US government continues in some form, even if there’s a nuclear war. But if you look back, and look at what some of the Cold War plans were for what to do… There’s a great book called Raven Rock, actually, that talks about this. But let’s take the Eisenhower administration as an example. As I understand it, the plans were completely insane and authoritarian. He had set up this system to, in a time of crisis, a small handful of his best friends would take power, nationalise industries, restructure the government, and basically start this crazy oligarchy. Now, COG plans obviously are very classified. That doesn’t mean that philanthropists can’t have an impact here. One thing you could do is fund a high-level study group of continuity of government against emerging threats — include nuclear war, include biosecurity, include AI — and do policy advocacy of key decision makers that ultimately make these plans, and emphasise the importance of having mechanisms to return to constitutional democratic government to keep the values that we care about. There’s actually an example of this in 2002: a right-leaning think tank AEI and left-leaning think tank Brookings worked together to do this continuity of government commission — so a precedent for doing something like this. Again, I think there are really concrete things we can do to make sure that if, god forbid, something like this happens, things go a little bit better than they otherwise would. Mark Lynas on whether wide-scale famine would lead to civilisational collapse Rob Wiblin: Let’s talk about the famine channel first. I’m worried about that, but I think a bit less worried than you. I guess there’s two things that we could maybe look at. One is: what is the probability that we have a really big food shortfall? And then, what’s the probability that that leads to some kind of cascading civilisational collapse? Let’s maybe do the second one first. So imagine that we do have a whole bunch of breadbasket areas that all have a drought simultaneously. And so food output globally goes down a bunch. Let’s say we only have 70% or 80% as much food as we have in a normal year. What’s the path by which that leads not only to a lot of people dying of hunger, but it also leads to civilisation as a whole falling apart? Mark Lynas: It’s reflected in prices. Obviously, the price mechanism is the rationing mechanism that we have to decide who gets to eat and who doesn’t in pretty much every society. And there was some forewarning of this in the 2008 food price spikes — which were more related to the oil price at that time actually than the scarcity of food, but I think the same thing would happen — and even then, the food-producing countries ceased to export. They put on export bans, which then affects the commodities trade. That means that importing countries experience both rapidly increasing prices and their result is food insecurity. So if the price is double, triple, you’d get more than a 30% reduction. By the way, if you lost that much harvest, you’d lose all of your world’s fair trade. So food-importing countries could potentially starve very quickly. And people aren’t just going to sit there and gradually get hungrier and hungrier. They’re going to move. They’re going to move in the millions to try and find whatever foods are available anywhere. And those kinds of dynamic effects are both very difficult to predict, but — Rob Wiblin: Don’t look good. Mark Lynas: No, they don’t. Exactly. No, they don’t look good. They’re not going to have an outcome which means that everything’s hunky dory. Arden Koehler: So maybe people moving around in order to find places where they can eat might lead to refugees and that would possibly lead to conflict. Is that the idea? And then that would lead to collapse? Mark Lynas: I guess. I mean, it’s again very difficult to predict the response of countries which do have food sufficiency. Are they going to be prepared to share, or are they going to put up borders and build walls? The evidence from the Syrian refugee crisis suggests that the response would be to close borders and build walls and allow people to drown in the Mediterranean in the thousands, as Europe has done. And look at the political ramifications of that. You then saw a rise in populist movements and almost fascist type political parties, which are now in power in European countries, and take us back to a time which is scarier than any since the 1930s, I would say. And this is Syria. This is one country. Imagine this happening across the majority of the world’s countries. Then those countries which do produce enough for themselves, they’re really going to have a tough time. So yeah, we can take glimpses of the future perhaps, from some of the really awful things that have happened in previous years. But there’s a multisystemic nature. It doesn’t really give you a sense of the magnitude of the result, I don’t think. Rob Wiblin: So when I envisage a situation where there’s a huge food shortfall like that, firstly, I think we’ll probably have some heads up that this is coming ahead of time. You start to notice the warning signs earlier, like food prices going up, and food futures going up. Because it’ll be a global emergency much worse than the coronavirus, say. You just start seeing everyone starts paying attention to how the hell can we get more calories produced? And fortunately, unlike 500 years ago, we are in the fortunate situation where most people today aren’t already producing food, and most capital today isn’t already allocated towards producing more food. So there’s potentially a bunch of elasticity there where, if food prices go up tenfold, that a lot more people can go out and try to grow food one way or another. And a lot more capital can be reallocated towards agriculture in order to try to ameliorate the effects. And you can also imagine, just as everyone in March was trying to figure out how the hell do we solve this COVID problem, everyone’s going to be thinking “How can I store food? How can I avoid consuming food? How can we avoid wasting food?” Because every calorie looks precious. And maybe that sense of our adaptability, or our ability to set our mind to something when there’s a huge disaster and just throw everything at it, perhaps makes me more optimistic that we’ll be able to muddle through, perhaps more than you’re envisaging. Do you have a reaction to that? Mark Lynas: My reaction is: imagine if Donald Trump is in charge of the response. It’s all very well to have optimistic notions of technological progress and adaptive capacity and things. And yeah, if smart people were running the show, that would no doubt be the most likely outcome. But smart people don’t run the show most of the time, in most places, and people are amenable to hate and fear, and denial and conspiracies, and all of those kinds of things as you’ve seen, even in the very short-term challenges of COVID. Rob Wiblin: Yeah, it is interesting. I’m not sure whether to update positively or negatively on society’s resilience watching COVID. There’s some places that have handled it really admirably well, and I’ve been impressed by how good their governments are. And then other places where it’s just been remarkably incompetent. Something that’s really struck me is how disruptive it’s been to have a virus that has an under 1% fatality rate, which in the scheme of pandemics really isn’t that serious, and yet it threw everything into disarray pretty quickly — which has maybe made me more worried about these flow-through effects. If 10% of people are starving, how much disruption does that create to everything? And I guess the fear of conflict: how much does that make it hard to get anything done? Mark Lynas: And it wouldn’t just be a global average of 10%. You have a global average of 1% for COVID. It would be, I don’t know, 50%, 90% in certain places. And I think that would be a psychological shock to us as much as anything. Even if you’re in a country which is likely to survive, just seeing that extent of famine would be… We haven’t seen anything like that since the mediaeval times. But yes, societies tolerated, with the Black Death, the loss of half of the population across much of Western Europe at certain times. And there were famines, which probably had death tolls of similar magnitude in those times as well. And what do people do? They worshiped, they castigated themselves, they fought wars. But they survived. So that isn’t necessarily the herald of extinction, but remember, these processes are getting worse, unless we’ve stopped emissions by that time, because the economy just can’t… How are you going to carry on burning huge amounts of coal, or any kind of fossil fuel, if society is so challenged by a mass famine, for example? Arden Koehler: I was reading your book, and you mentioned that one issue with talking about adaptability is that it’s going to be much harder to adapt without burning fossil fuels if we haven’t already transitioned into a carbon-free economy. Fossil fuels are the source of our ability to do so much. So maybe people will feel like, for the next five years, we really need to do this huge project in order to adapt — which is going to require us to use a lot of energy, which is going to require us to burn a lot of carbon — which might make it so that people are more likely than I had initially intuitively thought to continue releasing carbon into the atmosphere, even as things get really, really bad. Mark Lynas: Well, that’s the Dubai scenario, where you have a petrostate under a huge air conditioned dome in an intolerably hot environment which is entirely dependent on artificial water, artificial energy generally, and food from elsewhere. So yes, it’s certainly possible that whole cities can exist in an ambient environment which is intolerably hot, but you can’t do that over an entire subcontinent. You can’t build that dome. I mean, you can see this process already happening, just with air conditioning. So air conditioning demand is one of the biggest drivers of increased energy use and increased coal burning in India, for example, and other places too. So maybe that’s a kind of feedback, a smaller scale version of the feedback you’re talking about. Arden Koehler: Yeah. It’s interesting. It’d be like a human feedback loop or something. Rob Wiblin: I think one part of what Arden is gesturing at is that, if in 50 years’ time, people see that the world is on fire or things are going horribly because of climate change, some people have the intuition that that will cause us to stop burning fossil fuels, because we’ll have seen by that point how reckless it is to do so. But I think that underestimates human stupidity. I’m not very calmed down by that, because the international public good problem will still be there. The fact that the incremental emissions that you do at that point in order to deal with that situation, that’s going to cause problems more in the future than it does right away. I could totally imagine a situation where the world is falling apart and people maybe even want to burn fossil fuels even more because they feel desperate to find some way to cope with the situation and that just makes it worse. Arden Koehler: Or they’re like, it’s already so bad. There’s some sort of defeatism that could happen. Rob Wiblin: “What the hell?” Let’s return to this famine issue. So it sounded like you thought that, in some of these three or four degree warming scenarios, we could end up with food production halved or more in a given year, which is perhaps a bigger impact than I expected. Maybe try to convince me that that’s how big the effect could be. Mark Lynas: Well, you just need to look at the scale of the issue. I mean, in terms of food production, it’s only a few hundred millions of tonnes which make up the food trade, which is only a few percentage points of the overall production. So you don’t have to lose very much to eliminate all of the traded commodities of rice and corn and wheat and things that now constitute the majority of the calories for importing countries. Going back to classical economics, you’ve got this specialisation where countries which can produce food do, and then trade it for other things which are produced in countries which have a better capacity to do other things. So the other thing you’ve got to think about is, to what extent does that drive us into the Ricardian production model, where everyone closes their borders? You only think about your own population, you ignore famines abroad. Clearly, when politically, leaders are responsible to their own populations, that’s your main motivator. You don’t care whether people are dying in faraway countries. And, like I say, that did begin to happen in 2008. The first response was to close borders and to stop exports — both in order to protect the price, and to protect supply at home. Because if you’re exposed to international markets, then you get increased prices, even in a domestic situation where you produce enough food yourself, of course. So that dynamic between globalisation and a return of protectionism, I think, would be a huge driver of what actually happened in terms of food security globally. David Denkenberger on low-cost, low-tech solutions to make sure everyone is fed no matter what David Denkenberger: Certainly a concern here is that our initial modelling is assuming global cooperation. The UK has a lot of potato seed, and they wouldn’t be able to grow them themselves — would they actually trust another country to grow the potatoes and then give them back more food than originally? So ideally we could actually talk about some of these agreements ahead of time so that people could be pre-committed to it. But certainly we are concerned about cooperation breaking down. We’re going to be moving into more regional geographic information systems (or GIS) analysis, looking at the resources of individual countries from a resilient food perspective, and then actually working out what would happen if international trade turned off — it would be much worse. What we’re hoping to do is use that information to convince governments to actually cooperate. In the past — say 2007, 2008 — the actual food production shortfall was less than 1%. But because of countries doing export bans, rice price went up three or even four times. So it’s a major risk, restriction of trade. There’s even potential of restricting more than just food trade, if we lose that trust and cooperation. And that would be just catastrophic, because then you lose energy trade and minerals and components — and the supply-chain issues we’ve seen in COVID are nothing compared to that. Rob Wiblin: Right. I guess we’re considering a bunch of different scenarios here. I suppose with a volcano, probably almost no infrastructure is destroyed, all the ships are still running. In the nuclear war case, a whole bunch of stuff is destroyed, but still a remarkable fraction is still online. I guess you have to model these quite differently if you’re trying to figure out the threat to international cooperation in each instance. David Denkenberger: I think in the case of a supervolcanic eruption, you could certainly have continent-scale destruction of infrastructure. Similar with an asteroid impact — a similar order of magnitude, I would say. However, there are important differences. One reason why nuclear winter is significantly worse is because it’s black particles going into the stratosphere, and those absorb the sun and then they’re lifted higher, so they can stay for a decade. Whereas with the volcano, it’s sulfate — more whiter particles — and they fall out faster. But then also if you have a nuclear war — Rob Wiblin: Someone’s been annoyed with someone else. There might be some preexisting tension. David Denkenberger: Yeah. Of course it could be accidental, but still there’s an enemy involved, and so cooperation would be even more challenging. Rob Wiblin: Do you have an intuition for if there was a supervolcano eruption, whether humanity would mostly pull together? My sense was cooperation during COVID-19 was pretty good, and it could be even better during a volcano situation. David Denkenberger: Potentially. I’m very concerned that if people don’t know about resilient foods then they could conclude that most people are going to die. It could be an incentive for countries to do very bad things, like steal food from your neighbouring countries. So I am very worried about that, and that’s why I want to get the message out that we could actually feed everyone if we cooperate. One of the things we’ve done is look at the current costs of these resilient foods’ production. Because many of them are already in production, and that gives us some idea of how cost effective they might be — though of course the actual catastrophe could change things. And in the catastrophe, it might not be so much that it is dollars that we’re talking about, but how many resources go into it. And I think that’s correlated now with the current price, and so that allows us to prioritise. And one of the technologies that came out as promising is turning fibre or wood or cellulose into sugar. So we looked at constructing factories to produce this sugar very quickly, but what looks to be more promising is taking an existing factory that has a lot of the components we need and repurposing that to produce sugar. One of the most promising we found was a paper factory, because it already takes wood, and it takes a lot of energy to grind it up and do that pre-processing step. And then it’s not that much more work to break the cellulose into edible sugars. Rob Wiblin: OK. So basically, should there be a nuclear winter or terrible volcano or something like that, we would still have a whole lot of wood and other plant matter that is just out there in nature. And there’s a lot of energy embedded in that, but the problem is humans cannot eat wood, so we need to find some way to make it digestible. And you’re saying it’s possible to turn it into the kind of sugar that we would normally eat. Like it’s possible to break it down into glucose and then we can eat that? David Denkenberger: That’s right, and there are actually a couple startup companies that are trying to turn fiber into edible sugar. That’s great, because they’re doing some of this research, but they’re just not thinking how we would do it fast in a catastrophe. And so that’s the type of research that we’re looking at. Rob Wiblin: So what’s the process for taking a bunch of wood and then turning it into something that people can eat? David Denkenberger: Basically a lot of grinding in the beginning, breaking it up into pieces. We call it “lignocellulosic material,” and that comes from the cellulose, which is basically lots of sugar molecules stuck together. There’s also hemicellulose, which is similar, and then there’s lignin — which you can’t really do anything with lignin, so you need to separate those components. And then you apply an enzyme to break the cellulose into sugar and the hemicellulose into other sugars. Rob Wiblin: Is this with bacteria or do you use chemicals to break it down? David Denkenberger: It’s typically done where you’ll purify an enzyme that is produced by an organism like a fungus or bacteria, but then it’s done without that organism in a bioreactor. Rob Wiblin: OK. So I guess fungus eats logs, and there’s probably bacteria that eat logs as well. And I suppose they also probably can’t directly absorb cellulose or anything like that, so they themselves have to break it down into sugar somehow, right? And if I remember from my biology class, fungus extrudes some sort of acids or other compounds that break down wood into something that then the fungus can absorb and use to get energy. Are we to some degree mimicking that process to make something that humans can also eat? David Denkenberger: Exactly. As we mentioned, mushrooms are one way of turning wood into food, but it turns out they’re pretty inefficient. And so if we can just grab the enzymes from them and then turn all that cellulose into sugar, or nearly all, we get a lot more food out of it than with the natural process. Rob Wiblin: That makes a tonne of sense. How expensive would this be? Is this a way that we could plausibly make food at an affordable price today? David Denkenberger: Amazingly inexpensive. Even though we’d be doing this repurposing with 24/7 labour — we would have to pay more for that — and even considering the fact that we wouldn’t run these factories as long; we’d probably only be running them for 10 years during the catastrophe. That increases the cost, but not that much. And so if you were to feed one person all of their calories — which obviously they wouldn’t eat all their calories, but it’s a way of visualising it — it’s only about a dollar a day from cellulosic sugar. And that’s our target here, because we’re trying to feed everyone no matter what. We want to look at those resilient foods that are inexpensive. Rob Wiblin: Makes sense. What else can you give us an update on? David Denkenberger: Well, given that we might be able to actually have some sunlight here, one thing that we didn’t get to investigate in the book, but now have since, is seaweed production. And we found that it’s very promising. There are several species of seaweed that can still grow 10% per day, even with the lower light levels in nuclear winter and lower temperatures. Rob Wiblin: Even up to adulthood? They just keep growing at 10% a day? David Denkenberger: They just keep growing. So basically you have these long lines that are floated by these buoys, and you start with little pieces of seaweed and they grow and then you just chop off the growth and then they just keep growing. It is amazing. And it’s only certain species, and right now they’re often limited by nutrients. So at least in the near term, we would still have those nutrients. I talked last time about the potential overturning of the ocean — that if you cool the upper layers of the ocean, they sink and bring up nutrients. So it turns out that is not as large as I thought it was going to be, so the actual production from fish is not going to go up as much as I thought. But with seaweed, it’s just much more efficient, growing it directly, than having algae grow and then feeding it to fish. Rob Wiblin: Right, right, right. You don’t need an extra trophic level. So the point there is that during a nuclear winter, say, if agriculture’s interfered with, the atmosphere is colder, that cools the upper layer of the sea, and then that tends to sink down, which then creates this upwelling of nutrients that are on the ocean bed. And then that basically provides fertiliser to seaweed to grow faster than it would normally, so potentially you can get seaweed growing even faster than it was before. David Denkenberger: Yeah, potentially. I think what they’ve found is that what they call the “net primary productivity” — that is, how much biomass is produced per year — will still fall in a nuclear winter, but it won’t fall as much as on land because of that nutrient enrichment. Rob Wiblin: I guess also one benefit that anything under the ocean has is that the temperature falls much less under the sea, because the water buffers the temperature change. So while you might get a lot of plants on the surface dying from frost and things like that because of the winter conditions, under the ocean the temperature might only drop a couple of degrees. David Denkenberger: That’s right. Rob Wiblin: What kind of inputs do we need in order to grow much more seaweed? I guess you’ve got to be around a coast, and then you were saying to grow seaweed, you attach it to ropes? It’s a bit like rope-grown mussels or something like that? David Denkenberger: Yes. So it turns out we produce a lot of synthetic fibre for other reasons, like clothing. The main constraint here is twisting those fibres into ropes that we’re going to attach the seaweed to. We found that right now, we don’t produce that much rope — we would actually have to increase our rope-twisting capability by 300 times, which sounds kind of crazy. But it’s actually a really simple process, and people have done it in their garage with a drill, basically twisting these fibres. But these pieces of equipment are only like $10,000 and you can make a lot of rope, so it turns out it takes a very small percent of our manufacturing budget to make a lot of rope twisters. Rob Wiblin: I mean, most seaweed is growing out of the ocean bed on rocks and things like that, right? It attaches to something on the bottom and then it grows upwards. But that’s no good for us? I guess all of that seaweed that can attach to rocks on the bottom of the coast, that’s already growing, so we need to have some artificial environment that’s very conducive to seaweed growing. And I suppose the closer it is to the surface, the more light it’s getting, so it might grow faster if we are attaching it to ropes near the top? David Denkenberger: That’s right. So seaweed, why it can handle low light levels, is it often does grow 10 metres down in the ocean. But yeah, we want to have it near the surface in nuclear winter. Athena Aktipis on whether society would go all Mad Max in the apocalypse Rob Wiblin: As you said in the intro, your next book — Everything is Fine! How to Thrive in the Apocalypse — is about how we think about civilisational disaster scenarios and the effect that thinking about that has on us. What’s something that people worry about in the apocalypse or in a catastrophe that maybe they shouldn’t worry about quite as much as they do? Athena Aktipis: Well, the big one for me is this idea that as soon as something starts going wrong, that the fabric of society is going to fall apart and that everyone will just be like, “Every person for themselves.” Because if we look at what actually happens during times of disaster, people jump into action to help each other in this need-based way — even helping strangers, even going to extreme risks to rescue people. People just do this spontaneously when the shit hits the fan. And yes, if disasters go on for a really long time, if you have slow-burn situations, where people are starving and that lasts for months, for years — Rob Wiblin: Like siege warfare. Athena Aktipis: Right. Yeah. When people are in a state where they just literally don’t have enough food or water or whatever, our physiology starts to not function normally, and things can break down once you get into that famine territory. But if we’re just talking about acute disasters, and if we’re talking about situations where, ultimately, people are able to jump in and help each other and deal with fixing some of the problems that have arisen together, you see that in the first few weeks — especially after a disaster but before everyone is like, “Oh, now we’re going to kind of go back to normal.” So I think that people having the wrong assumptions about what happens in those moments is not only not supported by what we see in times of disaster, but it can also be really problematic. Rob Wiblin: It is a destructive attitude. Athena Aktipis: Yes. It doesn’t necessarily put us in a good place to be proactively managing risk together either, if you’re like, “Everybody’s going to turn on everyone.” Rob Wiblin: “Everyone for themselves. I’ve got to get my knife.” Athena Aktipis: Yeah. This sort of Survivor mentality, right? Like that show. In the end, it’s like everybody’s pitted against everyone else, even if they cooperate a little bit. Often, that’s not how it works. Usually, that’s not how it works. We humans survive because we cooperate and work together, and that is how we have survived forever. Rob Wiblin: How we got here. Athena Aktipis: Yeah. With hunter-gatherers, you don’t just have one hunter-gatherer foraging and hunting. They live in a group, and they share at a central place, usually around a fire with whoever’s gotten what. Rob Wiblin: That’s how our organism operates, basically. A single hunter-gatherer is a dead hunter-gatherer. Athena Aktipis: Yeah. I think a lot of that’s reflected in just how we’re set up psychologically and emotionally. We like eating with people. A dinner party is super fun. Or preparing food with people, or just spending time with others, creating things together. We really thrive on being social and taking care of others to a certain extent too. It’s something that a lot of people intrinsically enjoy. Really delving into that side of our human nature that desires to be interdependent and embedded and generous and helping others, and part of systems that are functioning well — where we’re there to back each other up and manage risk together — I think we like that stuff. The better we can understand the evolutionary and cognitive and emotional mechanisms that underlie that, the better a job we will be able to do to manage the risk of the multitude of apocalypses that we’re likely to be facing in the future. Rob Wiblin: Ways things can go wrong. Yeah. It’s been interesting watching the conventional wisdom on this shift over the last few years. I think back in 2019, people’s intuitions were really that even quite modest disasters would lead to the breakdown of law and order. Everyone’s just stealing from one another, just because of some relatively mild problem. Then I remember early on in COVID-19, some people were predicting there’s going to be like blood on the streets. It’s going to be chaos. Everyone is just going to look out for themselves. There’ll be a crime wave. And the exact reverse was true. Crime went way down. Violence went way down. I mean, to start with, people were saying home. It’s much harder to get into fights. It’s much harder to burglarise their house when there’s people in there. I think actually the US is almost the only country in the world where crime didn’t go down all that much. I think there were some kind of unique, US-specific factors there, although it did go down in the short term. But there’s been quite a big correction, and I keep hearing this point that people make that it’s remarkable how extraordinarily cooperative humans are when things go wrong. When everyone’s part of the same struggle, they’re fighting against adversity together, then actually we’re way more cooperative in that situation than we are just on a typical day while we’re on our commute to work normally. I wonder, is it possible that we’ve overcorrected in some way? Is there anything to be said for the idea that people can be uncooperative? Or is there any way to temper that? Athena Aktipis: Absolutely. It’s a matter of understanding in what ranges of parameters you get what behaviour. In situations where things are bad for a long time and people are starving, then you can get what we would consider the breakdown of society and families and all of that, because people are literally just on the brink of dying. Rob Wiblin: It’s quite zero sum. Athena Aktipis: Also, I think we don’t really know when you get to the point of people being in that starvation state, where they’re very likely to die, what of the behaviour that you see would we call “adaptive” from an evolutionary perspective — i.e., was selected because it provided an advantage — versus a byproduct. This is a very unusual state for the organism to be in, and there are things that are happening that are just the result of molecular pathways that are… Rob Wiblin: Accidental. Athena Aktipis: Yeah. We don’t know, and it’s not something that we can really study, obviously. But practically, what we see is in those situations that you do get a breakdown. The other situation where you definitely can have favouring of cheating is in groups where there’s a bunch of anonymity. They’re large. People don’t necessarily feel like they have a stake in the wellbeing of the group that they’re a part of. Rob Wiblin: There’s not peer monitoring, or perhaps justice done when injustices are committed. Athena Aktipis: Yeah. In those kinds of situations, individuals that are exploiting are going to do better. If people see that exploitation and cheating is happening, then that can very quickly unravel if the norms start changing about that. So I think we absolutely need to consider both sides of the coin here: What are the situations where cooperation not just makes evolutionary sense, but also is how people behave in practice? And then where are the situations where exploitation and cheating really are problems, both from an evolutionary perspective and a practical perspective? Luisa Rodriguez on why she’s optimistic survivors wouldn’t turn on one another Rob Wiblin: I think the typical story is more like people are at one another’s throats, or killing one another. It’s like zombie apocalypse almost, where it’s like the last person surviving gets all the food supplies. Yeah, what reasons do we really have to think that people would strive to get along and find one another? Luisa Rodriguez: I guess I do feel a bit “peace and lovey” when I talk about this, but I think you even pointed me to the research on how people react to crises, according to sociologists who have looked into post-crisis response. And again, this is hard to generalise from because it’s not actually a case where maybe the world’s going to end; it’s sociologists looking at how individuals and groups respond to tsunamis and hurricanes. It’s just almost unanimously not violent. Looting is a kind of famous trope, but almost never happens. Same with people fighting over resources. I think it did feel intuitive to me that there would be lots of violence because people would be facing death. But not only does that not look like the case empirically, it also just doesn’t look like a very… I mean, it looks like a strategy that might help some individuals survive if you just think about it theoretically, but it really doesn’t seem like a strategy that would be good for the majority of people to take on. Because most people will really benefit from cooperation that lets them grow more food. An individual will have a very hard time producing enough food for themself. And there’s a reason that we live in cities: it’s because we specialise and produce more stuff for the number of people. So cooperation has these clear benefits, especially in the context of agriculture. And insofar as there will be some selection for survival strategies, it seems like lots of people would benefit from taking this cooperation-y strategy. And there will be some cheaters or people who use violence to get a bunch of resources, but on the whole that’s not going to be a very persistent survival strategy, because you’ve got to have someone to steal from. Rob Wiblin: That’s true, yeah. Are there other case studies that we can point to where people cooperated to a surprising extent? Luisa Rodriguez: So both Hiroshima and Nagasaki suffered just horribly. I don’t think I fully understood before looking into the details, but basically, a third of the population died. I think the population started around 400,000, and maybe a quarter of those people died instantly. And then another set of tens of thousands of people died from radiation poisoning in the weeks after. So that already was horrible, of course, and shocking to learn, and especially to learn about the details. I just remember being really surprised that 90% of the city’s buildings were either totally incinerated or reduced to rubble. So just like this huge infrastructure loss. And then at the same time, I also learned that the recovery was just shockingly quick. So the analogy doesn’t totally work, but if you kind of imagine these as cities whose societies basically collapsed — how quickly they were able to recover is just really astounding to me. I think power was restored to at least homes that weren’t completely destroyed within like a month or so. Water pumps were restored within I think just a few days. Actually, maybe what surprised me even more was some intermediate services were back within the next two or three days. They had trains running on like day two. I remember learning that the bank, there was a bank where I think… God, this is awful. I think literally all of the employees were killed immediately, but the bank was able to reopen a few days later. And those services were actually just really important to getting things up and running again. Other things too, like telecommunications: they had phones back I think on day two or three. Rob Wiblin: It kind of beggars belief. But you’ve checked the sources on this? Luisa Rodriguez: Yeah. I think the thing that makes it less generalisable is obviously the fact that they just had tremendous support from very nearby cities. And those volunteers would’ve come in and literally, in some cases, hauled in new pieces of equipment. Rob Wiblin: Another interesting historical analogy that doesn’t quite fit, but speaks to the fact that people don’t turn on one another that much, is mass aerial bombing in World War II. There was this big theory in the military in the ’30s and ’40s that if you bombed a city a lot — like London, like Dresden, like Tokyo — that people would lose their morale and they would lose the will to fight the war, and it would cause the other country to surrender. Nothing could be further from the truth; it had no effect like that. Do we remember the Blitz in London as the time that people lost interest in fighting the Nazis? We do not, nor is that the case in any of the other countries that were bombed. They pulled together and they felt extremely angry with the people who were fighting them, and in fact, they coordinated more than ever in order to overcome this adversity. Something that’s different about that case is that you’re specifically fighting an enemy, an external enemy. But we saw some of that with the pandemic and in the war cases, which are some of the most destructive. Then you would have this thing, you could imagine in the post-nuclear apocalypse thing, that people would feel an enormous sense of comradery with their fellow citizens who have suffered this immense catastrophe. Luisa Rodriguez: Right. We are all suffering because of the same thing, whether it’s a nameable entity or another state or a pandemic. The other thing I have about this is, even if you get some groups where for whatever random reason there happens to be more violent people or something, and you actually do get some set of factors that means there’s violence — I guess this only applies if you have isolated groups, at least somewhat isolated groups — but it doesn’t really make sense to think of it as a single pool of survivors. If it’s many survivors, there’ll be geographic variation in what culture is like, just like there is now. If there are few survivors, they’ll probably be in groups that actually don’t contact each other that much, because like one’s trapped in Australia. I mean, maybe not trapped, but they’ve settled in different places and they’re not really exchanging much. And then I guess you have to think it’s a really dominant outcome that none of these groups will end up being cooperative. Rob Wiblin: You always just disintegrate. Luisa Rodriguez: Yeah. And I guess, I mean, you could have that view. It’s a really dark view. You have to believe that human nature is just really… But it just does not seem strategically robust. Rob Wiblin: Even historically, when you have a very aggressive group that tries to attack another one, they almost never try to exterminate the group that they’re conquering — they want to tax them almost always. And so it’s like, even if you’re a marauding group in the post-nuclear war world, what you want to do is get people into your control, so they farm and then you tax some of the agricultural surplus that they’re producing, ideally. The scenario where descent into conflict and just total strife and lack of coordination potentially makes sense is one in which there are no productive opportunities, where there’s no ability to farm or make food or really do anything. Luisa Rodriguez: Right. But then you’re dead anyways. Rob Wiblin: Yeah, probably dead. So certainly in the case where the infrastructure isn’t destroyed, and you don’t have a sufficient climate disturbance that agriculture becomes impossible. In fact, people will see productive opportunities all around them to scavenge things, to try to grow things, to move to a place where you can grow things. Even in places where it’s incredibly cold, and you can’t grow wheat or whatever, you can still do fishing. Luisa Rodriguez: You can still grow potatoes. Rob Wiblin: Yeah, grow potatoes. Luisa Rodriguez: Or if not, fishing. Yeah. Rob Wiblin: OK. So I think that might be where the folk sense of, “We would just descend into Mad Max” comes from, is the idea that there’s nothing that you can make. But I think that is empirically unlikely. Luisa Rodriguez: That makes sense to me. David Denkenberger on how resilient foods research overlaps with space technologies Rob Wiblin: Is there any potential to use space industries’ need to feed people in space as kind of a stepping stone for funding research and development into these different methods that then could be used for these resilient foods on Earth? Or is the space industry just so small that this wouldn’t be a big enough industry to really bootstrap any of these approaches? David Denkenberger: I think there’s potential overlap. It is at a quite different scale. I would say that one potential overlap with global catastrophic risk would be the really extreme scenarios where potentially everyone has been killed, that you can think of, “Can we make a refuge with 1,000 people that might be able to repopulate the Earth?” And Elon Musk is interested in doing this on Mars, so if we could figure out how to make an independent colony on Mars with fewer people, and with less expensive, less infrastructure-intensive food production, then I think that could have some existential risk benefits. Rob Wiblin: Yeah. I guess another option is doing that somewhere really remote on Earth, like under the sea, or in Antarctica or so on — where you’d also have to figure out how to feed people in a worst-case scenario. It sounds like you could potentially use a nuclear reactor to grow bacteria and then eat them — or otherwise just have stores of fossil fuels, or put it somewhere where you have access to fossil fuels — and then you could in theory eat that. David Denkenberger: Yeah, that’s right. So again, if you look at the plans for having an underground bunker — typically nuclear — the plans were to go through regular plants. And that’s really inefficient, so we could lower the cost of this significantly. Rob Wiblin: Yeah, so it was either you have this enormous initial cost of stockpiling all the food and making it big enough to store food to feed everyone for ages, or you’ve got this horrifically inefficient process of converting electricity into human-edible food that now we can do 10x better on. David Denkenberger: And the other thing if you use the stored food is that you’re breathing out carbon dioxide and you need oxygen. Whereas if you have a system that actually grows food — either plants, which is of course inefficient, but these space-based resilient foods if you want to call them that — they can act as the life support system, because they would actually take the carbon dioxide from the astronauts to make the food and then they produce oxygen. I guess another extreme scenario is a runaway climate change scenario, where it might get too hot for plants to live. So that’s another scenario where having these types of “space foods” could be a good food source. Rob Wiblin: Interesting. God, it could be a very bizarre future, I suppose. Mostly just eating bacteria grown in electricity. Again, it sounds like absolutely bizarre sci-fi stuff. I kind of can’t believe that it actually works, but sounds like in principle it could. Zach Weinersmith on what we’d practically need to do to save a pocket of humanity in space Luisa Rodriguez: So what’s a good argument for why we should settle space? Zach Weinersmith: Something we call “the cathedral of survival.” And the idea is essentially that, in the very long run, it would be good to have a second reserve of humanity in case something goes drastically wrong on this planet. Now, as I said, I don’t buy that as a short-term thing, but there is at least a kind of plausible, this is something that humanity should have worked toward in the next 100 years. And we do argue, for reasons we might get to, that it is a project of a great period of time for research and development. Then there’s no reason not to put in the first bricks of the cathedral now, even if we’ll all be dead when the project is completed. Luisa Rodriguez: But I guess you’re still not totally convinced it’s even realistic, at least not on the timescales that lots of enthusiasts seem to have in mind. Hence the subtitle of the book, which I thought was great: “Can we settle space, should we settle space, and have we really thought this through?” And a big reason for that is that it’s just really, really hard in lots of ways that people I think don’t appreciate — myself included. First, I guess there’s the fact that space is a terrible environment for a human body. Can you talk about why that is? Zach Weinersmith: Almost anywhere in space, the moment you step outside your suit or ship, you die. Nontrivial, right? And definitely in the places we are likely to go that is true, and that’s just the deal. To give just a quick example of why space is a really fussy place to live. To me, this is a fascinating detail. Spacesuits are kept at lower pressure than spacecraft. And the reason is that it’s hard to operate, it’s like a balloon inside, if it’s at full pressure. So we keep it at lower pressure. Just makes it easier to bend and operate the suit. In order to do that, you have to up the oxygen concentration so your lungs can still get enough. And we don’t like to do that in the craft because on both sides of the Cold War, there were tragedies related to pure-oxygen environments. Most American audiences know Apollo 1, but in the Soviet Union, there was a very similar incident with a trainee named Bondarenko. So it’s a real problem. So the joke we have is, if you’re on your Mars hab and your friend is dying outside the facility, you literally can’t go save them. Because if you put on your suit without pre-breathing oxygen for a while, you’ll just get the bends, like a diver surfacing too soon. So you’ll just wriggle and die while your friend also dies. And actually, the only three guys who’ve ever died in space were Soviet cosmonauts — Patsayev, Dobrovolsky, and Volkov — who all died due to a valve opening when they were moving toward descent. So it’s not a little thing. Everything is going to be annoying. The next thing is radiation. Radiation is real bad. I won’t go into the details, but the short version is that in space, you get higher doses of different kinds of radiation than you get down here, and with unknown consequences. Radiation is poorly understood even on Earth; it’s even worse understood up there. The data we have mostly comes from space stations, which are still in the Van Allen belts, so they get more radiation than we get down here. But it’s still different. We only have a tiny amount of data from the guys who got sent to the Moon, and they weren’t there for very long. It was on the order of weeks total. And it’s just, you know, we don’t know the effects of this stuff. And it’s scary. Probably the main practical effect is you’re going to have to bury your base under a lot of dirt. No glass domes for you. There’s a bunch more detail in the book if people want. Then the big thing probably is microgravity. So in the International Space Station, you experience free fall, as if you’re in zero gravity. And reliably, that degrades bones. So we know bones — especially like hip bones; bones you don’t use a lot — lose something like 1% of density per month. It’s crazy. And that’s with intense exercise, like six days a week on like a treadmill with a spring to pull you into it, and you still have this loss. Similar effects on muscles: they degrade over time. It’s considered very impressive if, when you come home, you can walk. And there are other reasons for that, but one weird thing that happens in space is when you lose that gravity, you get a massive upward fluid shift. So you lose like 30% of the volume in your legs, and your face is sort of just poofy like a baby. They actually call it puffy face. It happens. The sinister side of it is it’s probably associated with this phenomenon where astronauts tend to come back with worse vision. And in fact, astronauts over 40 are sent up with what are called “space anticipation goggles,” assuming they’ll come back with it. This happens even on short trips. As I recall, I think it’s permanent, or at least semi-permanent. So it’s a problem. And what’s most scary about this — you can always get glasses, I guess — but it’s possible that’s actually an early sign of broader nerve damage. So there’s equivocal evidence of cognitive negative effects on astronauts. We don’t have enough data. A big thing underpinning all this is that we don’t have anyone who’s gone longer than 437 days. I think the next person down is about a year. And it’s only like half a dozen people have gone that long. Most people are much shorter. So we really don’t have any kind of really long-term data. And by the way, a Mars mission is on the order of two to three years. Luisa Rodriguez: Yep. So another challenge is that if we’re actually going to settle space, we need to make new people in space. You said we can probably have sex in space. Is the hard thing gravity? Zach Weinersmith: I’m debating how graphic to get here. It’s funny, one of the things we did for this book is we read a lot of old books forecasting the future of space. And there’s a sort of golden age of talking about sex in space, which is from somewhere like 1960 to 1980. I think it was just the right time. And it’s like Arthur C. Clarke, I don’t remember if we put this in the book, but he had some quote that was like, “Space is about to become more erotic.” And you’re like, oh god, Arthur. So yeah, space is Newtonian if you’re in microgravity; it’d be easier on the Moon. But basically that means if someone bumps into somebody else, they both go flying. So, again, from this period, there were attempts to figure out how to manage that. We found two different proposals for what one guy called an “unchastity belt,” which is a sort of elastic waistband for two. And it’s funny; you hear that and you’re like, “OK…” And then you think, “But wait. Like, how, exactly?” And then you’re like, “Maybe I’m just not gonna…” Yeah. And then there’s another one called the “snuggle tunnel.” I forget who proposed that, but it was basically, imagine a large pipe with holes in it for ventilation, because CO2 tends to build up in your mouths if there’s not ventilation. And I could go on, but… Luisa Rodriguez: I’d argue that space did not get more erotic. Zach Weinersmith: No, it has not gotten more erotic. The dream of Clarke has died. I mean, worth noting that space notoriously kind of smells bad. And by the way, you change undies every something like four to seven days. The mood lighting is not present. There’s not a lot of private space. But when I say it probably could happen, basically I’m referring to anecdotal reports from men who said they were up for it. We found two men admitting to space onesomes. So that’s what I mean by that. Whether you could actually bring the baby to term, I mean, who knows? I say the human body is not designed for zero gravity. But you could note, and this is kind of goofy, but we looked up, does anyone do headstands while pregnant? And apparently this comes up in yoga, and it’s OK. And so apparently foetuses could do negative one gravity. I mean, they are kind of in a neutral buoyancy tank. So maybe it’s fine. I’d be more worried about some sort of cellular-level process that depends on gravity in some way or another that we’re not thinking about. But, you know, it is the case that evolution would at least design it so a woman could trip and fall and the foetus would be OK. So clearly you can alter the sort of acceleration that’s being put on the foetus, to say it in a weird way. So that’s why I say it seems plausible that the baby might be able to come to term, unless something we don’t know is happening. There are also little other off-ramps. The atmosphere in a space station is very different from what we get on Earth, so they tolerate a much higher level of CO2, because they have to. It’d be very expensive and mass-consuming to have a bunch of CO2 scrubbing going on. Ideally on a Mars base, you’d have a lush ecosystem to manage that, which is a tall order. I’d be willing to guess you could bring the baby to term, maybe. Although, you take that suite of stuff I just described and throw it at a foetus: who knows? It’s obviously never come up on Earth. But who knows? Probably, maybe. Who knows? But the real question… You know, often when this comes up, it’s like, “Can you have sex? Can you have babies?” But in order to have a settlement, babies have to develop through all the stages of human development to be adults who can have children — and that’s where it gets really scary. So you describe all these medical things: imagine applying them to a kid whose bones are developing, whose vascular system is developing, whose brain is developing. We really have no idea. And so the scary thing is, it’s not that we can’t get this data; it’s that without this data… You know, we have Elon Musk saying we’ll be there in 30 years. No one is collecting this data. There’s really haphazard experiments over time. There’s not much agency funding. As far as I can tell, there’s no funding from crazy billionaires. You know, we need this data. It’s going to be very painstaking to get — arguably unethical to get, because you at least have to experiment on primates before you’re willing to do it on human women. And it’s hard to imagine how you could even get good data unethically in a matter of decades. It should be a problem we’re pursuing now, if we’re really serious about space. Like, if tomorrow we found out Earth was going to be dead in 100 years, this would be part of the crash course, a big part of it. What also worries is that if you do execute on this settlement, and you’ve got kids being born in these conditions, where you would expect a higher than normal rate of abnormality — you know, kids with cognitive deficits, physical deficits, who have trouble contributing to this hostile environment — where they can’t get any care. On Earth, when you have special-needs children, some of us have complained about government services not being quite adequate, but there are at least services; there are ways to take care of these human beings. And that wouldn’t be true on any kind of medium-term Mars settlement. And what’s scary is we found three different quotes from advocates in this community willing to say some version of, “We’ll just have to have natural selection do its thing” — which, you’re like, holy crap, this is like a horror science-fiction novel. Luisa Rodriguez: That really is horrifying. Zach Weinersmith: They’re just saying the quiet part out loud, though. This is what would happen if tomorrow you snapped your fingers and there were a million people on Mars: you would be doing a mass experiment on babies, the result of which would probably be a large number of children who couldn’t be cared for. So you know, I always say we’re concerned about space ethics. People are imagining we’re going to be like, “Do you really want capitalism on the Moon?” And there are people who want to bark up that tree, but we’re like, what we don’t want is vast experiments on babies for no reason — which seems to be a reasonable ethical posture for anyone, anywhere, ever. Luisa Rodriguez: And then child development is basically just all the stuff we already talked about. Seems hard enough on an adult human, and then child development seems hard and complicated. One unsettling fact from your book is that there was a startup called SpaceLife Origin that was announced in 2018 and their goal was to have the first human birth in space by 2024 — so this year we could have had a human birth. But in 2019, their CEO left, citing serious ethical safety and medical concerns. So this does just seem ethically incredibly fraught and we have to do really like years and years — because child development takes time; you can’t rush it — of experimentation somehow ethically. Zach Weinersmith: People have this idea that space stations are kind of organised, scientific projects that are going down the line on big questions — and they’re not. Space stations are built for politics and then cool scientists jam stuff onto them. The result of that is if you’re someone like us who’s had to do research, you say, have we solved space psychology? There’s not going to be a textbook that goes through the 50-year experiment that we’ve been honing in on issues. What you have is 50 years of grab-bag stuff. And it’s the same in reproduction. So quail eggs have gone to space, and some rat systems have gone to space. Various plants have gone to space. Geckos and certain types of fish have gone to space. But there’s not what you would want. What you would want is something like a module devoted to rats having generations in space. And that would be a start, right? You still wouldn’t say humans could do this, but that would be what you would want to start answering these questions, to see if you get problems over time. And I will say some experiments seem to go just fine; some don’t. We see in some cases cellular deformation. One case had stillbirth — I think that was with rats, I want to say — when they got home, though. And some have head deformation and stuff like this. And what’s worrisome about that is we don’t know the culprit, right? There are many altered conditions in space. So you can keep a control group back home; that’s great. But to give you one example, we observe in males lower testosterone, and in females lower oxytocin. Very scary for reproduction. And maybe it’s just stress. I think we talked about this in the book, that the early experiments suggested that I think it was rats stop cycling: female rats stop cycling when they go up. And turns out that’s not true. You just have to wait longer. So probably what happens is you’ve just put them through enormous stress. And at least with a human, you’re like, “By the way, we’re going to space now.” A rat is just like, “What the hell is going on?!” But it’s like, not only are there scary things happening, but we don’t know the precise cause. So that’s why I say these experiments would take at least decades, because you’d have to start from simpler systems, and you’d have to slowly, ideally, work your way up to something like a Moon base with human beings in it, somehow ethically. Luisa Rodriguez: OK, so the next challenge is creating self-sustaining habitats wherever we go. How hard is that? Zach Weinersmith: Super hard, but super awesome. I’m an avid gardener, so this question of closed-loop ecology just brings me joy to talk about. So let’s spend a lot more money on it. Just ’cause Zach thinks it’s cool. Luisa Rodriguez: Because it sounds fun. Zach Weinersmith: OK, if you’re on Mars, say, you would much rather grow food on site, right? Probably in the form of plants. We could talk about animals later if you want, but likely plants. And plants are good for a variety of reasons. They help you ameliorate Mars soil. Mars soil is dead, awful stuff. But if you have organic matter and you’re careful about the chemicals, you could, in principle, turn it into real soil that you could use. Also, plants generate oxygen. They can be part of bacterial systems that help cleanse water and turn waste into healthy, safe stuff. And the question is how to do that, when on Mars you’re going to be in a sealed container, most likely. Almost certainly. We never do this on Earth: the greenhouse is almost never sealed, with the exception of times we tried to do it to learn about how to do it in space. The problem is, we haven’t done this very much, and we’ve never really done it at scale. On a short document, you could bullet point every time it’s been tried. The first couple attempts were by the Soviets. There’s a set of systems called BIOS: BIOS-1, BIOS-2, BIOS-3. And they’re kind of funny. BIOS-1 is, you can imagine just like an apartment or something, but there’s a huge algae vat running chlorella, that’s good at generating oxygen. So they tried this. One problem with chlorella is, in theory, you can get a lot of fat and protein from it. Apparently, it just tastes bad or is kind of sad to only eat algae. Like it’s a bad sci-fi novel. But also, they didn’t fully close the system. I think they brought in meat, and I don’t think they had all the oxygen. I’d have to look at my notes, but it wasn’t fully closed. It was kind of small, still kind of cool. They did BIOS-2 and BIOS-3. They kind of expanded the system, added more variety of plants. Never fully closed though. The phrase we found was, “Siberians have to have their meat.” So typically, you don’t necessarily want meat, because meat is a very inefficient source of calories. So generally, for that same reason, you don’t want them in space — because, if you imagine you have a bowl of soybeans and you could either eat it or feed it to a cow, obviously the better call is the soybeans. The best experiment we’ve done to date is Biosphere 2. If you’re wondering what Biosphere 1 is, Biosphere 1 refers to Earth. Biosphere 2 is kind of a cute name. So this was a facility kind of made by crazy people in Arizona in the early ’90s. It was a sealed greenhouse — and sealed, they claimed, as tightly as the Space Shuttle, which might surprise you is not perfectly sealed, but good enough. But a really tight seal. And it was a 3.14-acre campus. Just absolutely enormous. For people who don’t have an offhand sense, a football pitch is one acre. Huge, right? And eight people went in, and eight people came out two years later. It’s often remembered as a failure, but it did kind of work. Luisa Rodriguez: That’s pretty awesome. Zach Weinersmith: I think so. I think it’s awesome. And it’s easy to make fun of. They screwed up a lot of stupid stuff. At one point, they had to pipe in oxygen. There’s this story, and I’ve repeated this myself, which is that the concrete was pulling out oxygen. I was just talking to a guy about this, and it actually pulled out CO2 — but that’s kind of the same problem, because the oxygen is bound up in the CO2, so when that happens, slowly oxygen is leaving the system. So at one point, they’re not quite suffocating, but getting listless and it’s hard to climb the stairs. So they had to pipe in oxygen, which is actually a big controversy. Like, “Do we do this? Do we wuss out and breathe?” So that’s a problem. Food was a problem. They lost a lot of body weight. I think for women it was like 10%; men was like 18%. And they weren’t chubby to begin with. And a big thing — but actually, this seems small when you start to think about a realistic space base — the numbers are in the book, but they spent something like most of their time, six days a week, were just survival. Just growing enough food and processing it to live and eat subsistence. You know, it’s funny, I always think, I have a great-grandfather who was a farmer, and probably if you told him that they spent most of their time on the farm, he’d be like, “Yes, obviously.” Luisa Rodriguez: Yep. But now I’m like, “They what?!” Zach Weinersmith: They what? Right, right. Which you should think about every time someone’s like, “…and they’ll run the nuclear reactor, and they’ll cleanse all the soil, and they’ll XYZ…” You know? But again, it did basically work. And there’s a lot of stupid stuff they could have corrected for the second run. So they went two years: like, a lot of the trees weren’t even fruiting yet. If you’ve grown fruit trees, you know they can take three or five years to bear fruit. They had some animals that were just basically stupid calls. They got the wrong species of chickens and pigs and things. But you know, you can imagine a world where it kept going and it was well run, and we had like 30 years of running this, and we’d know a lot, right? We don’t. It’s like with babies, where what we really want is longitudinal data over a great amount of time. You know, there was a concern at the beginning that this sealed ecosystem would just turn into like green goo, and it didn’t. But there were problems. They had mildew; they lost all of their beans — which, as you might imagine, are one of the best sources of protein — and so they had to eat fodder beans meant for goats, and it was bad. And this is stuff that probably could be fixed, but maybe not, or maybe not without a lot of knowledge about how to manage ecosystems. So that is easily, easily the biggest, most elaborate experiment like this we have ever done. For a million people, Elon Musk is saying that you need a greenhouse the size of double Singapore. We do not have a science for this. The last textbook is from like 2003. What you really want is to run this experiment many ways, many times. To give you a trivial thing, Biosphere 2 only had a half-acre for intensive agriculture. By the end, they were colonising the other biomes with more ag stuff. And so probably what you really want, if you’re not run by slightly crazy artsy ’90s people, is to just start with intensive ag with an eye toward oxygen production. But again, we don’t know. You want some sort of massive software system that can think about this really intricate ecosystem design. We don’t have it. And the last thing to say on that is: like with babies, you can’t just get the data, right? You can’t even throw money at it. You should, but you’re not going to go fast, because it’s an ecosystem. You’re moving at the speed of saplings, right? It’s going to take decades. Nobody is throwing the right amount of money at it. We have little tiny systems. Again, if Jeff Bezos or whoever wants us on the Moon or Mars in 30 years, we need the answer to this, unless you want a massive armada bringing food all the time at enormous cost. Lewis Dartnell on changes we could make today to make us more resilient to potential catastrophes Luisa Rodriguez: I was interested in talking a bit about practical adjustments we could make now to make society more resilient in the future. We can imagine a scenario where humanity faces some severe shock, but it’s on the borderline between a situation where we experience a series of cascading failures versus get our act together while we still have access to pre-disaster technology and rebuild in a kind of orderly fashion. In that kind of case, what are some ways that we can organise things now that might make a difference between those two futures? Lewis Dartnell: I think the idea here is: can you engineer the situation so that you fail gracefully? If you have just experienced a catastrophic shock — rather than the whole system snapping and fracturing completely — can you cushion the fall slightly, or catch your fall, so you don’t regress too far before pulling yourself back up? Without wanting to bat away the question, because I do think it’s a good one, I wouldn’t know how to go about answering it. Because again, I think it’s so dependent on what was the event in the first place? What is the scenario we find ourselves in? How many people have died? What nation-states are now at war with each other over the resources that they are looking for for their own populations? We’ve already talked about saving not just libraries of useful information that tell people how to go back to slightly simpler states and slightly lower technological levels and pull them back up. But repositories of the most useful tools as well — things that, again, link back to appropriate technology: things that you could repair perhaps at the village level, rather than having to send back to a factory in China to get repaired. Start breaking some of those ties of the global transport of things around the world and make it a bit more local. To link this to current affairs, the EU as a whole — and the world in general as well — are starting to address where do we get our oil from? Do we continue getting it from Russia? Because this is now very problematic. Do we try to sever our connection with Russian oil and try to find it elsewhere, or — and you can probably guess what my point of view is — do we take this opportunity to fundamentally change the question and look at how we can not use oil at all? Can we go much more towards renewables, to hydropower, to wind, to solar? Can we break our reliance on something that we get from another nation-state, which they then basically use as political leverage? A lot of these topics and current affairs do link very, very directly to people looking at catastrophe studies and rebooting. They’re different aspects of the same coin. Rob Wiblin: I think there’s some stuff that we do today, which is slightly cheaper for us now, but looks catastrophic from a resilience point of view. One example is it seems like we’re making more and more things like tractors and cars internet connected, in such a way where these items begin to break down and stop functioning if you don’t have access to a computer to debug them. Or if they can’t get regular patches from the internet, they start complaining and then break down. People after the apocalypse might not be able to get their cars working. We could slightly do some better patent control for Tesla’s vehicles by having their software updates all the time. That’s one where I wonder, legally, maybe we should just say all of this essential equipment has to be able to operate even if it never connects to the internet again, because there’s a possibility that the internet will disappear and we still need to have tractors. Lewis Dartnell: I think that’s a great point, Rob, and it links quite closely to the right to repair, that there’s been a change in the legislation in the UK. That absolutely is a great idea, because again, it breaks the bond slightly between the manufacturer and the person that sold something to you — that you should be able to take it to anyone to repair, if not have a go at repairing it yourself. There’s a whole bunch of wonderful repair cafes and organisations set up to show people how to fix things when they break, to reestablish that connection between ourselves and our technology. So you’re right, Rob, there are low-hanging fruits. There are little things we can start changing. Rob Wiblin: How big of an issue do you think it would be that most farms are currently sowing seeds where you can’t then harvest the seeds of those crops and then sow them again, because they basically just become rapidly sterile, for reasons we won’t go into. It makes me wonder, after the apocalypse, once these fantastic factories that are producing these very high-yielding but sterile seeds disappear, will we have enough seeds to sow the next crop that we’re going to need? And will people be able to find the seeds that they need in order to resume agriculture? Lewis Dartnell: Yeah. I don’t think the solution is to stop using these hybrid crops, because the reason we use them is they are so fabulously high yielding — and we need that nowadays to feed the world population that we have. So I don’t think the solution is to change how we are currently doing things. I think the solution is more to just be sensible about this and keep a repository of a simple alternative to go back to — to have like a saved file on our computer in case we ever have to go back to it. That would be as basic as the ancient Egyptian technology of having some granaries, some warehouses, stocked with heirloom crop seeds, not these hybrid crop seeds, that you can crack open and go back to if you need to. If you needed to have sort of regenerative farming, where you can keep back seed corn and plant it next year, rather than going back to the hybrids each time. Because you’re right. You really don’t want to start losing crops, because each of those represents technology which has taken thousands of years to develop, of selective breeding. Christian Ruhl on thoughtful philanthropy to reduce the impact of nuclear catastrophes Christian Ruhl: So actually, Founders Pledge we are putting our money where our mouths are, and we recently made a three-year grant of $2.4 million — which by the way, that puts us as a major funder in the field now — which is crazy, right? That shouldn’t be the case — to this project called “Averting Armageddon” at Carnegie, led by James Acton and Ankit Panda. Basically, it asks a lot of these questions about war limitation and escalation management over three years, thinking really in depth about what can we do here? They’re asking what limits can we actually come up with? Which ones might stick? What are the technical obstacles, things like hotlines failing, and what can we do to make sure that these things go better? Luisa Rodriguez: Yeah, that does sound like a great grant to make. And just to reiterate that it is wild that you are now a major funder in the nuclear space with only $2.4 million. That brings us to this weird fact that very few institutions work on these kinds of right-of-boom interventions. Christian Ruhl: Yeah, that’s right. In nuclear issues, we’ve heard anecdotally that right-of-boom interventions are only a small fraction of the total amount of work. We looked into this a little bit, tried to quantify it, and it looks like it’s about 3.3% or so of philanthropic funding, at least over the last 10 years or so. It’s kind of challenging to actually come up with an estimate because the right-of-boom / left-of-boom distinction is a little bit fuzzy. But we searched through the funding databases, tried to find any grant that could plausibly be considered right-of-boom, came up with this rough 1-in-30 number, and then ran that by a bunch of nuclear experts. And they agreed, they said yes, literally nobody works on this. Most of them thought that estimate was probably too high. Luisa Rodriguez: That’s truly wild. I can imagine a few reasons for this, but before I make guesses, what do you think the most common ones are? Christian Ruhl: Great question. So I pointed to a report called Philanthropy to the Right of Boom; it’s also in the nuclear report: why do we care if something is neglected? Many reasons, but two of them are: there might be low-hanging fruit to be found, so policy advocacy might be easier when a space is less crowded; but importantly, neglectedness could also be an indicator that something is actually a bad idea, and there’s a reason people aren’t doing it, right? So we need to look into the reasons that something is neglected. I think there are a couple of bad reasons. So this way of thinking has historically been associated with kind of hawkish Cold War thinking, and it still carries that ideological stain. So it’s like a political, PR, ideological problem. We interviewed a lot of experts about this, and one thing that surprised me was the answer was always like, “Oh, it just kind of has had bad vibes since the Cold War. We don’t do this.” Luisa Rodriguez: Wow. Christian Ruhl: That’s what it comes down to. Luisa Rodriguez: Somehow I feel like I get it. I don’t believe it when I inspect it, but when I’m imagining the US government investing in something like systems to make sure they can communicate with another country in the event of nuclear war, I’m like, “How dare they plan for a nuclear war to happen?” It just feels kind of outrageous. It feels to me upsetting and yucky, just really unpalatable that anyone’s even thinking about this. I do kind of get it, but I would hope that the people thinking seriously about what to invest their resources in would inspect that a bit closer. Christian Ruhl: It does feel yucky. I personally feel that a lot, working on these issues. It feels abhorrent to think about using nuclear weapons, but at the same time, when we think about it rationally, this is something that we do need to think about. But good reasons that people might say, let’s not pursue these kinds of interventions: one is that for various reasons, decreasing the consequences of a nuclear war, you might think maybe that actually makes nuclear war more likely — so it makes it seem winnable, it makes it seem less abhorrent, maybe undermines deterrence in one way or another. I lump this all into one kind of category: decreasing consequences increases probabilities. To respond to that, we can go back to the car analogy again. Let’s imagine we’re in this town where for some reason they haven’t had cars. And cars are new, and it’s been a few weeks and we haven’t had any car crashes. So we’ve come up with traffic lights, rules of the road, driver’s licences, and so on — but we don’t have seatbelts, we don’t have airbags, we don’t have ambulances. And maybe we want to suggest, “Hey, we should think about these bags that could inflate and somehow prevent the crash from being as bad. I don’t really know what it looks like. Maybe we should spend some money on a research project to see what it looked like.” Then people might say, “Actually, what makes people safe drivers is that they’re worried about crashing their cars.” A couple responses to that: like, maybe. But there are also drunk drivers, and we know from history that nuclear decision makers sometimes get drunk. There are malfunctions, there are accidents, all kinds of things can go wrong. If we’re thinking about mitigating nuclear winter, and we’re worried about might this make nuclear war more probable, my simple answer to that is that decision makers generally don’t even think about nuclear winter in the first place. That means that if we’re able to decrease the probability of nuclear winter, that might not even enter into the decision-making calculus. If you look at the historical examples of how decisions get made, what do they think about? They’re not like, “Will 2 billion people die or will 1 billion people die?” They say, “This is horrifying. Many people could die.” We know that humans are scope insensitive from various experiments. And then they think about things like the upcoming election: “I wonder if I look weak if I don’t do this,” right? This sort of rational calculation doesn’t happen. And if you think about who might die from a nuclear winter, I think horrifyingly, it’s already the marginalised people, people in the Global South living in extreme poverty. And do we think that enters the decision-making calculus of leaders in rich countries? Actually, here’s a thing we do have some evidence for. We do have some evidence about the behaviour of leaders and political elite in rich countries — and it’s generally that they don’t do anything about the global poor, starving, and dying. Miserable deaths happen every single day, and except for a small handful of people, we let this tragedy happen right under our eyes. So the idea that that would change during nuclear war just doesn’t make sense to me. Luisa Rodriguez: That’s just not a key consideration for them. And on the one hand, that is incredibly sad and dark and depressing. On the other hand, it means that this kind of risk is actually less worrying. Yeah, I think I basically at least buy that for some of these. I can imagine the argument being a little stronger for others. So maybe there’s an interesting angle, which is that we prioritise especially highly interventions that seem especially unlikely to enter into the calculus of these decision makers. Christian Ruhl: Yeah. And what’s happening here is they’re making empirical claims about the psychology of decision making, often without empirical evidence. And I’ll make an offer here: I’d be willing to take a bet with anybody that if you ran an experimental war game about this, decision makers would be fairly insensitive to the numbers. If somebody’s worried about 200 million Americans dying or 100 million Americans dying, I think in the brain of most humans, that’s just like “the end of everything that I care about,” and it’s horrifying, and hard to even think about those numbers. Toby Ord on whether civilisation could rebuild from a small surviving population Rob Wiblin: Imagining a world where there’s a massive nuclear war and most of the surviving people are in New Zealand or Chile or Argentina or whatever: it seems like most people think that it’s very likely that those people would then be able rebuild all of the civilisation that we have now, and then potentially eventually more later on. That basically there’s a very good chance it would recover. Are there any reasons for thinking that we just might not recover? I know sometimes people mention, for example, that we would have used up all the fossil fuels, so that it’d be harder for them to build up again. Toby Ord: I think that this is a very uncertain issue. I think you’ll find that people have very different views on this probability of recovery. Some people think that it’s 99.9% chance we’ll recover, and really some people think that it’s like a 90% chance we won’t recover. And it’s very wide disagreement about a topic that we have very little actual information on. So I’ve tried to be fairly cautious about that, since I’m not an expert on it — to treat it as a live possibility, but something that I’m not putting all my weight on. So when people talk about civilisation collapse: it’s a fascinating area, I think, for people in existential risk. A lot of the literature on it is just about individual civilisations, such as a particular dynasty of Egyptian civilisation collapsing where they treat, say, Egypt as several civilisations as opposed to treating say Western Europe as a civilisation or something like that. It’s very fine-grained, typically. But what we’re talking about is, at the very least, a global collapse of civilisation. And then sometimes that is thought about as things moving to a pre-industrial time. But I’m talking about it in the book that the level that I think is most salient, that I’m using, is that there is no civilisation. So it’s in the same way that 15,000 years ago, there was no civilisation — and what would be needed to get such a severe level of collapse? It could be that I’m analysing it at the wrong level, and that the dangers are more from having smaller levels of collapse is sufficiently more likely, that even though we’re more likely to recover, there’s still a greater chance of screwing things up. It’s very unclear, but for my purposes, I’m thinking of it as there’s no civilisation anymore. And it’s very hard to see how nuclear war could get you to that level. Also there’s a kind of reassuring aspect that civilisation has independently developed more than five times in different parts of the world from pre-civilised world. Therefore it’s very hard to think that there’s only a 10% chance that civilisation would come back or something — since why did it come so many times so far? Rob Wiblin: Interestingly, we’ve only had one industrial revolution, though. So what if we go back to 1500? And I guess some people think that the Industrial Revolution was very contingent on particular aspects of technology and politics. Not everyone. Toby Ord: Yeah. So that’s a different approach. I think at Open Phil, their writings on this have been more concerned about that level: what would be needed to knock things back to a pre-industrial world and then the chance of getting industry back. I think that’s interesting and there should be more discussion of what’s the relevant level that does most of the work. And that one, people who think that the values that came out of this change as well, that they’re something that’s very fragile and if you got industrial civilisation back they’d have different values. That’s another potential argument, not one that I draw upon, but that some people make. The way I think about this is that it’s not that the collapsed civilisation would then go on forever, to the end of time, as an alternative to extinction. It’s rather that it would be more likely to end its state of non-civilised world by moving to extinction than by moving upwards to a civilised world or some kind of fulfilling of humanity’s potential, such that the event that knocked out civilisation would be the key event in human history. That’s the property that I think is needed in order to make that the existential catastrophe. Rob Wiblin: You have this really nice point in the book about how people talked about civilisational collapse in the past. So the Roman Empire collapsed in some sense: it became more politically fragmented and disorganised. But that doesn’t mean that many of the people there died or that the city ceased to exist. And even the eastern part of it kind of continued on apace and it didn’t affect China or the Americas. It actually got quite a localised political breakdown. Toby Ord: That’s right. And so from my view, I failed to see why almost any of these numbers on civilisational lifetimes and things like that are relevant at all. I think that there is some case someone could make that they’re relevant. For example, they could say that, “We’re so interconnected now that the level of interconnectedness of the world is similar to the interconnectedness of one of these previous civilisations.” I don’t think it’s quite true, but if they could make that case, then maybe that would kind of suggest why it’s the relevant unit of inquiry. But a useful example I think is this case with the Black Death that killed something like a third of the people in Europe and it didn’t cause any kind of regional collapse of civilisation. Things just kind of moved through. And so it seems like this is the kind of case where we do seem to be quite robust. Luisa Rodriguez on how fast populations might rebound Rob Wiblin: Taking now the assumption that we would recover, how fast could we expect the population to rebound if things go well? Luisa Rodriguez: So if you think the population would grow at the fastest level that it ever has — which is in the 1960s, that was about 2.2% per year — then you’d get about a tenfold increase in population every 100 years. So if you lost 90% of the population, you’d be back to current levels within 100 years. And then if you think that population is going to grow slower, so maybe the level it did when humans were just agriculturalists, then you could recover a population from 90% population loss to current levels in about 240 years. Which is still really, really fast. Rob Wiblin: Yeah. It’s kind of shockingly fast. It seems like it should take ages, but yeah, I suppose just the magic of exponential growth means that it doesn’t take quite as long as you’d think. Luisa Rodriguez: Exactly. Yeah. So even the most pessimistic scenarios we thought of were if 99.999% of people died, which is a huge number of people dying, at that 1% population growth — so agriculturalist-level of population growth — you’d still expect the population to reach current levels in about 1,200 years. Rob Wiblin: I suppose we could compound all these scenarios into an even more pessimistic one where you have 99.9% gone, and you’re only growing at the hunter-gatherer level, which was 0.1%, we would get back to the original population in 7,000 years. But I guess maintaining a 0.1% population growth rate the entire time just seems very strange. It seems like it should either be below zero or more above zero than that. That’s just like knife-edge level. Luisa Rodriguez: It would be really surprising if at 6 billion, we were still going at hunter-gatherer levels of growth rates. David Denkenberger on the odds civilisation recovers even without much preparation Rob Wiblin: It seems like even in an all-out nuclear war, where we didn’t have very good food alternatives, there’s a good chance that 10%, 20%, 30% of people would survive, or maybe even more than that, why wouldn’t we recover? David Denkenberger: I personally am fairly optimistic that we would recover. There’s been some talk about why we might not recover. For instance, if we don’t have easily accessible fossil fuels anymore. I think there are still routes of recovering, but it would be more challenging. Another one is potentially something that wiped out the grass family because so many civilisations have been built on the grass family, corn or wheat. The exception is potatoes, like in South America. Then you could say, well, maybe the trauma of the catastrophe would select for people who are nastier or not trusting. So maybe we wouldn’t develop that trust of people who are not related to us. I don’t know how high probability that is, but I think it’s at least possible. Maybe the disaster somehow permanently affects our IQ or some other characteristic that made development of civilisation not possible. I think Eliezer Yudkowsky said something like we are the least intelligent that it is possible to develop industrial civilisation — as in we just were able to cross the threshold. Because it’s interesting that it took us something like 100,000 years from anatomically modern humans to develop from where we are now. If we had just had the knowledge we have now, we could’ve grown at 3% population growth per year, and it would only take a few centuries, but it took us 100,000 years. Like I say, I’m optimistic that we would recover, but it’s just I think not a sure thing. Athena Aktipis on the best ways to prepare for a catastrophe, and keeping it fun Rob Wiblin: What kind of advice do you have about how people should think about the apocalypse in order to be happier while they’re doing it, and maybe think about it in a more productive way? Athena Aktipis: I think one of the easiest things that we can do is just think about risk management in a fun way. There are a lot of things that we can do at the household level to manage risk better. Then we can also build communities that do a better job of managing risk, and then you can take that up to the highest scales as well. Things like making sure that you’re ready for an emergency. There’s this idea of all-hazards preparation, which is basically just emergency preparation, but you can make it a little bit more fun by being like, “Hey, are you ready for the zombie apocalypse? Do you have 72 hours’ worth of stuff so you could just shelter in place?” Some basic emergency preparedness actually helps to make our overall systems more robust too, because if there is a natural disaster and all households are able to shelter in place for 72 hours, then whatever institutional-level support is available to deal with a problem can be focused on the acute problem — as opposed to having to split effort between the problem and the humanitarian crisis that can emerge in those situations. Then the other one is building these networks where we can get help in times of need, ask for help. Those often spontaneously emerge in disasters, but having some networks already set up ahead of time before things get really bad is an easy way to manage risk by doing what people call “limited risk pooling.” Where it’s like, if you’re in need and I have enough to help, then I’ll just help you without expecting to get paid back. Rob Wiblin: That’s interesting. I haven’t really heard that idea before. What sort of relationships might you want to build ahead of time? I suppose here, we’re thinking of disasters like, an extreme one is nuclear war possibly, if you manage to survive it. Also earthquakes would be another classic, or massive wildfires in some places. Athena Aktipis: Yeah. Flooding. Massive power outages in the winter or the summer. There are a lot. Rob Wiblin: Yeah, I suppose now we have to worry about cyberattacks that could shut down the electricity grid. Something kind of unprecedented. Athena Aktipis: Absolutely. Yeah. This then opens up this whole other area of my work, which is really about looking at cooperation in human societies. We have this project called the Human Generosity Project, where we’ve looked at almost a dozen small-scale societies around the world now and how people within those societies help each other in times of need. We have found that if you look at the risk management strategies that people are using, we see need-based transfers. It’s like, “Hey, if I’m in need, I will only ask for help if I’m genuinely in need.” And then if you receive a request, you will help if you’re able to without going below what you need. The one society where they don’t use it as much is this society I was telling you about in Mongolia, where they have these winters that are just horrible. They have to help each other ahead of time. They have to manage the risk proactively, because they can’t go to each other’s houses when there’s six feet of snow outside. Rob Wiblin: Oh, wow. So what do they do? Athena Aktipis: They will help each other build shelters and make sure that everybody has the resources that they need. It’s basically sheltering in place for the winter. They’re helping each other be able to do that for their families and their livestock. Rob Wiblin: I would emigrate. That sounds awful. Athena Aktipis: Yeah. But the bottom line here is that people around the world in small-scale societies do this, and also in rural societies here in the US. We’ve studied ranchers in the Southwest, here in Arizona and New Mexico near the border with Mexico. They have a system they call neighbouring, which is largely a need-based transfer system where they help each other in times of need and they don’t expect to get paid back for those things that arise unpredictably. These are things that already exist, and they’re really good at handling the kinds of things that typically we want insurance for — those things that we can’t predict and can’t control. And if you take it to the extreme, there’s certain things that market-based insurance actually cannot insure against. Rob Wiblin: Because everyone gets hit at once? Athena Aktipis: That’s one possibility. Everyone gets hit at once. But another is just that it hasn’t happened yet and there’s no way to really calculate what the probability is of the event or how severe it would be. In the absence of any of that information, you can’t calculate what an insurance premium could be. It’s an actuarial problem: you need the data in order to price the insurance. But with these need-based transfer networks, you can at least have them in place for any kinds of needs that arise unpredictably. Now, whether the system will be able to effectively handle those is another question, but you at least are able to set up systems that can deal with things that have never happened before — that we can’t even understand what the risks are like. Rob Wiblin: I live in a city, in London. What should I try to coordinate with my neighbours? I suppose a big issue is people don’t have very large places, so it’s difficult to stockpile enough water or food or anything like that. I guess in a war scenario, London’s a very juicy target, so things could get pretty grim. Athena Aktipis: Yeah. I think no matter where you live, having 72 hours of supplies is wise. And you can actually do this in a way that makes sense for having a very busy lifestyle, which is something that I love. Rob Wiblin: Apocalypse prep — on the go! Athena Aktipis: Exactly! Apocalypse-casual lifestyle. How do you do this well? For example, if you like couscous, couscous is a great food to have around. Not only can you prepare it very quickly on a weeknight if you need something to eat, but it stores well and you actually don’t even need hot water to prepare it. You can just add water to it and let it sit for a half an hour, and then you could eat it. So if couscous is something that you’re fine with, then you can just make sure to buy enough couscous at the store that you could at least have it be part of what you would be eating for those 72 hours. Whenever you buy a new one, you put it at the back and you just have some extras. And then it also makes it easier when you’re super busy and you don’t have time to go to the store. You’re like, “Oh, I’ve got my 72 hours of prep.” Obviously, next time you go to the store you want to re-up. But sort of thinking about being prepared, not as like, “Oh, I have to go and figure out how to buy really long shelf life food on Amazon. Who’s a reliable source for this?” No, you just look at the kinds of things that you like to eat that are shelf stable, and just have more of those on hand. Because that will make your day-to-day life easier, and also will put you in a better position if something totally unexpected happens and you have to shelter in place. Rob Wiblin: The UK has a pretty precarious food situation actually, because it relies on a constant stream of imports. It doesn’t produce anywhere near enough food for the population that it has. I would suggest having enough food for weeks, conceivably months, would be not outrageous, if you were able to do that here. I guess there are storage issues. I did at one point store a whole bunch of rice, but I didn’t store it well enough and mice got into it, and that was very embarrassing. I was a very amateurish prepper. But yes, now we have some rice and pasta in a thick Tupperware. The mice can’t get to it. Athena Aktipis: Excellent. I think that another thing is just not feeling intimidated by, “I have to do all of these things.” No, just start with having enough water around and having some extra dry food that is stuff that you eat anyway, and then you can work from there. It’s not like you have to do all of the things all at once or anticipate every possibility, because you just can’t. Having conversations with people about what they’re doing: Do they have their preps? It can be fun. If you get into a little social competition about it, it could be playful fun. I have this idea for a new kind of dinner party. I haven’t tried it yet, but I absolutely want to: you roll the dice to figure out whose house you’re going to go to, and then you show up at that house, and you have to figure out how to make a really nice dinner with just the shelf-stable prep food that’s there. Then you practice making fun meals and surviving in your mini-apocalypse dinner party. Stuff like that. I think we could make it fun. Then it just kind of puts our attention on how maybe we should just be ready for the unexpected, so that at least we have some more time to plan. That’s the thing. You might not be able to have enough food around to actually manage the risk of something catastrophic that would happen, but you can have enough food around so that you have a few days to figure out what your next steps are if something really catastrophic happens. Rob Wiblin: Yeah. Makes a lot of sense. When you start talking about stockpiling food to protect yourself against disasters, I think some people’s eyes kind of roll and think you’re a bit crazy, and start associating you with preppers who get maybe a little bit too into it. But earlier this year, my partner and I spent a bunch of time thinking about what would we do if there’s a nuclear war and we survive? What would that look like? It was very long odds, but maybe it was getting probable enough that it was worth having a conversation about it. It is really fun on some level. Athena Aktipis: Exactly! That’s the thing. Rob Wiblin: Maybe not everyone enjoys this, so it’s easier for them to have to stay in for survivalists. But there is something just very entertaining. Our lives can be a little bit boring on a day-to-day level. Imagining what would we do if just everything was destroyed? How would we cope? Athena Aktipis: Yeah. People have called it “survival porn.” There’s something just appealing about imagining yourself out there or inside or trying to survive in challenging odds. It’s like a little story that we tell ourselves that makes us have fun and feel good. Rob Wiblin: I saw on your website that you were involved in some zombie apocalypse medicine preparation thing. I don’t really understand the nature of this — what was it? And what was something useful you learned about surviving the zombie apocalypse specifically? Athena Aktipis: Well, I started and I’m the executive producer for this whole group of really fun and interesting scholars. We’re called Zombified Media. We have the Zombified podcast that I cohost with my friend and colleague, Dave Lundberg-Kenrick. We have this livestream channel called Channel Zed, where we have all sorts of shows that are based around the idea that the zombie apocalypse is going on now: Here’s how you get your pantry set up. Here’s how you set up your go bag. Here’s how you deal with the zombies outside in a way that makes sense given cooperation theory. We kind of used the zombie apocalypse there as a fun way to engage people. And it all actually started with the Zombie Apocalypse Medicine Meeting. We just had the third one of those. We do those every two years. It’s basically an academic conference framed around this idea that the zombie apocalypse is going on and we have to try to understand zombie behaviour. We have to look at what are the threats that we’re facing in the world now? What are the threats we’re facing in the future? Rob Wiblin: What is an unusual tip that you’ve learned for how to survive a zombie apocalypse? I guess there’s a wide range of different possible apocalypses of the zombie kind, so maybe you need to be more specific about the exact scenario you find yourself in. But yeah, any advice for listeners? Athena Aktipis: Absolutely. I’ve got one big tip, and it applies to all apocalypses, which is to build your Z team. This is who are the people who you would want by your side in the zombie apocalypse, or really in any hazard. That doesn’t necessarily mean that somebody would have to be literally by your side — but somebody who you would want to be sharing information with, that you would want to be there to back each other up if something went really wrong. That’s basically risk transfer. That’s the limited risk pooling. Start setting up those relationships. You probably have a bunch of people in your life already, you just don’t even think about them in those terms. But have a conversation with them about, “Hey, if the zombie apocalypse happened, what would we do?” — or if a nuclear event happened, if you want to be more serious — so you can start to have those conversations, and just think about how you can proactively manage your risk. And cultivating that Z team is a super fun way to do it. Hopefully your Z team members, they’re like, “Hey, do you have your 72 hours?” Rob Wiblin: Bare minimum to get in the team. Athena Aktipis: “If not, let me help you.” Right? I think we can approach it from this perspective of how do we bring more people into this idea of making managing risk fun? Also, we should absolutely not neglect the neighbourhoods and communities that we live in. Because in the event of an emergency, it’s likely that there’ll be some interdependence with the people who live near us. Then we can also think about scaling that up. There’s these sister cities programmes that arose for cultural exchange and educational exchange, but now when there are disasters in cities that are sister cities, oftentimes, there’s just this spontaneous outpouring of help. They’re kind of Z team members to each other. We can kind of generalise this Z team idea, this risk transfer idea, to a lot of different levels: try to grow those systems that increase our resilience at the individual level, household level, neighbourhood, community, national, international levels. “Risk management for the win!” is how I like to look at it. Will MacAskill on the virtues of the potato Rob Wiblin: So listeners are dying to know: what is it with you and potatoes? Will MacAskill: I just think they’re neat. Rob Wiblin: This might be a little bit in jest. When I say “listeners are dying to know,” I mean two listeners. Will MacAskill: Yeah, this did come up at the last EA Global London. So yeah, during May, June 2020 — kind of the peak of lockdown madness — I was working hard on the book, and I was often going down kind of rabbit holes that were often very useful. But one I went down was the history of the potato, and the potato’s long-term impact. And an early draft of the book had a lot of potato-related content. People didn’t really like it. Rob Wiblin: It’s mostly just a book about potatoes at this point. Will MacAskill: I do now own many books about potatoes. And it all stemmed from how the potato was one of the most important transformative technologies of all time. Rob Wiblin: Tell me more. Will MacAskill: Well, when it was first imported to Europe from South America, it was actually regarded with quite a lot of scepticism, because people thought that it would give you leprosy because the skin looked like the skin of a leper. But it was also just a radically new sort of vegetable. So it took a while to take off, but then those areas that were suitable for potato-based agriculture started using it. There’s one study that suggests that they had radically more urbanisation and population growth. You could get three times as many calories per acre from the potato as you could from wheat, or… Rob Wiblin: It’s just so fast growing, or just so efficient at converting sunlight into calories? Will MacAskill: Yeah, basically. It’s also quite close to a superfood. There’s this paper by Nathan Nunn and Nancy Qian in The Quarterly Journal of Economics, it’s got like thousands of citations, and it’s like, “Section II: The virtues of the potato.” It’s like comparing potatoes to turnips. At least, if you’re an agricultural worker requiring more calories per day, you can live on nothing but buttery mashed potatoes. Rob Wiblin: In reasonable health? Will MacAskill: In reasonable health, yeah. You get all the relevant nutrients apart from vitamins D and K, which you can get from milk. You also need an occasional supplement of lentils or oats for the molybdenum. But basically, to a first approximation, buttery mashed potatoes can just be your life. So it seemed to be actually just very good nutritionally as well. Rob Wiblin: Huh. Will MacAskill: The joke I started making was these analogies between the potato and AI. Because it was this discontinuous, technological advancement that was in some ways more “general” than previous vegetables, but it was also recursively self-improving. Rob Wiblin: Because we kept selecting the best potatoes? Will MacAskill: Well, no. Because by growing potatoes, it actually was also very good feed for livestock, which produced manure, which allowed you to make many more potatoes. Were you thinking that’s a bit of a stretch? Rob Wiblin: I was thinking, how many cycles of improvement do you get out of that? Will MacAskill: You do plateau. But I think that is also relevant. Rob Wiblin: Right. Could happen with AI. Will MacAskill: Exactly. Getting a bit of self-improvement doesn’t mean you go forever necessarily. Rob Wiblin: Yeah, totally. Will MacAskill: You can plateau. And it is relevant for thinking about automation more generally, or productivity improvements more generally. In agriculture in general, we’ve had these enormous productivity gains and automation, and actually that’s meant agriculture’s become a much smaller part of the economy rather than the bigger one — because it’s the stuff that’s hard to automate, but essential, that ends up becoming kind of the bottleneck and swells to become the whole economy. And that could well happen with AI as well — this economic model and how this might go and how that could be like a bottleneck. Kind of singularity-esque both. But people thought it was a little flippant. Rob Wiblin: Didn’t make it into the final cut. I think there’s almost no mentions of potatoes in the book. Will MacAskill: Sadly, it’s all gone. But I have been approached by people asking for perhaps a standalone article. Maybe it will still see the light of day. I have commissioned research onto this, because the other part of the issue is the core persistence study. If you take the persistence study and then just extrapolate it out, then if the potato had never existed or hadn’t been imported to Europe, a billion fewer people would be alive today. Rob Wiblin: Wow. It kind of makes it surprising that the civilisations in the Americas weren’t more powerful or weren’t more populous. I mean, if it’s so much more efficient at producing calories, you’d think that’d be a huge advantage in terms of getting economies to scale and being at the forefront of technological development. Will MacAskill: Yeah. That’s actually kind of maybe a good argument for some of the scepticism about the studies. So I have asked Jaime to look into it, because I’m confident the effect size will not be as large as it is stated in the paper. Rob Wiblin: Yeah. [Update three months later:] Will MacAskill: Well, I should say, in the last podcast, I made a comment about this persistence study about how potatoes had this enormous long-run impact. And I commented, “Oh, it probably doesn’t check out.” Turns out it checks out! So my interest in potatoes was vindicated all along. It’s the most important technology ever invented. Among the most important. Rob Wiblin: Nice. We’ll go and issue a correction to that episode. Will MacAskill: Please do. Luisa Rodriguez: Hey listeners, I hope you enjoyed that compilation! There are many more gems in the full episodes for you to check out if you’re craving even more lessons about civilisational collapse and resilience — there’s links to all of those in the blog post of this episode. You may also want to read about how you can use your career to work on these issues — on the 80,000 Hours website we have an article that explains the case for working on reducing existential risks, as well as problem profiles on great power conflict, catastrophic pandemics, nuclear security, extreme climate change, and AI-related catastrophes. Those too are linked in the blog post for this episode. All right, thanks to the production team for putting that compilation together. We’ll be back with a fully new interview very soon!