From Nuclear War to AI: Why We Keep Predicting the Apocalypse | Ben Landau-Taylor

Vance Crowe [00:00:00] You know, I talk to

Ben Landau-Taylor [00:00:00] People a lot and they're like, well, it's going in this direction, and we can see that it's going very fast. And like the thing about speed is it sort of presupposes, you know, how far away the destination is. Like you can ask is a hundred miles an hour fast. It's like, well if I'm going from San Francisco to Los Angeles, it's pretty fast. If I'm going from San Francisco to Mars, it's actually quite slow.

Vance Crowe [00:00:21] Welcome back to the podcast. I'm glad you're here today. Ben Landau-Taylor Returns longtime. Listeners probably remember me talking to Ben about how institutional decay was a real threat to our continuing civilization. Today Ben is back with an entirely new article that was published in Palladium Magazine called The Lifecycle of an Apocalypse. If you are somebody that has always pushed back against the narratives about climate change, about the way that acid rain is coming, about nuclear war, you are in for a real treat because Ben has a view on this and how it is shaping our view on AI. We're gonna jump right into this interview and we'll be back later to talk more. Ben Landau-Taylor, welcome back to the podcast. Yeah,

Ben Landau-Taylor [00:01:08] It's good to be here. It's been a long time.

Vance Crowe [00:01:10] I could not believe it. I was looking up in the archives before we started this podcast and it was 2021, but I remember our podcast vividly because you said some pretty shocking things about the way the world would play out. And it turns out, I think almost all of your predictions were right. We didn't have violence in the streets. The Epstein situation was a, I think you called it a dinosaur skull where there's the rest of the body, but we're probably not gonna find out about it. You were right about that and on and on. So I'm very excited to talk with you about your predictions of AI and what's going on with the apocalyptic thinking around this subject. Yes. Well, hopefully my predictions

Ben Landau-Taylor [00:01:49] That we are not on the verge of an apocalypse prove out to go equally well.

Vance Crowe [00:01:55] So why don't I let you lay out your article. You put one out in the life, you put one out in Palladium Magazine called The Lifecycle of an Apocalypse, and when I first started to read it, I thought you were going to be

Ben Landau-Taylor [00:02:06] A doist,

Vance Crowe [00:02:07] But you are not. Do you wanna lay out the general thesis of the article? Yeah, so I've been,

Ben Landau-Taylor [00:02:12] You know,

Vance Crowe [00:02:13] Hanging out

Ben Landau-Taylor [00:02:13] With a bunch of the AI doomers for a very long time. I started getting involved in that scene in like 20 12, 20 13. A lot of them are very good friends of mine and I've learned, you know, I, I respect them in a great deal in many ways. But when it comes to the

Vance Crowe [00:02:29] Specific

Ben Landau-Taylor [00:02:30] Idea we we have of the, you know, computer apocalypse that's supposed to be coming up, I don't, I've never really been persuaded by it. And the reason is because I've studied a bunch of the other apocalypses that have been predicted and were supposed to get us all. And a bunch of people have noticed that in when it comes to the sort of big pop culture, sort of the societal gestalt that the computer apocalypse is sort of coming in right as the environment apocalypse has been losing steam and sort of in the popular mind, you know, five years ago, 10 years ago, people were very viscerally worried. I think it's easy to forget how, you know, very immediately worried a lot of people were that the world was going to be literally destroyed by climate change. And that's been kind of like fading away and was, and the computer apocalypse has been sort of breaking through into the popular discourse around the same time. And as I was reading about the history of a bunch of these, there were a couple of interesting things that I noticed. This is not the first time that it's happened that we've had, you know, these different apocalypses that as the last one fades out, the next one comes in and it's, and it seems like there's been this sort of psychological or sociological sort of need for there to be some story about how technology and industry is about to destroy the human species. And this has been going on for a very long time, since longer than either of us has been alive.

Ben Landau-Taylor [00:04:06] And you know, it keeps changing. They keep not arriving. And I was looking into the history of this and trying to find where did these start? Because if you go back to the like 18 hundreds, you don't really see anybody getting worried that, oh no, we're all about to be destroyed by our own creations. And I was looking and the first, the first of these that I was able to find where you see, you know, not just owe some abstract prediction of what's going to happen when we run out of coal, but people getting really viscerally worried of we are about to destroy civilization. This starts happening for the first time right after the first World War, which kind of makes sense because that was this incredibly horrific destructive conflict that left parts of Europe basically post-apocalyptic was incredibly traumatic for everyone involved, for very un understandable reasons. And so then you, you have a lot of the intellectuals start being like, man, if we do another one of these with better technology, then we're just going to destroy civilization. We're going to destroy the human race. Or if it's not the next one, then it's going to be the one after that or the one after that. Because you know, the basic idea that people had was as we're go, as we get more and more powerful weapons, sooner or later we're going to get something that's powerful enough. You can read the sci-fi at the time. Olaf Stapleton has a story where they end up just killing the entire population of Europe with an advanced poison gas. And then later on there's some, you know, more advanced power, power technology that explodes and turns the entire world into a volcanic wasteland.

Ben Landau-Taylor [00:05:47] You know, that's the science fiction. But the thing that people were more concretely worried about, a lot of people thought it was going to be the bombing of cities by airplanes. This was a new thing that in, you know, in like the late twenties, early thirties, they started being like, oh, our planes are good enough now to drop bombs on cities. Some people were very worried about this. The British parliament had a discussion where a bunch of the different parliament members think that if this happens, that's just the end of civilization. That's, you know, the term they use. A bunch of them say this will end civilization if we have another war with the bombing of cities. And one of them, Clement Atley, who after the war would later on become the Prime Minister, is talking about how we need international control of civil aviation as the only way to prevent the destruction of civilization by airplanes. Which is, you know, it's a common theme you see in a lot of these. We need the international control of the carbon emissions or the airplanes or the nuclear bombs or the AI or that's always sort of the only way to prevent the apocalypse. And then it turns out these guys were not totally wrong. Like the bombing of cities was a very big deal. World War II broke out, it had a lot of these super weapons, you know, it had the atomic bomb, it had the fire bombing of cities. It did not destroy civilization. It like was very bad for a lot of people, quite deadly. But civilization as a whole bounced back quite fast. And you know, Tokyo and Hiroshima and a lot of the cities that were hit worst with the bombing, it took about 10 years before they were larger than they had been before the war.

Ben Landau-Taylor [00:07:24] They grew because they recovered so quickly in a way that a lot of the sort of airplane doomers didn't really expect at the time. And

Vance Crowe [00:07:35] So between the World War, so could I, could I just restate what you've said so far?

Ben Landau-Taylor [00:07:38] So please,

Vance Crowe [00:07:39] Pre-World War I, people weren't living in this like, oh, the world's about to end. It was really a lot more about the excitement of the future. World War I comes along, we see the death and destruction that can come along with industrial scale war. And so now people flip into an apocalyptic mindset where now they can picture things going horribly wrong. And instead of technology being something that takes us a step forward, there is some trade-off step that really kind of pervades the culture throughout starting after World War I. Is that correct?

Ben Landau-Taylor [00:08:14] Yes, that's right. And the, and the thing that changed in people's minds was before the war, people had the idea of, oh, we're getting a better understanding of everything in the world. We, you know, we're, as we're getting more rational and more reasonable, this is giving us power over the physical world, but we're also understanding society better. And so we're going to get better government, better social organization because the same stuff that lets us figure out better machines, that's going to work on society also. And then the, the, the World War makes people think, man, maybe all of our better technology is, you know, just coexisting with the same sorts of, you know, we're just the same sort of people we've always been and the promise of we're going to figure out how to have better societies and how to not have war, maybe that's not really going to pan out. And then between the World Wars p you, like, basically every intellectual is just running around and being like, so how do we figure out a way of reordering society to not keep having these bigger and bigger and more destructive wars? Then they have the second world war and people just give up. They're like, okay, it's just the nature of man to destroy himself. And what's going on is that most people are just baseline convinced first that the nature of man is to destroy himself with industry and technology. And then second they'll reach around for whatever specific, the nuclear bomb, the climate change, the overpopulation, the under population, you know, they're, they're sort of reaching for what is the most plausible thing to fill into this preexisting con conclusion that it is man's nature to destroy himself with technology.

Vance Crowe [00:10:00] I learned here how to communicate with intent and with interest, very lecture light and practice heavy to collaboration. The intimacy with the cohorts, the caring nature that they shared across the group. I don't think that would've come across in a larger group or in a classroom setting. You know, just relax and, and talk was I think very effective. And I have some vision, some clarity for how I can take what I learned going forward and change the trajectory of my life. Interest-Based Communication, learn more at Articulate Ventures slash IBC. You know, it strikes me because I used to share an office with a Russian woman and she said, oh Americans, you know, all of your stories end with a happy ending and all of our stories end more realistically where everyone dies. But she was like, but the funny thing about Americans is you all think that we're about to, to go over the edge and that everything's gonna end. And I think that she's kind of right, but I'd never heard anybody explain why I thought maybe that was just a part of the American ethos. And maybe what you're saying it is, it was because what we became after World War II started amplifying this technology and then people had this kind of, oh well the trade off is existential doom that's coming on the horizon. Yeah. And I think

Ben Landau-Taylor [00:11:22] It's not just America. This seems to

Vance Crowe [00:11:23] Be

Ben Landau-Taylor [00:11:23] Basically throughout the western world, you don't really seem to see the same ideas in like China. They don't seem to have reacted this way to the world wars. Their experience of the world wars was very different. Their intellectual environment was very different. But you know, if you go to like France or Italy, I think it's the same.

Vance Crowe [00:11:41] So how do you think this is playing out in society? You, you, we move, we went all the way through World War ii. What happens after World War II and people give up on those on, on stopping war. Yeah.

Ben Landau-Taylor [00:11:53] So

Vance Crowe [00:11:54] After that

Ben Landau-Taylor [00:11:55] You see there's a bunch of these different, you know, things get filled in for what's going to be the specific technology that destroys us. And they all go through this sort of cycle of an intellectual comes up with some argument and then it gets sort of popularized and filed down. And a couple of the, the argument always has a couple of missing steps. Obviously none of them have been true because we haven't been destroyed yet. And the missing steps follow an interesting pattern. But the, but it gets popularized and then it has about like 10 or 20 years of having lots of currency in the popular mind before people kind of, it's not that it gets disproved, people just kind of lose patience with it and they're like, well I haven't been destroyed by that one. And then another one sort of arrives to fill in for the missing demand. And they've generally gone through two sort of big categories where there's been a bunch of different reasons why nuclear bombs was supposed to be the thing that was going to get us. And there's been a bunch of reasons why it was supposed to be environmental collapse that was supposed to get us. And each of these has had a bunch of different variations where with the nuclear bombs right after the war, it was going to be just that, you know, the, the bombs explode and kill everybody or they destroy enough of the infrastructure that you can't keep having civilization and thi everything just sort of falls apart. And then pretty quickly, you know, people who are, you know, thinking con more concretely about this realize that like, no, it's not actually going to make humanity go extinct.

Ben Landau-Taylor [00:13:31] Like if we had a big nuclear war, you know, I'm here in San Francisco so I personally would be vaporized, but you know, that's not going to make humanity go extinct. That's not going to destroy our civilization. It feels a little bit offensive, but you can actually just have a really big war and kill a ton of people and then everyone else just sorts of picks up and moves on even though you might be dead. You know, that's funny that something that you're saying there is so simple

Vance Crowe [00:13:58] And yet is really countercultural. I mean people genuinely believe if you get into a world war, it's all over the, the whole game is done. And I I, that was definitely pounded into me and my youth when we were up against the Russians and you know, there was the big red scare. Is that just a result of the government propagandizing people in order to get more power or is

Ben Landau-Taylor [00:14:20] This

Vance Crowe [00:14:20] Something deeper in the psychology of humans?

Ben Landau-Taylor [00:14:23] I think it is something deeper and more spiritual just basically about how we never really recovered from losing our pre-World war I faith in progress. And then also it is a lot, a lot of it is does come from propaganda, but it's not really directly government propaganda. It's actually, so when the nuclear war won, it actually, a lot of it comes from anti-war propaganda where, so, you know, I, the there was in 1950 was sort of the second reason we were supposed to all die from nuclear war, which was the radiation poisoning the idea that you can make these nuclear bombs that may leave the area radioactive for a very long time, the cobalt bombs and that, you know, these are the ones that get, you know, briefly mentioned in Dr. Strangelove as that's how they're going to leave the world uninhabitable. And now the actual mechanics of this, you can't actually cover the whole earth in radioactivity with them. That's not how atmospheric movement works. So there is some fiction where like it ends up making literally the entire world radioactive. We've probably all seen stories like this of varying levels of scientific sophistication, but you can definitely make some cities very bad places where you would get cancer if you stayed there a long time. But you, or, or like in theory, no one's ever actually built one of these bombs. 'cause when you're building bombs, you want to kill people right away, not like many months in the future. But the, so the, the idea, this is another one where in principle

Vance Crowe [00:15:59] You could

Ben Landau-Taylor [00:15:59] Make a bomb that leaves an area radioactive for a while, but you can't actually irradiate the entire earth that way. So that one was popular for a while and then kind of by like the seventies people weren't really viscerally worried about it. It was just sort of a fiction trope. And then in the eighties was when you see the more propaganda wand of people trying to very deliberately convinced, like you were talking about the idea that if you have the world war, it's all over. And this was the idea of nuclear winter. And so there was a big campaign of a bunch of scientists, this was organized by Paul Ehrlich and Carl Sagan who got everyone together. They put together a bunch of computer models simulating how the, you know, their models of how the cities would burn and loft of the ash up into, way up into the atmosphere and then how long it would stay there. Computer models were very new and exciting at the time and Sagan and Ehrlich both just said many times on the record to reporters, the reason we are doing this is we have to convince the p you know, the generals and the politicians and the people making the decisions that they can't have a war because a war would be too terrible that you know, that you know, that it would wipe out humanity. So it's not just, oh it's a big war, but like we have to convince them that this is so horrible you can't actually do it. So this was, you know, a very deliberate, very, you know, very sophisticated campaign. There were some,

Vance Crowe [00:17:29] It seeded into culture, right? Like movies, like war games with Matthew Broderick, right? Where,

Ben Landau-Taylor [00:17:34] Where

Vance Crowe [00:17:34] The computers took over and they're launching nuclear weapons and the conclusion is the only way to win is to not play the game. Right?

Ben Landau-Taylor [00:17:41] And there are some earlier versions of this, like there's the bulletin of atomic scientists, which is quite soon after the war war gets put together by originally a bunch of the actual nuclear physicists. And then over time they leave and a bunch of just regular PR guys sort of end up colonizing the organization. But you know, they have their sort of clock of like it's, you know, 30 minutes to midnight, it's 20 minutes to midnight, it's 15 minutes to midnight. And what exactly is midnight? That's sort of unclear, but the idea is, you know, they're trying to make the nuclear war sort of too terrible to contemplate. So yeah, so this was a anti-war campaign was to try to get this idea in, in sort of into the water as, as a way of convincing people to not do a war was by sort of puffing up the idea of how bad it would be when it comes to the nuclear winter. Since then we have, you know, learned that it was not nearly as dire as their like computer simulations where, you know, you, you had to sort of make up and guess about a bunch of the different atmospheric parameters we've seen. There have been some oil fires. I think this was in the wake of the Iraq invasion, if I remember right, that some of the big oil wells caught on fire and oh, they set 'em on fire as they were running away, right? In Kuwait. Yeah, yeah, that's right. And Sagan was still alive at the time. He was pretty surprised when those ended up not lofting the ash way up into the upper atmosphere, the way that they were expecting in their simulations.

Ben Landau-Taylor [00:19:18] And there was some big wildfire, I think a Canadian wildfire where that one was big enough to get the sort of lofting the ash way into the atmosphere and they found that it dissipates something like twice as fast as was predicted in the nuclear winter calculations. So again, we are looking here at a physical mechanism which is basically plausible. Like it's true that ash can get into the atmosphere and in at least some quantity it could end up blocking sunlight. But the particular, you know, there were a bunch of assumptions that had to be filled in. 'cause you know, it, you, you can't just sort of reason figure out all those very complex atmospheric dynamics from first principles that you have to actually observe this stuff. And you know, it turns out the observations were do not line up with the things that they were, you know, as, as assuming when they were putting to putting together the, these predictions in order, you know, as part of this anti-war campaign. Which, you know, that's also what we saw when we were looking at, you know, the i the cobalt bombs. That's what we saw when they were looking at the, you know, the, the interwar guys were looking at the aerial bombing of cities. And so nuclear winter was very scary for a while, but again, it sort of fell out of the discourse. A little bit of this was because we got some more data in and the people who were studying it very closely started thinking, oh well it's probably not actually like they, you know, as bad as they were saying.

Ben Landau-Taylor [00:21:00] But mostly it's just that it's been a long time and there hasn't been a nuclear war or a nuclear winter and people just kind of, you know, got bored of it and moved on after noticing that they hadn't been destroyed.

Vance Crowe [00:21:12] Well it's hard to keep the world's anxiety up about this and to make it so if you are the

Ben Landau-Taylor [00:21:18] Person

Vance Crowe [00:21:18] Talking about it, that that brings a spotlight to you. If the people aren't worried about it, then the spotlight's not there. So probably that means that spokesmen or people that would normally leverage that go away.

Ben Landau-Taylor [00:21:30] Yeah, and you know, like the, the people who were organizing that are retired or dead by now and you know, the, we've got other activists working on different campaigns and then the other big strains sort of in parallel with the nuclear one was of course the environmentalist one where this one sort of started really picking up in the late sixties mostly, where basically the idea was that the normal operation of industry was going to end up destroying civilization by again. We see a bunch of sort of different mechanisms of the same underlying, if you're just running all the factories, that kills everybody. So the first one was the overpopulation apocalypse, which you got in the population bomb Paul LIC

Vance Crowe [00:22:21] Again.

Ben Landau-Taylor [00:22:22] Yes. Paul Ehrlich shows up again after the, after the, so this was, he actually did this one before the nuclear winter campaign. I think this one was about 1960 or so, 60 where he was predicting basically that the exponential population growth was going to outstrip the ability of the world to produce food. And so you would start seeing massive, massive famines in the 1970s. And this was like the big justification of lots of the population control policies that were put in place in like China and India and also just got a lot of individual people to be very neurotic about having kids and have much smaller families. Turned out to be totally wrong. Food production has actually, so our ability to get better yields in food is such, has gone up at about the same rate as the population. So it's not just that we are producing so much more food as we have more people, it's that we are feeding the world using

Vance Crowe [00:23:25] About

Ben Landau-Taylor [00:23:26] The same number of acres as we were when Erlich wrote this book in the first place. 'cause our technology has gotten so much better and the technology getting much, much, much better is basically the reason why all, why all of the environmentalist apocalypse scenarios didn't pan out. 'cause the other one that, that was a little bit later than Icks book was the limits to growth, which was sort of the canonical argument of a couple of the other mechanisms. There was the resource depletion idea, which is that we're going to run out, you know, society is using exponentially larger amounts of resources like, you know, oil or cobalt or rare earth minerals or whatever. And so for all of these they were like, okay, so we don't know exactly how much there is in the world. Geologists have their estimates, let's be conservative and say that the real amount is five times more than the geologists have found. And they do their projections and show that like we should be peaking and like running out of all this stuff in like the 2010s or something. Oh

Vance Crowe [00:24:32] Yeah, I remember peak oil that was, I mean, on the news

Ben Landau-Taylor [00:24:34] Every

Vance Crowe [00:24:35] Night,

Ben Landau-Taylor [00:24:36] Right? And

Vance Crowe [00:24:37] The reason that

Ben Landau-Taylor [00:24:37] These ones end up being totally wrong is, well one, we just haven't surveyed most of the earth for almost any resources. Oil and gold are about the only ones where we've been reasonably thorough. But the, but but even with those, our technology just tur keeps getting better at extractions so that, you know, when they were doing these projections, they were looking, they would look at something like, you know, shale oil, they would just see, you know, from the perspective of the 1970s that's just junk. And then later on we get the technology that we can, you know, we figure out how to do fracking and all of that stuff that had been just junk. The geologists knew it was there but you couldn't get anything out of it. Now that's where we're getting a huge fraction of our oil and you know, we, we figure out how to like transport and use methane or natural gas as they call it, as a fuel. And so that's something like a third of fossil fuel usage these days and they just, you know, the, the technology to actually be moving, moving that and using it as a distant fuel was just barely beginning to be worked out when that book was being written. So all of these projections of we're about, we're supposed to run out of all that stuff. No, like for all of that, we've just had a very easy time exponentially growing that. And you know, it, it turned out to be kind of like the caveman worrying that they're just going to run out of Flint. Like you know, there is a finite amount of flint on the world but like

Vance Crowe [00:26:08] The world

Ben Landau-Taylor [00:26:08] Is really big and you are not close to hitting it.

Vance Crowe [00:26:10] Here's an idea from

Ben Landau-Taylor [00:26:11] A

Vance Crowe [00:26:11] Book I'm writing about communications and negotiation skills for highly logical people. Everything

Ben Landau-Taylor [00:26:18] You

Vance Crowe [00:26:18] Need to know about how your presentation is going is written on the faces of your audience.

Ben Landau-Taylor [00:26:24] I once

Vance Crowe [00:26:24] Spoke to a room of Pennsylvania dairy farmers, a mix of modern farmers and plain folk, Amish and Mennonite. When I looked out I could

Ben Landau-Taylor [00:26:32] See

Vance Crowe [00:26:32] That everybody was locked in and really engaged in the speech. So I started to frame up my argument for hundreds of thousands of years humans were hunters and gatherers and only about 10,000 years ago we started farming the modern farmers. They loved this framing, but the plain folk started looking away. The warm eye contact had turned cold and I didn't know why, but I knew something wasn't working and I was able to detect that from the audience just by their eye contact. So I sped up the slides to the next concept, caught their eye contact again and brought them back afterwards. A young Amish man came up to me and said, I liked what you had to say but I totally disagree with you about how long human beings were hunters and gatherers who've not been around on the earth this long. Only then did I realize that his coldness made sense. I had spoken against something that was against their beliefs and caught it in time only because I was trying to look the audience in the eyes. I call it, look at the dragon when you present you should look at one person in the eyes at a time, finish a single thought and then move to the next person. Their faces tell you whether they understand, whether they're engaged, whether they agree. And then you'll know, should I be speeding up? Should I slow down? Should I clarify things or change course altogether? Dragons guard the treasure, your audience's eyes are scary. Things that you have to face in order to get what you want. Look at the dragon as one of the social heuristics I teach about to become a better presenter in my forthcoming book about communications and negotiation skills for highly logical people sign up early to receive excerpts and provide some feedback at Articulate Ventures slash book.

Vance Crowe [00:28:23] I also remember around this time is people talking about things like acid rain and the ozone layer hole. Yes, this, this becomes a big deal. Yes, the

Ben Landau-Taylor [00:28:32] Pollution is the other thing that they start really worrying about. And this is another one that they start doing some of their sort of exponential projections of in the limits to growth of, you know, we're releasing more and more just toxic sludge into the environment that's causing all of these problems. It was causing all of these problems. You had horrible smog in the cities, the rivers were becoming poisonous. And then this one, again, we've just invented the technology to be able to clean all that stuff up, which didn't really exist in the sixties and seventies, but we figured out, oh, if you put these sorts of filters on your factories and you make sure that when you're burning the trash you burn it outta really high temperatures that you get rid of all of the dioxins and you know, it was

Vance Crowe [00:29:17] Reasonably

Ben Landau-Taylor [00:29:18] Expensive to, you know, put all that stuff throughout the entire industrial stack and you know, we had to stop using chlorofluorocarbons 'cause those were messing up the ozones, which to different chemicals for refrigeration. But

Vance Crowe [00:29:31] Basically it

Ben Landau-Taylor [00:29:32] Turned out that, you know, retooling the industrial stack to just not be emitting this stuff was very achievable. And you know, we've seen, you know, places like China sort of speed run this in the last 20 years where they build up their industry and when they're building it up, they don't really care. They're just like, we're going, you know, we're going to make our cities filled with smog, that's fine, that's just the cost of industrialization. And then later they get rich and they're like, all right, well now that we are rich, one of the things we want to buy with being wealthy is clean cities where you can go outside without having to like wear a gas mask or worry about getting asthma. And you know, we had, you know, like, so basically we

Vance Crowe [00:30:16] Figured out the

Ben Landau-Taylor [00:30:16] Technology to clean up all this stuff. We don't have the horrible smog in la, the rivers are nice again. We figured out how to build landfills that actually treat the toxic sludge rather than just leaking it all into the groundwater. So that was another one where we just got the technological solution and we're able to clean up all of this stuff.

Vance Crowe [00:30:37] Now do you feel like this was a function of, because people got all stressed out about it, we came up with these answers or would we have come up with these answers even if people hadn't been throwing up the alarm? I think that there was,

Ben Landau-Taylor [00:30:52] It,

Vance Crowe [00:30:52] It depends case by case.

Ben Landau-Taylor [00:30:54] There's some variation,

Vance Crowe [00:30:55] But I, I think

Ben Landau-Taylor [00:30:58] A lot of them were just sort of regular quality of life things that, you know, the reason we got rid of the smog was, was not because people were worried that it was going to destroy the world. It was that just going outside in the smog is just really unpleasant viscerally. And, and so like that wasn't, oh here's this future thing that could happen. It's like, I live in Los Angeles, I walk outside, it's really bad and if the gov and I want the government to institute the new standards so that, you know, we're, we just don't have to deal with this problem we're asked. The ozone layer was much more of the, we have projected ahead to see this thing. It's not really that bad now, but we can see how that's going to be bad if nothing is done. So we're going to have to switch away. And that was much more of the, the alarmism ended up actually being the causal mechanism by which the problem got solved. So it's different when you're looking at different types of pollution basically.

Vance Crowe [00:31:53] So chalk went up for the ozone layer that that panic did work out. I think people now use that as evidence for why we should be stressed out about things like climate change. So some people be like, well what happened to the hole with the ozone? And they'll be like, well it's 'cause we made a big deal out of it. People changed, we should do the same thing with climate change. Is climate change in the same apocalypto doom cycle? Yeah. So you, you

Ben Landau-Taylor [00:32:17] Saw the sort of global warming thing take off after like sort of after the pollution just got solved and people stopped sort of really viscerally worrying about, you know, you still had that Captain Planet stuff when I was a kid. But it was, you know, kind of old and on its way out because, and it's not that people ever explicitly sort of went to the public record and were like, good news, we've, we solved the pollution problem. It turns out that we can just deal with that. It just kind of faded out and a lot of the sort of general vibes of it sort of got repurposed to the global warming idea, even if a lot of the arguments don't really carry over because the mechanisms are fairly different for the sort of much more local problems you get with a lot of these toxic emissions versus the much more global way that, you know, in, in increasing carbon in the atmosphere changes the overall climate. So with that one, again, you have a real mechanism of like, in fact the, you know, the carbon emissions, you know, do cause the earth to be warmer and like I can just notice that my hometown just doesn't have the horrible winter days that it would have a few times a year when I was a kid. It's like, yep, definitely warmer. And then you also have this leap to therefore civilization's about to be destroyed, which is never what what you would see when you would actually read the sort of most rigorous clima climatologists. Like, they're like, okay, things are going to get moderately hotter, that's going to be pretty bad if you are in the really hot places or the parts that are like really low lying on the seaside, it's going to be pretty good if you're in like the Canadian Tundra or you're looking at commerce in the Arctic circle.

Ben Landau-Taylor [00:34:02] Like, but like, you know, the idea of, oh it's going to, you know, you had all of those projections of like by 2007 the city of New York is going to be totally underwater from rising sea levels types of stuff. And you've got a lot of people like using the phrase Cook the Planet and not meaning it as a metaphor. They were imagining that it would literally be too hot for there to be any human life and this stuff was totally scientifically groundless. There was no, you know, there's a real mechanism, but there is this assumption that it would go all the way to the extreme of destroying civilization. You know, partly through this game of telephone of like activists trying to blow it up so that people would, you know, you know, make the treaties that they would want. And partly from people making movies like Water World and Wally and that stuff, just because it makes for a good compelling story, if you can tell Bad Max in the rising sea levels have got rid of all of the land story. Yeah. And that loop

Vance Crowe [00:35:04] Just keeps building because not only do people see it in their culture, in their movies, in their television shows, then there are corporations in order to respond to consumer demand, they start taking on these sorts of belief systems and what can we do about it? So now people are participating in it in their work. Now they're tied, their careers are tied somehow to Right and meeting these mandates, things like that. Yeah. And

Ben Landau-Taylor [00:35:27] It's partly because they're catering to consumer demand and it's partly because a lot of the people in those companies just they're seeing the same stuff everyone else does and they

Vance Crowe [00:35:35] Believe it.

Ben Landau-Taylor [00:35:38] So, you know, so that one again, you know, it had, there was this real physical mechanism and then also there were these untethered predictions of the, imagine the physical mechanism to its maximum conceivable extent which destroys civilization. And then just over a couple of decades people seem to have, you know, about five, four or five years ago in America, Europe seems to be a little bit behind. People just kind of got bored with it. They were like, well we haven't been destroyed yet. There was no sort of public

Vance Crowe [00:36:12] Intellectual

Ben Landau-Taylor [00:36:13] Sort of explicit refutation of it, but people have just kind of moved on.

Vance Crowe [00:36:21] Yeah, I mean, gosh, it's been a long time since I've seen people promoting that. I think it was starting to go out during COVID right now all of a sudden people are worried about diseases and George Floyd and Racial justice and then Me Too. But those things never had the deep hold on them in culture in the way that climate change did. Certainly Right. Mean, you know. Yeah, yeah. I was just asking are those like inversely related then? I

Ben Landau-Taylor [00:36:49] Think

Vance Crowe [00:36:50] So. I think

Ben Landau-Taylor [00:36:51] That the Overlap with the all the 2020 stuff is mostly just coincidence. I think it was just kind of, this was about as long as people had for them before they just kind of lost patience with it. And it just happened to be around the time all of that was going on. And a bunch of the intellectuals, a bunch of my, you know, friends who are very into of this sorts of things are, are pretty worried about pandemics as a possible thing that could destroy humanity. And you can actually see this among the, especially intellectual ones going quite early, the Var Bush had a book Modern Arms and Free Man, which he wrote I think in the fifties, where he's worried about pandemics as sort of like, you know, the big catastrophic thing that we have to be prepared for. Because again, like if you think about the physical arguments, like yeah, you know, having big disastrous pandemics is certainly possible. Like, you know, we had one that was as pandemics go, not as bad as they could be, but still a pretty big deal. But it never really got its hold in the popular imagination. And I think the reason for it being that, you know, it's not man causing his own destruction, which seems to be the hole that we're trying to fill here. And what has filled that hole, of course is the, is the AI and the, you know, the idea of making the machine that is smarter than the man. Which again, this has been a sort of background idea in pop culture and science fiction for about a hundred years.

Ben Landau-Taylor [00:38:28] But, and it's also been something that like a lot of the more serious futurists have been worried about going back to something like the fifties or so, but it's only quite recently that it's sort of gotten beyond the futurists and the computer programmers themselves and into the pop culture.

Vance Crowe [00:38:47] Yeah, it's funny because with AI, doism, you're right, the 1950s, like Dune hated AI, the, you know, they, they basically were saying this, they had to defeat it and this is what caused their civilization to change in the way and Hal 9,000, right? The, that that was shown as AI could take over and, and kill humans and trying to pursue its goals, unendingly and the paperclip stuff. And so it's been going on. But back then, the idea that you would ever be able to talk to a computer was science fiction, almost fantasy that you would be able to ask it a question and it would reply back to you or that it would be able to take some action in the world. I mean, hell, we've only really had powerful computers since, you know, the late nineties. So to the idea that these things could ever e even even in the, even in 2020, it'd be very difficult to predict AI and say that it was an actual serious thing that we had to worry about. Right. And you know, there were

Ben Landau-Taylor [00:39:47] A,

Vance Crowe [00:39:47] You know, Yudkowsky and all

Ben Landau-Taylor [00:39:51] Those guys over then they were

Vance Crowe [00:39:52] Saying

Ben Landau-Taylor [00:39:52] That stuff in 2020, but like, you

Vance Crowe [00:39:56] Know, at the time

Ben Landau-Taylor [00:39:56] They were considered these sort of like, you know, kind of odd fringe guys, you know, they were beginning

Vance Crowe [00:40:03] To get some intellectual

Ben Landau-Taylor [00:40:05] Currency by 2020, but it really hadn't broken through to like, you know, my mom or anything. And I, and a lot of that was just because yeah,

Vance Crowe [00:40:15] Having

Ben Landau-Taylor [00:40:16] The computer that can talk like a man just makes it really visceral.

Vance Crowe [00:40:22] So you mentioned Yudkowsky. Tell the audience a little bit about him and kind of the influence that he's had on the fear people have about AI. You mentioned him in your, in your writeup in Palladium, right? Yes. And so he's

Ben Landau-Taylor [00:40:35] Been, you know, a philosopher of artificial intelligence who around like, like in, in the, like a little bit before 2010,

Vance Crowe [00:40:44] If I

Ben Landau-Taylor [00:40:45] Remember correctly, started getting very worried about, you know, the, all of this stuff about like how the artificial intelligence, you know, the recursively self-improving artificial intelligence. You get the computer that is good enough at thinking that it can think about how it works. And so it can write this sort of next generation of itself and sort of quickly s spiral into something incredibly smart, incredibly powerful and outstrip people's ability to control it. And then from here, you know, he

Vance Crowe [00:41:19] Is

Ben Landau-Taylor [00:41:19] Certainly not the first to, to think, oh, and then once you have it like smarter than humans than it can, then it's sort of naturally going to come into conflict and it's going to win because it's smarter. But he was the one who really put together this vision of like how it is going to improve itself to the point where it then does all that. And so his, and so what's notable about the about it though is that basically his, the, all of the current artificial intelligence companies with the partial exception of Google DeepMind are basically downstream of his ideas of he just convinced a whole bunch of people that this stuff is powerful and inevitable and coming quite soon. And then a few of those people were like, oh, so then I should try to build it. And that was the genesis of Open AI and anthropic, whoa, whoa, whoa, make that jump for me. So I read this in your, in your article and I, I remember Elon Musk saying like, oh, we don't wanna build, AI could get outta control. It could be bad, it could be bad. And then all of a sudden he is like, eh,

Vance Crowe [00:42:29] Now I'm building some,

Ben Landau-Taylor [00:42:31] Right. And so Elon Musk got onto this largely through the work of Max Tegmark is a mathematics and cosmology professor at MIT who was one of the people who was convinced by Yudkowsky's arguments. And so he put together this organization, the Future of Humanity Institute, basically to popularize those ideas. And Tegmark was very sharp about how to do this. So he was like, all right, so you know, people like Yudkowsky and Nick Bostrom and a bunch of these philosophers have put together some quite rigorous arguments and in order to get these to go from being seen as science fiction sort of right next to the UFO people into like actual respectable discourse, I need to get like big names to endorse it. And so he started around just any famous person you can get. So he got Morgan Freeman to endorse it, whatever, but the big one was then he got Stephen Hawking to say like, yes, this makes sense, we should be taking this seriously shortly before Hawking died. And that was sort of like breaking the floodgates. And so shortly after that, you know, was when Musk started doing his first public posting about like, oh, AI is summoning the demon and all that stuff. And so Tegmark's Future of Humanity Institute put together a couple of conferences in Puerto Rico where that included, you know, so Musk went to those, Sam Altman went to those, Eliezer went to those.

Ben Landau-Taylor [00:44:10] And it seems like, so the genesis of OpenAI, you know, had been like largely came out of people who had, who were like interacting over the, over these ideas of like, here's the argument about how it's going to be powerful and inevitable. And like most of the people who were persuaded of that do not run out and build it. Most of them are what we think of as sort of like, you know, they're generally called the Doomers these days where, you know, they're around this whole less wrong community or various other institutions and organizations trying, like being very worried about this stuff sometimes doing various forms of activism. But some people see something like this and they're like, okay, so if it's going to be powerful then I want it. And some of them are like, well, if it's powerful then whoever builds it is going to control whether it gets used for good or for bad, and I don't trust you with it. So I need to have my own. So the, and that was specifically the genesis of Anthropic, which began as a bunch of people who were working at Open AI and who basically were like, man, I really don't trust the people in charge of OpenAI to like use this power responsibly, but I do, trust me. So we're going to leave

Vance Crowe [00:45:29] And

Ben Landau-Taylor [00:45:31] Do our own, you know, AI company that we are in charge of so that we can do it responsibly and safely. And this also seems to have been, you know, after Musk got kind of, it's

Vance Crowe [00:45:45] A little

Ben Landau-Taylor [00:45:45] Complicated but forced out of OpenAI and then he's like, okay, well then I'm going to make my own AI company, you know, housed within my industrial empire because I don't trust Anthropic and I don't trust OpenAI. So there's this sort of, if it's going to be, you know, this infinitely powerful thing that of course I need my own. And so by sort of, and, and, and so it ended up being that a lot of these initial labs, the connection isn't quite as strong with DeepMind, which isn't really, you know, which was the first of these before OpenAI, before the sort of chat bot model. They were making a lot of progress, you know, with a lot of their gameplay AIs where it seems like Demis Hassabis, their main founder doesn't seem to have been directly inspired by Yudkowsky, but one of his co-founders was and was getting into arguments with these guys on LessWrong.com very early on. And a lot of their funders were like, just Orthodox, you know, PERSU people persuaded by Yudkowsky's arguments about AI and you know, and so they were, and so those people ended up being like, oh, so this is going to be powerful, so I should put money into DeepMind. This was a way back in like back 20 13, 14 or so.

Vance Crowe [00:47:04] Oh yeah. But DeepMind that was like, I remember my buddy showing me a picture of DeepMind trying to recreate the scream and you could barely make it out. It looked like one of those pictures that you used to have to squint your eyes to make it 3D. So to imagine that that somehow was going to turn into a threat, those people were very forward thinking and at least knowing what was going to be possible. What was their argument though? What were they thinking was going to get out of control that would make this into an apocalyptic threat?

Ben Landau-Taylor [00:47:34] Yeah, so the basic idea is that thinking isn't magic. It's a physical process which like any other physical process can be understood. And, and then when you understand it, you can do engineering with it. And, and, and, and so there basically there's been a few different approaches to how are we going to actually do that. And the actual course of it has not really vindicated that as much as it might seem like, because so much of it has turned out to just be throwing larger and larger and larger amounts of compute at it with, you know, using basically the same fundamental techniques. And so it basically ended up giving them this, just this confidence that, so there have actually been several waves of, you know, the people working on AI software being like, oh wow, we are like getting quite close to building the machine that thinks like a man and can then surpass humanity. And this has led to a few different sort of like big waves of optimism and investment and people doing really good work and advancing the state of the art of software. And, and then the previous ones have then ended in what is called AI winter, which basically means they realize, oh wait, we are, we've, we've, we've got a lot of cool advances that can do a lot of very new and useful things that are very powerful, but it is not going to become the sort of like full machine man and then, and then sort of the field as a whole sort of loses confidence and a lot of the money dries up and then a decade later they get an, they get another wave of enthusiasm and there's another big spurt of progress in artificial intelligence.

Ben Landau-Taylor [00:49:15] So this is, and like what's happened each time is they get this sort of, you know, the same mechanism as the people going from, oh, like I can imagine how like increasing the temperature of the earth is a physical process. I can, if you do take that to its maximum conceivable extent, you get this thing, they sort of, they imagine the process of better understanding machine cognition to its maximum feasible extent. They, this gives them a huge amount of optimism about the power of the thing which causes them to go out and build a bunch of really cool stuff and that it turns out to, you know, not be the sort of super intelligent mind that can, you know, do anything and outwit anybody. But nevertheless, they've built a bunch of cool things and software has moved forward a bunch. And so it seems like basically what this gave them was this sort of deep spiritual conviction that the work they were doing was going to give them infinite power. And that coupled with like a pretty solid understanding of the state of the art led them to do a bunch of really good work and make a bunch of progress on making the software better. So it's the sort like basically the confidence that the software was, you know, this sort of spiritual conviction that the software was close to sort of totally transforming the nature of the world and being the source of infinite power pushed them to make all of this quite cool and useful technology.

Ben Landau-Taylor [00:50:50] And then a bunch of them are also like, okay, and then this means that we are there for, you know, in 2027 is when we're going to hit the sort of, you know, superhuman god mind or whatever. But that stuff still feels like a, a big leap to me. But it is that conviction in, in the power of the thing that is, that they're doing that has like led them to take on bigger, bigger and much more scientifically fundamental challenges than a lot of other people and recruit a lot of the best talent with the pitch of like, just do the most fundamental stuff and build the infinitely powerful God mind and also to raise money from people who share that conviction. So when you look

Vance Crowe [00:51:34] At AI, I'm, I'm reminded, I think it's an XKCD cartoon

Ben Landau-Taylor [00:51:40] Where

Vance Crowe [00:51:40] There's like the AI has the intelligence of a chicken and then it gets up to the intelligence of a human. And then it's just slightly more and the at on this exponential growth curve that it's this superhuman thing. I just saw that that popular graphic show up yesterday and somebody saying like, we are at the AI is as intelligent as a human and the next step is just

Ben Landau-Taylor [00:52:02] Getting

Vance Crowe [00:52:03] A little bit smarter than a human and then it flies off without us. That is not a persuasive case

Ben Landau-Taylor [00:52:08] To you. So I mean, it looks

Vance Crowe [00:52:11] To me like

Ben Landau-Taylor [00:52:11] Machine intelligence is much weirder than that. Like there's a bunch of places in which it's much smarter than me and there's a bunch of places in which it's much dumber than me. And you know, that's been true, you know, since the day I was first interacting with a computer, there were many areas in which it's much smarter than I've, you know, even when I was like logging on with them, with dos, there was a bunch of types of, you know, just calculation and you know, very, and like speed and all sorts of things where it would, and, and, and also like if I, when it comes to like being able to unload a dishwasher or like any sort of embodied task, they are much very far behind. And when it comes, like there's a lot of times when I'll be, you know, it'll, it'll get confused on lots of things that will not trip up a human. So it looks, so it looks to me like the

Vance Crowe [00:53:03] Idea

Ben Landau-Taylor [00:53:03] Of putting it all on one axis doesn't really make a ton of sense. It's certainly the case that there's a bunch of stuff. You know, in 1998 there was a bunch of stuff that, you know, we had the, we had replaced the, you know, the computers are better at doing this than the people are. So we're going to have the computers do those instead of people. And that has expanded and expanded so far. But the idea of, you know, oh, we can imagine because it's ge you know, doing some of these things and some of these things we can sort of take this process to its maximum cons, conceivable extent, and therefore it is better at the, at the humans at all of the things.

Vance Crowe [00:53:39] Like

Ben Landau-Taylor [00:53:41] The idea

Vance Crowe [00:53:41] That this

Ben Landau-Taylor [00:53:41] Is happening soon, just doesn't, it, it seems like it's largely an article of faith. It doesn't, like

Vance Crowe [00:53:48] I have spent,

Ben Landau-Taylor [00:53:50] I don't even know how many hours talking to a lot of these guys about where exactly does this conviction, not that the thing is philosophically possible. I think the idea that it's philosophically possible is, you know, totally, you know, it's slightly less

Vance Crowe [00:54:03] Low. The same is, hey, we could have a bunch of nuclear bombs dropped on 80% of the cities and things could

Ben Landau-Taylor [00:54:08] Go really,

Vance Crowe [00:54:08] Really bad.

Ben Landau-Taylor [00:54:09] But

Vance Crowe [00:54:09] It's unlikely is what you're saying, the these two things are, it is possible, but the probability of it is not strong.

Ben Landau-Taylor [00:54:16] Yeah. And that, and that they, and that there is this very strong

Vance Crowe [00:54:20] Psychological

Ben Landau-Taylor [00:54:21] Propensity to conclude that because we are seeing progress in a direction and we can imagine the sort of maximal conceivable extent of it to therefore conclude from that that it's happening soon. So like people,

Vance Crowe [00:54:35] You know, I talk to people a lot and they're like, well it's

Ben Landau-Taylor [00:54:36] Going in this direction and we can see that it's going very fast. And like

Vance Crowe [00:54:40] The thing about

Ben Landau-Taylor [00:54:41] Speed is it sort of presupposes, you know, how far away the destination is. Like you can ask is a hundred miles an hour fast. It's like, well if I'm going from San Francisco to Los Angeles, it's pretty fast. If I'm going from San Francisco to Mars, it's actually quite slow. And there's this idea

Vance Crowe [00:54:57] That people imagine

Ben Landau-Taylor [00:54:58] That they know how far away this sort of like smarter than human at everybody, you know, the next stage on that, wait, but why graph that you're talking about. But there's just, when I have not been able to find anyone who can articulate a, any actual reason other than I just kind of feel a vibe for why they think that that thing is close rather than, you know, like Leonardo da Vinci was like sketching machines that would fly by something kind of like helicopters 400 years before we actually built a helicopter. He could sort of see where it was going. Roughly. The aerodynamics don't exactly make sense, but he was sort of like in the right ballpark, but that doesn't mean you're close. And

Vance Crowe [00:55:43] I

Ben Landau-Taylor [00:55:43] Have been really trying to get them to give me a solid argument for why they think it's close and they have not been able to do it. So it really just seems like it's this back filling in. Within

Vance Crowe [00:55:53] Your argument or I guess kinda your perspective, does

Ben Landau-Taylor [00:55:57] This include humans using

Vance Crowe [00:55:59] AI at, at some scale that is able to destroy civilization? You know, there

Ben Landau-Taylor [00:56:05] Just

Vance Crowe [00:56:05] Earlier this week there was the French refrigerators that were hacked into using AI and shut down and all that food was wasted. I mean, you can imagine a madman using thousands or millions of agents being able to enact his will on a scale that could never be possible before the AI. Yeah. And

Ben Landau-Taylor [00:56:24] This and this stuff like this is definitely going to be a big deal. And this looks to me like basically people between the world wars looking at aerial bombing of cities where like in fact aerial

Vance Crowe [00:56:34] Bombing of

Ben Landau-Taylor [00:56:35] Cities is a big deal. And then they're like, and therefore it's going to literally destroy civilization. And I really just don't see that.

Vance Crowe [00:56:43] Yeah, that's a, that's a fair counterpoint. When

Ben Landau-Taylor [00:56:46] You think

Vance Crowe [00:56:47] About this though, you know, you, you described LA and Mars being two different destinations and how fast you're going and how you get there. Can you imagine a situation in which the destination is the AI takes over and it's just that it's so far away it doesn't seem to be one to deal with? Or is it you're saying, ah, it's implausible to me that there is a scenario where the AI wipes us all out. I

Ben Landau-Taylor [00:57:12] Mean it's certainly, it. I can certainly imagine it. It's certainly permitted by physics and you know, when it comes to what technology is going to look like 400 years from now, 2000 years from now, like who knows? You can see all sorts of crazy stuff. I mean we're certainly going to see a lot of crazy stuff. It's going to, like, the one thing we can be certain of is that there's going to be a lot of absolutely crazy things that no one today is even imagining. And I'm not saying that these things are ruled, are like totally ruled out. It is certainly conceivable that, you know, in 80 years someone is going to actually build the cobalt bombs and then just build, you know, so many of them and distribute them by like quadcopter drone rather than by ICBM in a way that actually could cover the entire earth. It's like when you have these things that are permitted by physics, they certainly can happen at some point that's just very different from the claims that you see people being like, oh wow, like this is imminent and so we need to make these specific changes to counter this very specific, you know, thing that's definitely happening, you know, within the next five years or whatever. It's just a,

Vance Crowe [00:58:22] I mean,

Ben Landau-Taylor [00:58:23] To

Vance Crowe [00:58:23] Be honest, I'm,

Ben Landau-Taylor [00:58:24] I'm totally in your

Vance Crowe [00:58:25] Camp. I, I, I agree with this and I think that one of the things that comes along with if this is the problem and these are the solutions, what you end up saying is we need to take these freedoms away from people and we need to grab it for ourselves so that we can centralize it in some authority. There's almost never a time when somebody has a cure for the apocalyptic danger that doesn't involve taking freedoms away from

Ben Landau-Taylor [00:58:48] People. Would

Vance Crowe [00:58:49] You agree with that? Yeah,

Ben Landau-Taylor [00:58:50] I mean you with, with every one of the cases that I've mentioned, the solution has always been let's form an international body to regulate this stuff. And you know, it seems like the chlorofluorocarbons and the ozone layer is the only time when that seems to have actually been a good and helpful idea. But most of the time it seems like it really doesn't solve the problems.

Vance Crowe [00:59:13] No.

Ben Landau-Taylor [00:59:13] As a person who used to work at the World Bank, I

Vance Crowe [00:59:15] Can tell you the more you centralize international power, what you're really doing is giving the elites who already have the power that are going to make those institutions and they're going to staff them. You are giving them increasing amounts of power and oversight over

Ben Landau-Taylor [00:59:29] Something

Vance Crowe [00:59:29] That could really democratize things. I mean

Ben Landau-Taylor [00:59:32] The,

Vance Crowe [00:59:33] The potential of AI to lift people out of poverty of, out of the khap economy for example, is huge unless it gets regulated and they say, oh, we're only gonna give this to you in, in dribs and drabs, whereas the people at the top are gonna get to use the full thing. Is that also a danger? You're, you are on the lookout for

Ben Landau-Taylor [00:59:53] Oh yes. I think it

Vance Crowe [00:59:54] Is.

Ben Landau-Taylor [00:59:54] Like I,

Vance Crowe [00:59:56] I I think it

Ben Landau-Taylor [00:59:57] Is a much more likely scenario and there is a lot of, there is a lot of inertia pushing it in that direction. 'cause one of the reasons why there has been such a big, you know, skip from one of these apocalypse scenarios to the next is that they are, you know, the justification for a lot of these international regulatory sort of, that's how you keep your career going. If you look at the careers of a lot of the people who were, you know, were doing the professional environmentalist activism 10 years ago and that like, as the interest in that has dried up a, you know, a decent minority of them are doing the AI act activism now and like, because that's just how you keep these sorts of careers in this space. You know, you have to like whatever is the current, you know, justification for getting this sort of international regulatory track that you can keep your career going through. So that's a huge, there's a lot of inertia towards that and it's quite plausible we could end up there. It depends on a lot of things. So

Vance Crowe [01:01:02] Let's talk about

Ben Landau-Taylor [01:01:02] The doism

Vance Crowe [01:01:04] That's, that's going on because I think there are a lot of people like I, I know I've tweeted at Yosha Bach who's been trying to push back on a lot of this doism and I've said, man, the fear has really spread into rural America. If you go into the Midwest, for example, the farming communities where data centers are potentially gonna be built, you see people that have almost no interaction with AI at all, having very strong feelings about resource depletion, energy costs, domination over over farmland. It is very pervasive. There is definitely a doism going on inside of rural America that is much, much larger than their use of AI. Is this done on purpose or this is something that just has occurred? It's a natural phenomena. Yeah, so this is, so I think part of it there

Ben Landau-Taylor [01:02:00] Is natural and part of it is very engineered where the

Vance Crowe [01:02:03] Part of

Ben Landau-Taylor [01:02:03] It that is engineered is that there's these very well organized, very well funded campaigns by people who believe that the AI is going to destroy humanity. And they're like, well we have to, you know, make sure people understand that so that that they can stop it. So they've been throwing around like millions and millions and millions of dollars into all of these media campaigns and all of these activists and all of these, you know, bringing over the congressional staffers and, and all of this stuff. So there, so when it comes to the messaging, there is a very, there is a large professional group of people who is like full-time just trying to produce the stuff in order to cause that. And then on the other side there is, in most of these people, they just have this preexisting, you know, belief that the nature of the world, the sort of destiny of mankind is, is that we are going to destroy ourselves with technology. And like you cannot, if you were to just have a campaign around annual random nonsense, I don't think it would work. I think if you were to try to do something like this of like, you know, say you had the, you know, something about like natural pandemics or asteroid strikes, I just don't think it would work. But this happens to fit

Vance Crowe [01:03:27] With

Ben Landau-Taylor [01:03:27] This preexisting belief that most of these people have that, you know, it is our nature to do some type of apocalypse to ourselves. And you know, these, they, they had been concerned about the environmentalist stuff, but as that has kind of faded away, there's this like hole that is, there's sort of casting around for what's the most plausible thing with the best arguments and the best endorsements from the people who society recognizes as its intellectuals that I can fit in there and this is the current best candidate to fix that. So I think that, you know, like these other ones in another sort of 20 years or whatever it might be, what, you know, the AI is going to have been a very big deal and reshaped society in a lot of ways and not caused an apocalypse. And people are just going sort of drift away from the apocalypse fears like the other ones. But then something else is going to come into that and that's going to keep happening until we sort of resolve the spiritual trauma of the first world war.

Vance Crowe [01:04:30] Oh, that's fascinating. So your thesis is the AI doism is actually a, a consequence the trauma of World War I

Ben Landau-Taylor [01:04:42] Yeah, that all of these, this, yeah, all of these beliefs in the apocalypse is just this, it just comes from this general insight of our, you know, man is getting more powerful and at some level of power we don't really know specifically what we'll be able to destroy the world. And that has just, yeah. And that fundamental thing seems to have been behind all of these specific ones and is why the fact of any specific one failing to deliver the apocalypse hasn't really changed people's fundamental belief. 'cause they can still see that we are getting more powerful and that's going to keep being used in all of the ways for all of the things that humans have done since, you know, Kain figured out how to use a rock.

Vance Crowe [01:05:32] One of the things that I have noticed, I've given a couple talks on AI or on related technology at some universities, one at Iowa State and one at the University of Florida. And after both talks, the students in the class, you know, somewhere between 18 to 22, somebody would say, you are the first person I have talked to that thinks positively about AI, which means these kids that are in college trying to learn, trying to push forward knowledge what's going on, being up to date, are being inundated by their entire world. Their entire bubble is saying AI is a bad thing, don't use it, don't make it a part of your, you know, who

Ben Landau-Taylor [01:06:11] You are

Vance Crowe [01:06:11] With work and don't access it. Yeah. And

Ben Landau-Taylor [01:06:16] You

Vance Crowe [01:06:17] Know,

Ben Landau-Taylor [01:06:17] You,

Vance Crowe [01:06:17] You, you see this with,

Ben Landau-Taylor [01:06:18] You know, a lot of industry, like if you think about what, what, what

Vance Crowe [01:06:21] Are they hearing about like

Ben Landau-Taylor [01:06:22] Factories for example. It's basically the same stuff. So it's just this very pervasive, Jason Crawford writes a lot about this, how, you know, there's there, there's, there's all of this very foundational stuff, which is why we can have literally all of the things in our lives and people just get the propaganda from the people who want to dismantle it and they don't really look at, you know, the majesty of its regular operation and all of this stuff that it does for us. And I, I dunno, I find that very sad.

Vance Crowe [01:06:55] Well you, so you wrote this really lengthy, beautiful article in Palladium, you publish it. What was the response to your article?

Ben Landau-Taylor [01:07:05] Very little response. Like, there were a few individuals who like messaged me being like, I quite like it. And there's like some of my intellectual collaborators who, you know, I think were very glad that I had articulated it all in one place and cited sometimes. But when it comes to, you know, the doomers themselves, they have basically just not acknowledged that I wrote it.

Vance Crowe [01:07:28] Man, that is terrible. I, I read your article and I mean I found it to be very compelling and really something that kind of alleviated stress. In fact, there are a lot of boomers that I know that are trying to be up on tech that are a little nervous about it that I'm definitely gonna send this to because it puts that fear, that kind of existential dread that we all have about it into some kind of a context. And I myself am vulnerable to it, right? I see how quickly AI is changing, how much of an impact it's having on my day, how good it is. And I do have some pretty, pretty strong concerns about how much the world will change and how to know where that puck is going and to arrive there. But I have to say your article made me feel a lot better about that kind of latent dread that I feel.

Ben Landau-Taylor [01:08:18] Well I'm glad to hear that. That's a lot of what I was trying to do with it.

Vance Crowe [01:08:21] And so tell me, what do you think is on the horizon for AI? Will the doomers continue to get airspace? Will this turn into, we are required to register our AI agents and you have to have a government code for 'em before you're allowed to put 'em out into the world? What do you project is gonna happen? Yeah,

Ben Landau-Taylor [01:08:39] So there's going

Vance Crowe [01:08:39] To be

Ben Landau-Taylor [01:08:40] A lot of energy behind that. And so, you know, when it comes to this sort of political wrangling, I don't think I can give like strong, here's what's definitely going to happen happen. But my sort of pattern matchy guess of the sort of most boring story I can tell the thing that happens if there aren't any surprises, there's usually some surprises is basically that. So I think there's a lot of sincere, sincere doomers who basically are just like, man, the thing that we want is to make sure that nobody has this thing at all. Like in its sort of most powerful form and we want a lot of the other stuff you can get along the way and you know, we, we would love for this stuff to be like curing diseases and so forth. If you can do that without making a super cluster that like becomes even more intelligent than humans. And I think that those guys are just going to get outmaneuvered by the sort of like centralized, you know, regulatory blob is going to

Vance Crowe [01:09:47] Take a

Ben Landau-Taylor [01:09:47] Bunch of those guys, sort of boost them for a while in order to give a pretext for them to get a bunch more powers and like, you know, be able to regulate, you know, who's allowed to have the supercomputers and who isn't allowed to have them and make sure that there's tracking on every chip so that it, so that there's, you know, said, oh of course we can't have any like, you know, unauthorized supercomputers and if we can't track where all of these things are at all times, then how could we ever insure anything like that? And then it just gets used, you know? And then once they have that sort of in place, then all of this other, you know, like what we saw with the environmentalism stuff, it sort of got used as a pretext for, you know, this elite to just sort of like do its own rejiggering of the international economy and you know, oh, like yes, you know, everyone, we're making sure everyone's hitting their ESG goals. Is this like protecting the pandas? Not particularly, but that's not really important right now. And so I think it's going to end up being, the sort of most boring story I can tell is that this ends up being a pretext for further industrial con like further state control of the chip production and distribution and for increased surveillance at the level of individual chips.

Vance Crowe [01:11:06] Man, you are speaking my language here. I I am right there with you. I see this not just in AI, but you know, even in things like agriculture, you go out and you get a farmer that's excited and wants to do good things for the environment, the government is gonna be like, we love you. Why don't we give you a platform? Why don't we give you access? We're gonna

Ben Landau-Taylor [01:11:23] Let

Vance Crowe [01:11:23] You promote this idea. And then, oh by the way, when you get the public behind it, we are going to grab all of that energy and we're gonna suck it into the bureaucracy and you farmer that were trying to do it, that we were kind of guiding you to say that you were gonna be the one in charge. We're actually gonna push you aside and now we're gonna keep that control. And I see that same thing happening with AI people that are maybe like, to your point, genuinely worried about this are gonna be promoted by the people that want to get that fear up because the more that fear is up, the more there is to grab and pull into their own orbit of strength.

Ben Landau-Taylor [01:11:55] And we've already seen this happen, you know, this is, you know, this is how they ended up launching these companies that are doing exactly the thing that they say is what they don't

Vance Crowe [01:12:04] Want to happen. And

Ben Landau-Taylor [01:12:05] Like those guys, you know, they got a bunch of money from the Doomers, they got a bunch of their best researchers from the Doomers and they're like,

Vance Crowe [01:12:13] Thanks see ya.

Ben Landau-Taylor [01:12:14] And now we're just gonna run off and do our own thing and they leave, you know, all of the other guys and the machine intelligence research institute just sort of sputtering in the dust, be like, wait, how'd that happen? So

Vance Crowe [01:12:26] The other side of this, this

Ben Landau-Taylor [01:12:27] I think

Vance Crowe [01:12:28] That is, is totally crazy is once this technology is out and it is globally, there is, there's no stopping, right? As long as somebody can get electricity and, and the compute out there, they are going to be able to build their own models, they're gonna write better code. To me, we are in an arms race no matter what because whoever has access to synthetic intelligence and the, and the ability of agents to take action if, if their competitors don't have that, the, you won't just lap them, you, it'll be like, by the time you're in LA I'm already in Mars,

Ben Landau-Taylor [01:13:01] Right? And this does not mean that therefore everybody is going to use it, you know, and we, with the printing press, the Ottomans were like, man, we don't want that here. And so they, you know, like just basically banned. So you couldn't use it for, you know, anything in Turkish or Arabic if like the Greeks or the Jews wanted to have something in their own language with the printing press that was okay, but they just, you know, for hundreds of years they were just like, no, no printing press here. And that meant they fell behind everybody. So I think there are going to be, you know, probably a bunch of European countries are going to do that sort of thing with AI and then they are going to fall behind.

Vance Crowe [01:13:39] Well Ben, I have to say that it is tragic, an absolute tragedy that it took me five years to have you back on the podcast because this was again, another mind melting conversation. If people find what you're talking about compelling, they wanna see other things that you're putting out there, where would you direct them to go? Yeah, so my writing is on my

Ben Landau-Taylor [01:14:00] Website: benlandautaylor.com and I am also been working lately on the students podcast at thestudentsmedia.com.

Vance Crowe [01:14:11] Wonderful man. Well I will put links in that in the show. And what is the next article you're working on? Where are you sussing out patterns from history that you wanna share with the world? Yeah, so I was just,

Ben Landau-Taylor [01:14:24] I'm

Vance Crowe [01:14:24] Just getting back

Ben Landau-Taylor [01:14:24] From recently was in Brazil, covering the election there, looking at the mission party, one of the very exciting third parties who they didn't do very well in the election, but a lot of their

Vance Crowe [01:14:35] Political

Ben Landau-Taylor [01:14:36] Organization is very interesting and that's got me thinking about the strategy of these sort of outsider insurgent political parties in modern history. So I'm writing about that and should hopefully have that published in Palladium pretty soon. Well

Vance Crowe [01:14:53] Ben, thank you so much for coming on the show. We'll have you back on again soon. Great. Yeah, it's a

Ben Landau-Taylor [01:14:58] Pleasure to be here.

Vance Crowe [01:15:00] Thank you so much to Ben Landau-Taylor for joining us on the show. This is a wild ride and I have quite enjoyed it. Changing gears, if you are thinking about the future yourself, you may be wondering, what am I going to get? My loved ones for Christmas. If you've ever thought about doing something deep, something that will really have a lasting impact on your family, I would encourage you to consider a legacy interview. This is where I sit down with people's loved ones to record their life stories so that future generations can know their family history. Learn more. Go to LegacyInterviews.com. All right, we'll be back next week with another fascinating interview. I grew up with my dad and my uncle all the time and I heard all the old stories like my mother not having electricity and water. It's one generation removed today's world just so, so different. And

Ben Landau-Taylor [01:15:49] That gap is there. We always

Vance Crowe [01:15:50] Talked about it with my dad, but we

Ben Landau-Taylor [01:15:52] Never really did it. I would've

Vance Crowe [01:15:53] Liked to have had it of my parents. You know, we always wanted to do something like that, but somehow we never did. If you keep waiting so long, the memories do fade. I'm not saying I'm getting old. You take the opportunity when you can. Legacy Interviews, learn more at LegacyInterviews.com.