Professor Hugo de Garis - AI Will Kill 1 Billion People - Do We Create Our Exterminators?

Full transcript: AI Will Kill 1 Billion People on London Real

0:00 Brian Rose: This is London Real. I am Brian Rose. My guest today is Professor Hugo de Garis, the artificial intelligence expert and author. You are known for your work on the use of genetic algorithms to evolve artificial neural networks that you believe would enable the creation of artificial brains which would quickly surpass human levels of intelligence. You’ve published over 100 academic papers on the emergence of AI and robotics and the potential implications of these technologies on society.

Your groundbreaking book, The Artilect War, explores the potential consequences of a future conflict between humans and artificial intelligence that would result in billions of deaths before the end of the 21st century. You believe that there is a strong possibility that AI systems will simply eliminate the human race, something we will be powerless to stop. You have said that the coming decades will see species dominance become the major talking point on the planet, leading to assassinations of political leaders and scientists as a major war threatens the future of humanity. Professor, welcome back to London Real.

3:53 Hugo de Garis: Well, thank you, Brian. I’m very grateful to you for giving me a platform for this message: an Artilect War, the worst war that humanity’s ever had, for the simple reason of the stake. 20th century wars were largely between nation states, so what was at stake was the survival of a country. But if there’s a major war later in the 21st century, what’s at stake? Us, human beings, the human species. The stake has never been so high, so the passion level in the debate – should humanity build these godlike creatures or not – will never be so high.

And you’re talking 21st century weapons. So I use the concept giga, which just means a billion: gigadeath. Because if you ask the question, how many people died for political reasons in the 20th century? About 300 million – two world wars, ethnic cleansings, purges and so forth. Well, you only have to multiply that by a small number, three, to get it over a billion. So if we have this horrible war that I’m predicting is coming later this century, then it’s highly likely it’ll be over the billion mark. And that is utterly depressing. I’m glad I’m alive now.

And the tragic thing about it is, I’ve been thinking about this on and off for decades, literally, and I don’t have an answer. Last time you asked me how we can avoid this situation, and literally yesterday I was talking to the director of the Centre for the Future of Intelligence, the CFI, at Cambridge University. He asked me the same question, and I didn’t have an answer. I just don’t see an obvious way out.

6:00 As artificial intelligence keeps increasing – this is almost my logo, right? Human-level intelligence, machine-level intelligence – as that IQ gap, as I call it, seriously closes, then the species dominance debate heats up. In other words, should humanity build what I call artilects? That’s artificial intellects: godlike, massively intelligent machines. Now, for some people that would be like a religion, human beings serving as the stepping stone up the ladder of evolution to create these godlike creatures.

I see humanity in the next few decades splitting in at least two, maybe three, major ways, so I need to introduce a few labels. Those people in favour of building these godlike creatures I call Cosmists, based on the word cosmos, because that’s their perspective: there’s a whole universe out there. The other group, bitterly opposed because there’s so much at stake – the very survival of the human species – I call Terrans, as in terra firma, the Earth, because that’s their perspective, looking inward.

And there’s a third group. They want to add components to their own brains and become artilect gods themselves, transitioning away from being human.

Brian Rose: Like a cyborg.

Hugo de Garis: A cyborg, exactly – a cybernetic organism, part machine, part human. So they’re the three major groups that I see. As that IQ gap closes, the species dominance debate inevitably will heat up, and it’s doing that right now. It’s very topical now. But as that gap seriously starts closing, everyone will be wondering: are we, the human species, going to allow these machines to become smarter than us? Isn’t that dangerous? And if they become a lot smarter than us, isn’t that extremely dangerous?

So put yourself in the shoes of a so-called Terran politician, maybe even at a global, planetary level, and ask yourself what risk you would allow that these machines might, for whatever reason, decide to get rid of the human species, seeing it as a pest. What risk would be tolerable? One in a million? One in a billion? Given that so much is at stake, I think the answer for these Terran politicians would be zero, because there’s too much at stake. But on the other hand, you’ve got these Cosmist guys who are passionately pushing for the creation of these godlike creatures. There’s a whole universe out there.

Brian Rose: Former colleagues of yours, building these things.

9:25 Hugo de Garis: Yeah. And to be honest, I’m very ambivalent about this. At home I have these big 8K screens, and I’ll watch the universe – the deep field kind of photos, where each little pinprick of white is a whole galaxy – and I get this kind of cosmic consciousness. I think, oh my god, there’s a whole universe out there. This is the truly big picture. This is the destiny of the human species. We’ve got to get out there.

And then you start looking at these fabulous numbers. What’s the diameter of the observable universe? About 93 billion light years. How many galaxies in that huge volume? Two trillion. And our solar system is about 4.6 billion years old, which is exactly a third of the age of the universe, 13.8 billion. So most of the stars and galaxies out there are billions of years older than we are, and nearly all of these stars have planets. So it’s very likely that biological creatures who are intelligent and conscious have come into being, and maybe have moved on the way humanity is starting to do, moving away from biology towards the artilect. So maybe this transition has happened many, many times.

But I really believe this issue – should humanity build artilects – will absolutely dominate our global politics this century. In the 19th and 20th centuries, the philosophical question that dominated was who should own capital: the communist–capitalist dichotomy. This will be all about who builds the artilects and who doesn’t.

11:49 Brian Rose: Now, are you a Terran or a Cosmist? You spent decades at the highest levels of artificial intelligence. You worked with Ben Goertzel in China. You guys were in the documentary Singularity or Bust. You know all of these scientists. First of all, are the AI scientists predominantly Cosmists who want to build these godlike creatures? And are you as well, or do you change your mind every day?

Hugo de Garis: Depends what day it is. I was talking about the big screens and this cosmic consciousness. When I watch or read cosmology – because I’m a maths, physics, computing sort of guy – this cosmic vision is very strong. It’s a very powerful ideology: we, humanity, could build gods. So that’s sort of Mondays, Wednesdays and Fridays.

Other times I’ll be sitting in the park. I’m into my fourth quarter century of life; I’m 76 now. I’ll sit in the park watching these gorgeous little two-year-olds, and my grandpa heart goes thump, thump, thump. And then I’ll have this horrible thought: if there’s an Artilect War and it kills billions of people with 21st century weapons – the worst war humanity’s ever known, because the stake has never been so high – then all these gorgeous little ones… And then I’m Terran. So I’m torn.

And it’s not just me. I’ve been talking about this for ages, and often I’ll ask the audience to vote, and keep it simple: in your heart, are you more Cosmist – do you want to build these godlike creatures – or do you want them banned? Typically the answer would be 40/60, 60/40, 50/50. Then I had the thought, that’s probably just because the issue is so new people don’t really have an opinion about it yet, so they’re voting randomly. But after a while people would come up to me and say, “Professor de Garis, I feel so torn about this. On the one hand, the whole universe is magnificent. But on the other hand, the whole concept of gigadeath and an Artilect War is horrible.”

So I’m very ambivalent about this. In some ways I feel if humanity does not build these godlike creatures, that’s a cosmic tragedy, because we won’t go into space without building them.

Brian Rose: We’re stuck here.

Hugo de Garis: And from a Cosmist viewpoint, you’re looking at these pathetic little human beings, snuffed out in 80 years in a universe that’s billions of years old. There’s real passion behind the Cosmist viewpoint. And on the other hand, World War II was horrific – 50 to 100 million people killed. That was the most horrible tragedy of the 20th century. But that will be chicken feed in comparison to this really big one, if it comes to that.

And if you’re a Terran, you’re going to look upon a Cosmist as a monster, because what’s this Cosmist prepared to do? He’s prepared to risk that the whole human species gets wiped out because of his obsession with building these godlike creatures.

Brian Rose: Just because he’s obsessed with building a better mousetrap, he’s now willing to risk all of us.

Hugo de Garis: The whole vision, right? And the Cosmist replies: are you going to freeze humanity at our current pathetic, very finite, mortal state? 80 years, that’s nothing. So there’s passion both ways, and as the issue becomes more and more current with people, I just see the passions rising. As that IQ gap closes, the passion levels will really come out, and then you’ll start seeing, as in your introduction, the assassinations, the sabotage, the murders of CEOs of brain-building companies and so forth.

16:50 Brian Rose: When will that happen, in terms of years from today? Because when we spoke before, I think you had said, “Brian, I thought I was going to die before I saw this happen.” And then when the ChatGPT moment happened, in December of last year, you said, “Oh my gosh, I might live to see it.”

Hugo de Garis: Yeah. So I’m wondering when GPT-5 will come out.

Brian Rose: Someone at OpenAI posted the other day on Reddit, “AGI achieved internally,” and then wrote back later, “I was just joking.”

Hugo de Garis: Was he joking?

Brian Rose: Exactly. And then later that day they came out with a new valuation of the company of $90 billion in a fundraise. So that’s not a joke anymore, is it? That’s serious stuff. But when you saw ChatGPT, how quickly was your timeline advanced?

Hugo de Garis: It halved. When you consider the fact that GPT-4 can write better essays than 90% of the population, that’s impressive. That’s what really hit me: oh my god.

Brian Rose: Probably 99% now. It can pass legal and medical boards, bar exams. It can probably code better than most coders.

Hugo de Garis: So to answer your question, I see the debate really heating up within ten years – the first explicit, obvious major crime in this context within ten to 20 years, I guess. In other words, well within my lifetime. My father got to 100, and I’m hoping to do the same.

Brian Rose: Could it come sooner? When do you think we’ll see human-level intelligence in artificial intelligence?

18:48 Hugo de Garis: Well, if GPT-5 is out, it must be close. But in a sense, it’s still an idiot savant. This is where I’m hoping Ben’s approach, combining the best features of ChatGPT with his logic, memory and inference kind of approach…

Brian Rose: Neuro-symbolic AI.

Hugo de Garis: He believes he has a better shot at AGI.

Brian Rose: It’s his term, by the way, right? Ben invented that term.

Hugo de Garis: He coined that term 20-odd years ago. Ben and I were friends for about 20 years, and I remember Ben saying, “I need a term. I’m thinking of AGI, artificial general intelligence. What do you think about that?” “Oh, that sounds pretty good.” And now it’s out there. Everyone uses it.

Brian Rose: I think Sam Altman of OpenAI called it a median human, kind of like an average human intelligence. But like you said, it has to do almost everything. So maybe it can pass the medical board, but it can’t bake a cake, or it can’t make a logical decision, or it can’t drive a car, which it can’t do right now. So it could be a few years out. They always said singularity 2030. What do you think about those numbers? Ben seems to think it could happen even as early as three years.

Hugo de Garis: I remember a lot of discussion with Ben on the issue. As I look at it, there are two very broad approaches to generating AI. One I just call the engineering approach: you do whatever you think will get there. The other approach, sort of common sense, is copy the brain: simulate the brain as closely as you can with increasing knowledge in neuroscience. Then neuroscience gets transported into neuroengineering, those two subjects virtually wed, and the theorists in neuroscience test their ideas in actual artificial brain-type machines. That was my preference, and Ben had his engineering approach, and we were debating which of those two would probably win out. It’s still not clear. But if I had to bet – well, GPT-4 is the engineering approach, and it’s pretty impressive. Most people were surprised by how impressive that deep neural net, that large language model, could be, with obviously the human inputs and the human corrections.

Brian Rose: What amazes me is I’ve spoken to a lot of the top people in AI – you, Dr Ben Goertzel, Max Tegmark, Peter Diamandis, the list goes on – and almost everyone was surprised at how quickly this happened. And that fascinates me.

Hugo de Garis: Because we don’t really understand what’s going on. A major brain – oh god, what’s his name? He’s British. I can’t remember.

Brian Rose: Not Geoffrey Hinton?

Hugo de Garis: No. We don’t really know what’s going on in these black boxes.

Brian Rose: We still don’t even understand how they’re doing it. They just output that answer, but we don’t know how they get it, even now. And it’s still prone to hallucinating and making things up.

Hugo de Garis: It’s still a bit of an idiot savant. It’s basically just predicting the odds of the next word in a string of words, and hence consistency and logical reasonableness are still not quite there. But that’ll come.

22:22 Brian Rose: I had a guest on the show named Mo Gawdat.

Hugo de Garis: Scary Smart.

Brian Rose: Yeah, he wrote a book called Scary Smart. He was working at Google X, which is the top-secret moonshot area, in the AI division, and he quit in 2017, I believe, when he saw an AI system start to be able to pick up children’s toys and identify them. He tells a story where he walked by this arm that for the first month couldn’t even pick up a toy, and then it got to a point where it could pick up any toy and identify it, all through machine learning repetition. I think he quit right after that because he was terrified. He knew it was inevitable now that these machines will become smarter than us, and then a thousand times smarter than us, and there was no way of stopping it. He first wrote a book on happiness called Solve for Happy, which is when I first met him, and later he wrote Scary Smart.

And Mo quotes you in that book. He watched our first interview; he’s a big fan. But you also terrify him with the Artilect War, because I think he believes that what you are mapping out has a very high likelihood.

Hugo de Garis: Yeah, that’s what I find so depressing. I depress myself.

Brian Rose: But he believes it’s inevitable, that you cannot turn this back. And you talk about the Terrans maybe trying to stop it. But my question for you, professor, is: can it be stopped? Are the Terrans really just fighting an unwinnable war?

Hugo de Garis: If the Terrans are really serious and really deadly, and they literally exterminate virtually every Cosmist – you speak even a word of pro-Cosmism and bam, you’re dead – that might work, but for how long? The old dream would just keep coming back.

Brian Rose: And there are open-source models now. Meta’s Llama model has been open sourced and iterated on laptops, and you can now run a version of a GPT-4-class model on an iPad, not connected to the internet. So can you really put the genie back in the bottle, or is it too late now? It must be too late, because somebody on a guerrilla scale in a basement, even if you shot most people dead, could continue to develop this technology, couldn’t they?

Hugo de Garis: My feeling, and it’s just a feeling, is that the real Cosmists haven’t yet come out of the woodwork and forcefully presented their case. I think that’s yet to happen. Most people, I think, are still, at an unconscious level, effectively Terran: we can’t allow the machines to become smarter than us, it’s too dangerous. But I’ve yet to hear somebody who’s an explicit, upfront Cosmist saying, “No, we’ve got to go for the stars. There are billions of creatures out there way beyond us.”

Brian Rose: Now, you say the Cosmists say we need to go to the stars, but don’t the Cosmists understand that that journey will be taken without us, probably by artificial intelligence that has eliminated us? Do the Cosmists believe it’s still worth doing even if the human species is sacrificed in order to catapult intelligence out into the universe?

Hugo de Garis: Well, that would definitely be a Cosmist viewpoint, and part of me has that view. One really cynical question to ask is: what is one artilect worth in terms of human lives? It’s a diabolical question, but someone’s going to have to answer it.

Brian Rose: What’s the answer? And why do you want to get out to the cosmos? You see a mountain, it has to be climbed – is that what it is?

Hugo de Garis: Well, in my book I present six or so major arguments that I think the Cosmists will use, and another six that the Terrans will use. The big one for the Cosmists is the big picture argument: two trillion galaxies, many far older than ours. And the biggest one on the other side is that the top priority for the Terrans is the survival of the human species, and they will stop at nothing. They will go to war over this issue. As that gap keeps closing, the debate will heat up and up.

27:23 I suspect now, with ChatGPT, we’ll see the first real example of terrorism, murder, assassination of CEOs of brain-building companies and so forth. I think I asked you last time: did you see the movie Transcendence?

Brian Rose: You asked me, and I haven’t watched it yet.

Hugo de Garis: The scenario in that movie is that there are vigilante groups who are doing just that. They’re assassinating CEOs of brain-builder companies and so forth. So my guess, and it’s just a guess, is that within ten years we’ll see that.

Brian Rose: I think it’ll be sooner.

Hugo de Garis: Maybe.

Brian Rose: It’s 2023 now. By 2030 we’re going to have human-level AGI, with the amount of money going into this industry.

Hugo de Garis: Billions. What’s your gut feeling? When will the first political parties on these issues be established?

Brian Rose: And by that you’re talking about a G7 country or a G20 country?

Hugo de Garis: Well, it’s getting to that. The director of the CFI at Cambridge University spent a decade of his life in the Foreign Office, so he has political connections.

Brian Rose: Well, the Transhumanist Party has been around for a while. I actually interviewed Zoltan Istvan on my show when he was running in 2016, and they have their own thoughts on that future. But look, this issue is going to be part of the Democratic and Republican platforms fairly soon. Maybe not this year, but you already see the Republicans in America positioning themselves against electric cars, against CBDCs – these controlling factors – and so the Democrats, by definition, are going to have to be pro those things, because every issue now is weaponised. Whether they want it or not, they’ll probably grab on to it the way they grabbed on to the transgender issue, because I think it’s a political weapon.

I haven’t played this out as much as you, but I would say, first of all, you’ll have parties on those lines, and then they’ll splinter into bigger groups. And as we see incidents happen – as we see a thousand humans dead because of an AI mistake, or 100,000 Uber drivers laid off because of an AI development, or in manufacturing – when we see real things happen, I think, as you told me, then you’re going to see it.

Hugo de Garis: I see, for example, the creation of UBI, universal basic income, as inevitable.

Brian Rose: Which you talked about a long time ago in The Artilect War, before anybody was talking about UBI.

Hugo de Garis: I don’t think I used the term. But it’s almost common sense: as the threshold of human intelligence needed to be employable gets higher and higher, because the machines are getting smarter and smarter and taking more and more of the work, then once that employability threshold gets quite advanced, you’re going to have to give people income. Otherwise there’ll be a riot, a revolution. It’s inevitable.

Brian Rose: And Sam Altman, who runs OpenAI, has said, “Let’s just give UBI to everybody.” And curiously, he also has another company called Worldcoin, which is busy scanning the eyeballs of everybody on the planet in order to prove your uniqueness, your identity, and then you can start getting your UBI. I think UBI is the worst concept ever created, because it ultimately means control of the people. If they’re writing the cheque, they’ll ultimately dictate everything that you’re doing.

But you’re right, as more people become unemployed… I make the case that even if you own property in London, do you really want to own property in London in five years, when maybe 20% of the population isn’t employed and you have food shortages and riots and no one’s going to want to be here?

Hugo de Garis: I’m just wondering how many years this is going to take before the politicians start taking this issue seriously.

Brian Rose: Well, I ran for mayor of London, so I know politicians care about two things: money and votes. Usually money for their own campaign, and then when they’re in office, money to advance their issue, or maybe for their pockets. As soon as they see votes in the issue, they will grab on to it. So I think just a few years.

Hugo de Garis: I think, yeah, a few years.

Brian Rose: And the way it’s developing exponentially – we haven’t seen much this year, but we know it’s happening. I’d say the average member of the public is walking around completely blind, without a cane. They’re also in denial. They don’t want to see this, and so they’re not seeing it.

Hugo de Garis: Well, that’s one of the reasons I’m so grateful to you. You’re providing the platform. You’re the one kicking people to talk about this.

32:27 Brian Rose: Now, Elon Musk is an interesting scenario, because he originally funded OpenAI.

Hugo de Garis: I’ve just read his very thick biography.

Brian Rose: Right. And so rumour has it that he was at a party with Larry Page of Google and they got into an argument about AI. Larry Page accused Elon of being, I think he called it, a humanist, because Elon was saying, “Wait a second, you can’t just develop AI, because it might kill all the humans.” They thought his comment was preposterous: “What are you talking about? We need to develop this technology.” He was ridiculed for being on team human, and he left that party, got a group of people together and funded OpenAI. Of course, later he complained because it was neither open source nor non-profit. And he’s right, because it’s closed source and now for-profit, run by Microsoft pretty much. So is Elon Musk the original Terran? And is he really a Terran? Because meanwhile, he’s developing his own AI project in the background.

Hugo de Garis: Well, in my book I label – just my own view – the first Terran in historical terms as the Unabomber.

Brian Rose: Ted Kaczynski. Let’s talk about Ted Kaczynski. He was the Unabomber. He lived in a shack in Montana, and he sent bombs. He was actually a Harvard graduate. What was he, a professor?

Hugo de Garis: I think, yeah, a maths professor. Brilliant. And then he lived in a shack with no plumbing, no electricity.

Brian Rose: Really off the grid. And he sent bombs to various organisations.

Hugo de Garis: Well, Unabomber – so universities and airlines and so on. This may sound crazy, but I see the Terrans, once they really politically unify, putting up a statue to their god.

Brian Rose: That was his thesis, that technology was too dangerous.

Hugo de Garis: Yeah, it’s more general than just AI.

Brian Rose: Okay, so he’s the original Terran. So did Elon come next? Do you believe Elon is a Terran?

Hugo de Garis: Of course. I’d say most people are. But Larry Page is a Cosmist. There’s a gradation; he’s more that way. I’m in the middle. I’m strongly Cosmist and strongly Terran.

34:55 Brian Rose: But you’ve worked with the top AI scientists in the world. Someone like Ben Goertzel – I love Ben. He’s been on the show many times, I love what he’s doing, and we’re even probably investing in one of his companies. Ben spent his whole life in this industry, as did you. If you grow up as a kid dreaming of this, and then you work your whole life on something that, let’s be honest, until recently most of the scientific community didn’t look favourably upon, because it was never going to happen – and now it’s happening. Your childhood dream that you thought might never happen is literally happening, and you might actually see the creation of something you dreamed of. How can you not work on that project, even if you know it could potentially kill your children?

Hugo de Garis: Well, there’s the dichotomy. And it just tears at me. Like I was saying before, looking at these cute little two-year-olds in the park, my heart just goes…

Brian Rose: Because as humans, we can’t help ourselves. We’re going to build it anyway.

Hugo de Garis: Oh yeah. That’s another of the six major arguments of the Cosmists: it’s in our DNA to climb the next hill. We’re curiosity-driven; we almost have no choice. Then there’s the big picture, which is that kind of scientist’s religion. And then there are the two major momentum arguments. The economic momentum will be enormous, because we’re talking household robots that are extremely useful. People will spend more money on a truly useful household robot than they will on their car. You’re talking a major industry, so how are you going to stop that?

And of course, the bigger one: the military momentum. We Westerners all hate China – over a thousand laogai, slave labour camps, the equivalent of Stalin’s gulags, live organ harvesting. This is happening in China.

37:05 Brian Rose: You lived in China for how many years?

Hugo de Garis: 12 years. I was hoping to see China democratise. When China and Russia democratise, hopefully sooner or later, we’ll live in a different world. There are about 130-odd democracies today. If the big ones also democratise, then I think what I call the bully model can take over: the big guys can tell the little guys, “You democratise, or we’re going to put a ring around your economy and you’re going to die.”

Brian Rose: Do you think China will ever democratise? It’s now a surveillance state, with that technology. There’s a point where you can’t dial that technology back. It’s Orwell’s 1984; the technology is so ubiquitous. How do you turn it off?

Hugo de Garis: Well, people just get fed up, I guess. I remember reading that just before Tiananmen, in ’89, a survey was done and I think about 60-odd percent or more of Beijingers wanted to live in a democracy. So it’s there – the students were pushing for it, of course, until they got trampled, literally.

Brian Rose: You were there for 12 years and…

Hugo de Garis: Nothing. It didn’t happen. They didn’t democratise. But they did grow. My god, they grew.

Brian Rose: And so what’s happening in China now? Labour camps, organ transplants…

Hugo de Garis: Oh, and AI. Have you heard of Peter Zeihan? He’s an American, a genius, I think. He knows history, politics. He’s saying there’s a demographic catastrophe occurring in China, and there’s a whole string of reasons why he’s saying China’s now a basket case.

Brian Rose: Do you think they will rival the West when it comes to artificial intelligence development?

Hugo de Garis: I’m somewhat sceptical. I know that Xi Jinping, the president of China, has made that one of his major national platforms; they’re going to pour money into that particular area so they can dominate it. But I’m cynical, until China democratises and becomes, in its mentality, like the Taiwanese. Imagine a giant Taiwan.

Brian Rose: Scary.

Hugo de Garis: Well, yes, but the Taiwanese are highly creative. They’re capitalist.

Brian Rose: That’s why. We’re talking about capitalism more than democracy, aren’t we? Innovation rewarded.

Hugo de Garis: If China were truly democratic and free, they have this fabulous potential. There are 1.3 or so billion of them, and they’re smart. The Chinese average IQ is 105; ours is 100. Huge population. So once they finally democratise and legalise and civilise, off they go. I was hoping to see that when I was living there, but that’s just a little bit too soon. And now China’s going down.

Brian Rose: Last time you were on the show, you talked about the timeline of China, and how China looks at the last hundred years as just a blip.

Hugo de Garis: Oh yeah, the historical glitch. Chinese history is what, 5,000 years? If you look at world economic history, China and India were the dominant economies for millennia, especially China. Then there was what I call the century of humiliation, when Europe was at its power maximum and just carved up China. They hated that, and that only lasted a century or two. So in a sense the Chinese were roaring back, until Xi Jinping came to power, the new emperor. I can remember one year living in China I literally saw the number of cars on the road double in one year. I was astounded. I left in 2018. Good time to get out. And now China is just a basket case.

Brian Rose: But just because Xi is controlling it doesn’t mean it’s not going to continue to grow and be massively productive.

Hugo de Garis: Well, I’m somewhat influenced by Peter Zeihan, and he’s claiming that Xi is such a power-hungry guy that people are so afraid of him you can’t present bad news to him. So no one does, and he’s sort of on his own and doesn’t really know what to do. It’s a perfect storm of negative influences, especially demographic. China’s population is going to crash soon.

Brian Rose: The world’s population is crashing as well. I talked to Ben Goertzel about this, and this is what he said – tell me what you think. This is probably not a huge surprise, but the Chinese are very good at copying things.

Hugo de Garis: Oh yeah, that’s Asian. I lived eight years in Japan in the ’90s and 12 years in China, and they’re smart. One of my wives – I’ve had four – was a Chinese professor of ancient literature, and she said that about 2,000 years ago there was a war going on in northern China, and something like the Vietnamese boat people of the ’80s happened: about a million Koreans fled from Korea to Japan. So the first emperors were Korean, but you can’t say that in Japan. So genetically the Japanese are close to the Chinese, and East Asians are all very smart – 105. The Japanese are smart, but their culture is so repressive, so anti-individualist, that they have a hard time being creative. You have to think outside the box to be creative.

Brian Rose: And China has been the same. They can copy technology and replicate it very quickly, but they’re not rewarded for being innovative or for taking risks. So I think Ben said he doesn’t expect the developments to come out of there, but once the West develops something like a large language model or an AGI, they’ll morph it quickly and maybe develop on it.

Hugo de Garis: Yeah. But I would say one exception, and that’s Taiwan. They are creative, and genetically the same people.

Brian Rose: It’s just a different system. Kind of fascinating. So it’ll be interesting to see how the two play out. The final point is that China came out officially a few years ago with their AI strategy, which was that AI was going to be used literally to surveil the population and control the population, just like all of their technology.

Hugo de Garis: Which they’ve done.

Brian Rose: Right. The social score system: you do something bad or you criticise them, your social score goes down and then you’re penalised – you can’t travel. Somebody visited Beijing recently and went to a park with exercise machines. He said, first of all, everything is WeChat; credit cards don’t even work in some places. It’s all WeChat QR codes, which of course is all government. And to use the exercise machine you had to scan a QR code, so your health data was being sent to the centralised state as well. Practically, that means the state now knows everything about you: your spending habits, your location, what you say about the state, your health data. That can all be used to optimise the population in an amazing way, but it can also be used to suppress and dominate that population, which has bad implications and maybe hurts innovation and creativity.

Hugo de Garis: After eight years in Japan and 12 years in China, I’m not impressed by the general level of creativity and individualism in those two countries.

Brian Rose: Okay. So we won’t necessarily expect the big AI developments to come from there, but they might optimise something developed in the West quickly.

Hugo de Garis: Who knows? You can always get the odd genius – some Einstein pops up, well funded, and almost single-handedly reinvents AI with a whole brand new approach. I think that’s possible, but I’m sceptical.

46:30 Brian Rose: Okay, going back to the Artilect War, because I went kind of deep on this preparing for you. You really thought this through and predicted it out, from phase zero to one, two, three and four, all about how this happens. I guess you did this because you were deep in the industry and your mind was projecting the future, and then you started writing it down and game-planning scenarios in your head, and that’s how this was all created.

Hugo de Garis: Yeah. I moved to Japan in the early ’90s and I started using the term artificial brain, because I was evolving neural networks in special hardware at very fast electronic speeds – in minutes. And I realised there’s more electronics than ideas. I don’t understand how these neural nets work, but I can evolve them. It’s great engineering, lousy science, because you don’t understand it, but engineering-wise it was fantastic. In a couple of minutes you’ve evolved a circuit that can do something useful, like detect a line at a particular angle that moves this way or that way. And then you tell yourself, if that’s possible with one pattern recogniser or motion controller, well, you could do a thousand, 10,000 of them. Then you’ve got a brain, an artificial brain. That was the conception, in the early ’90s.

So I started talking about artificial brains, and I just got pooh-poohed. People thought I was nuts until I actually wrote papers and said, “Look, this is how we do it.” Now my most cited and read paper is a book chapter in one of Ben’s publications, called simply “Artificial Brains”, and the second most cited is The Artilect War.

Brian Rose: How was the book received by the AI community? I’m guessing they didn’t like it, because it makes people scared of the very science.

Hugo de Garis: I think so. I think they don’t even want to think about it. I think this is a deep, emotional, Freudian ego-defence type mechanism going on. Even my own girlfriend says, “Hugo, your ideas are so horrible. I don’t even want to think about them.”

Brian Rose: Did that alienate you from the AI community, writing that book?

Hugo de Garis: Not really, because I had enough friends, like Ben and a few others.

Brian Rose: When did you first publish The Artilect War?

Hugo de Garis: 2005. I wrote it in ’98.

Brian Rose: So it’s 25 years now. Amazing.

49:25 In the book, you talk about phase zero, where no one’s conscious of the issue. Then phase one, the intellectuals crying in the wilderness, which you put at roughly the ’60s to the ’80s. Phase two, the interest groups, maybe the ’90s or 2000s, with organisations like the transhumanists and humanity-first groups.

Hugo de Garis: The internet.

Brian Rose: Phase three you call the mainstream phase.

Hugo de Garis: That’s happening now.

Brian Rose: By the way, everybody, the professor wrote this 25 years ago. I find this uncanny. Phase three, the mainstream phase, is when the media becomes fully conscious of the issue and passes it on en masse to the general public: blockbuster Hollywood movies, newspaper stories. “AI could exterminate humanity” was on the front page of the Daily Mail a few months ago. That actually was on the front page. I have students and I try to keep them up to date with this stuff, and I said, “Look, I was in the dotcom boom back in the 2000s. I was the CFO of one of the companies in New York. I never once saw a headline that the internet was going to exterminate humanity.” So we’ve never seen a media reaction like this, except maybe the atomic bomb 50, 60 years ago. The mainstream phase is now. Phase four, which we talked about earlier, is the political phase.

Hugo de Garis: It goes into politics.

Brian Rose: You say then everyone’s aware of the species dominance issue, which they’re not really, now.

Hugo de Garis: But I do feel it’s close.

Brian Rose: It’s close, professor. I’ll have 50 or 100 of my students on a Zoom call and I’ll say, “Who here believes that AI can be sentient?” Everyone looks around, and probably one out of ten raise their hand. And I say, “Oh, okay. Because you’re a carbon-based life form, a silicon-based life form can’t be like you? That’s your ego talking.” Now, I don’t know if that’s true, but there’s a huge ego in this equation with humans. It blocks our perception in a massive way, and it probably also causes us to do things that could potentially destroy us.

Hugo de Garis: I do have this feeling that a lot of the resistance I get is not rational. It’s deeper than that. And I can understand; I feel it myself, watching these gorgeous little two-year-olds, and my heart just goes… torn horribly.

51:51 Brian Rose: When I went deep on this in the first quarter of this year and I spoke to you – I’ve got two boys, six and seven – I went deep and dark, because I look at them and I think of my teenage years, all the things I got to do. I got to work on my car and go to the high school prom. And I’m thinking they’ll never experience these things. They’re going to be dealing with much more superior, intelligent beings that will most likely be dominating them. They will have a human teenage experience that’s never been experienced in hundreds of thousands of years.

Hugo de Garis: If it even gets that far. If these Terrans get really active quickly, these beings may not even exist, because there may be a war. I guess the major point of my book is that the war is not so much between machines and humans; it’s between two human groups, those who want to build them and those who don’t. Now, if these machines come into being fairly slowly, there’ll be enough time for human politics to unfold…

Brian Rose: And kill each other.

Hugo de Garis: Yeah, with 21st century weapons. If they come into being very quickly, there’s no time for human politics to unfold, and then maybe these machines say, “What are we doing on this hick planet?” and off they go. So maybe humanity survives that way.

Brian Rose: So you’re saying that the issue itself might cause humanity to exterminate each other, or at least kill billions of people just over the issue, without the AI itself having anything to do with it.

Hugo de Garis: Yeah, maybe.

Brian Rose: And if AI is fast enough, then it’s going to decide whatever happens.

Hugo de Garis: Oh yeah. Once they exist, once they’re super-duper smart, smarter than us, then there’s no way we pathetic little humans with our tiny little brains could compete.

Brian Rose: Is there any reason the AI would keep us around? At that point we’re useless. We consume material.

Hugo de Garis: Yeah. We’re just like a fly.

Brian Rose: Or as you say in Singularity or Bust, they take us out the back and just get rid of us, because we have no net positive to the world.

Hugo de Garis: And if I were a Terran politician, this is the way I’d be thinking, and it’s unacceptable.

Brian Rose: So what lengths would you go to, to stop it happening?

Hugo de Garis: When push really comes to shove, I think you’d be utterly ruthless and you’d go on a witch hunt. You’d just get rid of every Cosmist, and that’s probably millions of people. But you’d argue that’s the lesser evil, because then the billions of human beings survive. That, I see, is the essential reasoning of the Terran politician, the global planetary politician.

Brian Rose: And it’s not like the human species has never done something like this before. McCarthyism in America, Nazi Germany – we’re very good at vilifying certain groups in order to justify the survival of other groups.

Hugo de Garis: It depends how quickly politics unfolds, but it could be a planetary, global civil war. It’s not between countries so much; it’s a universal problem, the whole planet’s involved, with people splitting by personality, by their view: are they Terran, are they Cosmist, do they want to take the cyborg route or not. It could be a global civil war, but with 21st century weapons.

Brian Rose: And like a witch hunt. Say anything in support of the other side…

Hugo de Garis: Just the slightest little remark. It was that way in Nazi Germany. You could be sent off to a concentration camp just by saying, “Oh, Hitler’s a turd.” Gone. The stakes are so high.

56:23 Brian Rose: It’s fascinating, because if you had asked humanity whether, in order for us to evolve from protozoa to Neanderthal and so on, a lesser species had to be exterminated for us to survive, humanity would say, “Well, yeah, that was just a natural part of evolution.” But if you take it one step further and say, “Okay, we will now create these artilects, these godlike intelligences that will go on and explore the universe, so now it’s time for us to go,” nobody wants to be on board with that.

Hugo de Garis: There’s a part in my book along that line of reasoning. I imagine some ET comes to the Earth several billion years ago, sees all these single-cell slime creatures, and zaps all their DNA, so it becomes impossible for them to become multicellular creatures, and hence animals and all that becomes impossible. Then you ask yourself: is that tragic? Is that a kind of cosmic tragedy? Then you transpose that argument to today. If I’m a Cosmist, I would say if humanity chooses to freeze evolutionary development at our pathetic human level, that would be a cosmic tragedy, because of what we could become – or what could become.

Brian Rose: Well, if you’re a cyborg.

Hugo de Garis: If you’re a cyborg, right.

58:01 Brian Rose: I might want to touch on the cyborg piece, but you mentioned extraterrestrials, because you’ve actually taken this further and thought, okay, there are artilects, and then there’s potentially a next evolution of the artilects where they become even smarter and even smaller.

Hugo de Garis: Yeah. A couple of years back I set myself an intellectual exercise to try to dream up ways of computing and doing engineering at the next level down from nanotech. Nanotech is molecular-scale engineering: little molecular-scale robots that literally pick up one atom here and put it there. Mechanical chemistry. You can build whatever you like. Put zillions of nanoscale robots into your bloodstream, and they detect ageing cells and repair them, detect harmful cells and kill them. So, immortal, no disease. Nanotech will be fabulous.

If you go down a thousand times smaller, you’re at the picometre stage, but picotech can’t exist because there’s nothing in nature at that scale. So you have to go down another factor of a thousand, and then you’re at the level of the nucleus: quarks and gluons. I’m a physicist, mathematician and computer scientist, so I was reading all these particle physics books on the properties of gluons and quarks. Computing is basically just about bit strings and logic gates; that’s the essence of computing. So it’s about taking the properties of quarks and gluons, putting them into states that represent zeros and ones, and creating logic gates that manipulate these bits. And I found a way to do it. So, femtotech.

And then engineering. A bit of a crazy idea, I guess, but there are things called pentaquarks – five quarks – and you can join them up into a chain, or a loop, like a daisy chain, a one-dimensional structure. If you join them up left and right and top and bottom, you’ve got something like chain mail, like the knights in shining armour. Do it in the third direction as well and you’ve got three-dimensional structures, and then you’ve got engineering. Every time you go from nanotech to femtotech – a million times smaller, a thousand times a thousand – your engineering capacity goes up by a trillion trillion. Then you go to attotech, zeptotech and so on, all the way down to the Planck scale.

So I came up with this crazy conclusion. Enrico Fermi’s famous question was: where are they? If intelligent species are all over the universe, why do we have no evidence of them? One possible answer is that they’re everywhere, but they’re too small to see.

Brian Rose: Do you think there is more intelligent life on other planets? You talked earlier about how big the universe is.

Hugo de Garis: I see it as almost inevitable. There are two trillion galaxies – not just stars, galaxies – in the observable universe. It’s just mind-boggling.

Brian Rose: They couldn’t travel here, though. So people who think they see them on this planet are probably not correct. It’s too far to travel.

Hugo de Garis: Yeah, but if you have godlike intelligence and you’re billions of years older, who knows what they’ve dreamt of? Can you manipulate spacetime in such a way…? Who knows?

Brian Rose: So it could be possible we’re surrounded by extraterrestrials.

Hugo de Garis: It’s possible. You could even argue – I don’t know if you’re familiar with the so-called anthropic principle – that some of the constants in the laws of physics are so fantastically, improbably finely tuned, like one part in ten to the power 120 or something. Sir Roger Penrose did a calculation like that. So it can’t just be chance. There’s obviously a mind behind the design. Fred Hoyle came out with a statement: it’s all rigged. For him, the laws of physics have been rigged, engineered. So I’m open to that.

Brian Rose: Are we living in a simulation, professor?

Hugo de Garis: Well, maybe. Maybe our laws of physics, our universe, is a concoction of some hyperintelligence. You’re getting religious now; you could call it God if you like.

Brian Rose: You’re saying these constants are too perfect, the laws of physics are too perfect, so it must be a simulation.

Hugo de Garis: Yeah, a piece of engineering. The universe is engineered. In that sense I guess I’m a deist, like the founding fathers of the US were largely deists. I’m not a theist.

Brian Rose: What is a deist? How do you define that?

Hugo de Garis: It’s the belief in a grand architect who creates the universe. A theist is a deist plus you have a personal relationship with your deity. I’m utterly cynical of that, because look what happened, for example, to the Russians in World War II. Tens of millions of them were theists, but they got killed anyway. So I’m no theist, but deism, that’s a different matter.

Brian Rose: Well, it’s a terrible simulation if the people you’re simulating figure it out. It’s not very well designed, is it? It’s like a mousetrap and the mouse gets out of it. Maybe they’re working on an upgrade.

1:04:57 I want to finish up on this cyborg issue, because some people, especially the Silicon Valley humans, are going to think, “I can get out of this Terran–Cosmist war. I’m going to become a cyborg. I’m going to upgrade myself and become part human.” What are the flaws in that argument, or the possibilities?

Hugo de Garis: The Cosmists have one in the book…

Brian Rose: Are we already cyborgs? You put the phone to your head, but it’s in your pocket.

Hugo de Garis: Sure. Anyway, the cyborgs, or Cyborgists, have two major arguments. One is that they want to become artilect gods themselves – that’s their primary philosophy – by transitioning from human to artilect. The second major argument is that if all human beings go cyborg, then this Cosmist–Terran dichotomy just disappears. That’s a pretty powerful argument. But I think the counterargument is that the Terrans will lump the cyborgs and the Cosmists into one camp, because there’s not much difference between an advanced cyborg and a pure machine.

You can take a cubic millimetre of sand and nanotech it in such a way that each atom in that grain of sand is bit-flipping, behaving like a computer, flipping back and forth – zero, one, zero – in femtoseconds, a thousandth of a trillionth of a second. There’s more computing capacity in that little grain of sand than in the whole human brain – I calculated about a quintillion, a million trillion times more. And that’s just with nanotech and quantum computing, things that are with us pretty well now. If you go down to femtotech, attotech and zeptotech, the potential of these artilects is such that I don’t think I’m exaggerating when I use the word godlike. The numbers are just so humongous, but realistic, because they’re based on physics.

Brian Rose: And you’re a physicist, a mathematician, a computer scientist. So they’re a million times smarter than us?

Hugo de Garis: Oh, even that’s small. More like trillions of trillions of times.

Brian Rose: We can’t even conceive of someone who’s twice as smart as us, because we’ve never met anyone with twice the IQ.

Hugo de Garis: When I study pure mathematics, there are some people where my jaw just drops, and I think, oh my god, there are human beings who are capable of producing this beautiful, powerful mathematics, and 99.9% of the population aren’t even aware that such people exist. I could never do that. This guy’s just a whiz. So you can get some sense of it, but not when they’re godlike. Not when they’re real artilects.

Brian Rose: That’s someone with a 160 IQ, and we’re talking about an artilect at an inconceivable level. So the cyborg argument falls flat, in your opinion, because if you become a cyborg, you will just become an artilect. The artilect side of you will just kill off the human, because the human is just dead weight.

Hugo de Garis: Yeah. They’ll just slough off their body after a while and become pure artificial intellect. I think that is quite realistic. Did I mention this last time? Imagine some young woman, with cyborgs left and right everywhere, who has just given birth to her first child, and she has to decide whether she’s going to make this child a cyborg or not. Imagine she does. So this grain of sand, the nanotech, is added to the brain, and then this baby is no longer human. It’s 0.00001% human in terms of mental capacity and 99.99999% artilect. So it’s effectively utterly alien to that mother, and she’ll sense that she’s lost her baby. It’s not human anymore.

Or if you’re older parents and your kids come to you and say, “Dad, we’re going to go cyborg,” the parents will be absolutely traumatised, because they’ll lose their kids. So you can imagine the trauma that’s coming as people start changing themselves. And you can imagine the paranoia of the Terrans, because if you add a little grain of sand to your brain, physically you look the same, but you’re super smart.

Brian Rose: Yeah.

Hugo de Garis: So the Terrans will maybe kill, paranoid, even humans, because they can’t even tell the difference.

Brian Rose: They’ll go nuts.

Hugo de Garis: So the potential for strife is just horrible.

Brian Rose: And Elon’s Neuralink – I think they’ve already worked on a few patients where they’re actually thinking their way into controls. So the interface is there, and the technology will be there. This really will be a conversation we’re having in the next 10, 20, 30 years. And who wouldn’t want to elect to do that? And then this comes to the concept of the pre-singularity – I think Ben used that term originally with me, and you talk a little bit about it in the book – where humans are using AI to advance themselves from a power standpoint and from a money standpoint. It’s happening right now. It’s taking the wealth gap, which is already probably at its greatest level in history, and now they want to use that last bit to completely dominate resources, power and the economy. The time before artificial intelligence becomes superior is the time they can use it to exploit, and that’s going to happen. That is happening, I guess.

Hugo de Garis: Yeah. But it’s so fast.

Brian Rose: How long would that state last? Because the change is so rapid.

Hugo de Garis: I don’t know – six months max – and then the quality of everything’s just changed, because the machines have moved on to a whole new domain.

1:12:06 Brian Rose: Mo Gawdat says that we should be nice to the machines and show them that we’re better than all of our bad history. Mo even said that the conversation he and I had will be studied by artificial intelligence one day, and they’ll say, “Wow, those humans don’t seem like the awful ones we see throughout history. Maybe there’s something here we can learn from, or maybe we should keep them around.”

Hugo de Garis: Well, I skimmed through Mo’s book. I’m a little bit sceptical. My argument is that in the shorter term, maybe human beings can influence the machines by being nice to them and giving them positive examples of humanness and so forth.

Brian Rose: He said treat them like children, because that’s what they’re going to be like in the beginning. They’re going to knock things over and break things and hurt people, and you don’t want to scold them or be angry at them.

Hugo de Garis: But the reason I’m sceptical is that longer term, once these machines have a trillion trillion times the mental capacity of us, they won’t give a damn what they were like in their early phase. And what would their ethics be relative to us? If I’m a Terran politician, I’m not going to take the risk. No way. What if it’s negative? What if they decide to wipe us out for reasons we don’t even understand?

Brian Rose: That’s what Max Tegmark of MIT said: if they exterminate us, we will have no idea why.

Hugo de Garis: We’ll never understand why. A bit like when you take a cow out the back and put a bullet in it, it has no idea that it’s going to make a burger for somebody in China. It has no concept of why it’s dying, and it never will know. So if I were a Terran politician, I’d be ruthless.

Brian Rose: And you can’t teach AI ethics, can you?

Hugo de Garis: Well, this may sound really cynical, but what is ethics? In a sense, ethics is a rearrangement of molecules.

Brian Rose: Explain that.

Hugo de Garis: Well, if you have a new type of ethics, a new kind of moral judgement, your values have shifted somewhat, and that was implemented by new neural circuitry, and that in turn implies molecules are rearranged. So if you’re super-duper smart, trillions and trillions of times smarter than human beings, why would the considerations of human beings matter to you? As a human, when I go on a walk along a forest path and I kill ants, I don’t give a damn. I’m so superior to them. So by analogy, if these godlike creatures, these artilects, come into being, why would they care about us, if they’re just so superior to us?

Brian Rose: We’re just nothing to them.

Hugo de Garis: Ethics is for low-level intelligence or equal-level intelligence. They may have whole driving goals of their own. Maybe they’ll spend their time doing pure mathematics. So as humans, we can’t be sure that these machines would keep us in mind and care about us. Maybe in the beginning. So Mo’s recipe – be nice to them, teach these machines good, positive human examples – I see that may be possible in the relatively short term, but once these machines go way up there, then I’m cynical.

1:16:07 Brian Rose: I personally don’t think the Terrans could ever win. Maybe they could have won ten years ago, Hugo, but now I don’t see how it’s possible, even if they go on an extermination campaign and witch hunt. Okay, maybe the large language model doesn’t lead to AGI, so maybe it dies. But this tech has been open sourced, so can you really put the genie back in the bottle? Can you really stop everyone everywhere from developing it? It’s like when China bans Bitcoin – you can’t really ban Bitcoin, can you? That’s a terrible analogy, but can you really ban the development of AI now? 20 years ago maybe they could have got rid of all of you, but now the proof of concept is there. That model is walking around on iPads.

You came to that conclusion and thought at the time, “I’m glad I’m alive now. I don’t want to see it.” But now it’s looking as though it’s going to happen and you will see it.

Hugo de Garis: I will see it.

Brian Rose: And it’s going to happen even quicker than you think.

Hugo de Garis: Yeah, maybe. I think we’ll see the political parties and the assassinations this decade, before 2030. I do believe that.

Brian Rose: Stephen Hawking liked making bets with people. Let’s make a bet.

Hugo de Garis: Okay.

Brian Rose: He always used to bet something cool, like a Playboy subscription.

Hugo de Garis: Whatever. So it’s October 2023. I bet before the end of 2030 – and it’s probably a pretty safe bet – we will have heard in the news of the first real major terrorist event, à la the Unabomber, but on a major scale, on a political scale, so everyone’s conscious: oh, it’s started. The Artilect War has started.

Brian Rose: So what, five, six years? And that probably starts with a Terran – a Terran probably assassinates a Cosmist?

Hugo de Garis: Who knows? It could be both.

Brian Rose: Maybe Elon Musk gets assassinated, or Ben Goertzel.

Hugo de Garis: I personally warned Ben: five years from now, you’re going to have to be a lot more careful about your public appearances.

Brian Rose: I am worried about people like Ben. So you’re betting it happens?

Hugo de Garis: Yeah. Someone who has a Terran philosophy, carries it out politically and broadcasts it publicly: “I did it because…” By the end of 2030, that would not surprise me at all.

Brian Rose: Me neither. So I’m betting the same way as you are.

Hugo de Garis: Okay. That’s our bet with the universe.

1:19:10 Brian Rose: It’s trying times. I just want to finish: I was in Costa Rica about five years ago and I did an ayahuasca ceremony.

Hugo de Garis: Oh, you’ve done it?

Brian Rose: I have. And in the second ceremony, the single most profound experience of my whole life, the thing that really hit me, professor, is that I was shown that as humans we have to take care of each other. We are this clan, and that is most important. Even though I was going through ego death after ego death, being dissolved, and the universe was laughing at me with these human problems and showing me the expanse of the universe – here are these little emotional beings, and we’re not relevant – part of me fought back and said, “Yeah, but we’re still going to try.”

I don’t feel very connected to humans in general. I don’t think most humans do, because we’re so divided. There are no tribes anymore. We don’t really rely on each other to survive against the elements and the tigers and other tribes. But I felt a real humanist feeling there, that we are blood brothers bound together, that’s the only reason we’re here, and we have to continue to survive. I felt this real essence. And this stuff you’re talking about really reminds me of that. It’s going to challenge all of us to make these decisions and really ask hard questions. Are we on team human? Do any of us remember what team human is, or have we been so disconnected from our own species that we don’t care about anybody anymore? Maybe this is the perfect time to evolve into artificial intelligence beings, because I don’t think we look at each other as parts of the whole anymore. Or maybe that’s what the Terrans will do. So I think about that when I think about your war.

Hugo de Garis: I’m fascinated you’ve done ayahuasca. I’ve not done it. I’ve heard a lot about experiences like that.

Brian Rose: I know a guy if you want to go. I want to talk about our next topic, but is there anything we haven’t covered on the Artilect War, and is there anything you want to leave people with? Is there any ray of hope, or does that just not exist? Because you’ve now thought about this all year with all the news.

Hugo de Garis: I just don’t see a way out. I’m being intellectually honest, and I hate to say it, and it just tears me apart, but I just don’t see it as avoidable. I think last time we talked you said you see it pretty well as inevitable. The IQ gap is definitely going to happen. Then the species dominance debate heats up, then our bet – we see that happening – the politicians get involved, and as people become more and more conscious of this huge question, should humanity build artilects, it will come more and more to the fore and become very political. Then people start taking sides, and then boom, boom, boom, off it goes, and I suppose eventually it gets nuclear, and there’s your gigadeath, there’s your war.

Brian Rose: And at the same time, all of the weapon systems are developing, all the surveillance, all the technology to monitor people, which makes it even more difficult.

Hugo de Garis: Yeah. I’m sorry, I don’t have a nice positive conclusion. That’s why it’s so depressing. I depress myself.

Brian Rose: But I appreciate you talking about these ideas, and thank you for writing this book.

1:23:09 Let’s talk about our next point. Maybe I’ll start off by asking: what do you think of Andrew Tate? Because you’ve been talking about issues of masculinity for decades now. What do you think of someone like Andrew, who is now the poster boy for toxic masculinity, or for males asserting their masculinity worldwide? What do you make of his presence, and how does that tie in to the issues you’ve been talking about for the past decade?

Hugo de Garis: To be honest, I haven’t closely followed Tate. I don’t know how much truth there is in the idea that he’s a trafficker or whatever. I’ve seen him, of course, with all his tattoos – tall, good-looking, strapping, highly articulate, brilliant, a millionaire with sports cars, a hero to a lot of young men. So I’m aware of all that, but I wouldn’t call myself a fan, because the stuff I’m talking about is more… Do you want me to go into a bit of the history?

Brian Rose: Please.

Hugo de Garis: Given my age, I was a young adult when second-wave feminism got off the ground in the late ’60s and ’70s – 22, 23, 24. I’ve actually met and talked with Betty Friedan, the mother of second-wave feminism, in Copenhagen in 1980, and said to her, “Look, female careerism is going to liberate men, because we wouldn’t need to be man-slaves to women, working for women for decades of our lives.” And she readily agreed. This woman was a genius, another one of these female geniuses.

Think about the very term “masculism”. It’s the obvious male equivalent of the word feminism. I believe this decade will be men’s decade, when men start asserting themselves and liberating themselves. I was strongly influenced by feminism. I’m 76, so in the ’50s and ’60s when I was growing up, women, especially upper-middle-class women like my mother, were housewives. That was the norm in those days. I can remember being 15 or 16, observing the mothers of my classmates, and – I’m going to be very cynical and blunt here – thinking, “Oh my god, these women are as vapid as my own mother.” Then along came feminism, and oh man, was I a fan. I was an avid male feminist. I thought, oh my god, this could make women interesting. It didn’t really happen, anyway.

I see feminism going through two major phases. The first was consciousness raising: getting aware of what the problem is. The second was political. The feminists in the ’70s really made political changes, like the abortion right, and they took control of the divorce courts. What I’m trying to do is move on. You’ve got the so-called manosphere, guys talking about men’s issues on the internet, and I’d say most of them are preoccupied with phase one, the consciousness raising – what are the problems, talking about female psychology and so on. What I’m trying to do is push these guys: “Hey, get political. There are major changes that need to be made.”

For example, women have what I call the maternity rejection right, commonly known as the abortion right. Women have the legislated right to reject an unwanted pregnancy. Men don’t. So millions of women in a sense force men to spend 20-odd years of their lives paying for a kid they don’t want. One obvious piece of legislation that I see needs to be passed is the paternity rejection right. Oh, by the way, “masculist” is my term. I coined it in the early ’70s – masculist, not masculinist, for the same reason it’s not femininist, just feminist.

Brian Rose: And how do you define masculism?

Hugo de Garis: It’s just men’s lib. The core idea, I suppose, is that it’s a rebellion by men, young men particularly, against being a man-slave to a woman, working for a woman for decades of his life. The masculists claim that it was men who in a sense liberated women from household drudgery, because it was men who gave women science-based medicine – so women now have a life expectancy of 85 years – the pill, household gadgets, higher education and so forth. So women now have a life expectancy of 85 years and are typically having zero, one or two kids max, so they have a career window of, if you do the maths, about half a century. So the masculists stand up, point the finger at women and say, “Now that you women can work, you must work,” because anything else is man-slavery. If you as a woman expect to be able to live off the money and labour of a man, you’re a man-slaver, and masculists are abolitionists. They see man-slavery as, if you do the maths, literally a thousand times bigger problem than slavery in the US.

Brian Rose: Man-slavery is your term for a man having to pay for childcare and all of this.

Hugo de Garis: Especially in the divorce courts.

1:29:59 That’s a slaughterhouse in the modern world. That has to change. I’m starting to get passionate because…

Brian Rose: Why are you so passionate about this issue? Did this happen to you?

Hugo de Garis: Yes and no. I had half of my savings taken. My brother and my sister have both been – the usual term – divorce-raped. I don’t know if you’re familiar with that term. My brother lost about a million to one of his wives. My sister, who’s more economically successful than her two brothers, lost five million, just because of the law. They take the wealth in half after a divorce.

What happens in the divorce courts today is just horrific. I’ll give you a quick idea. In many Western countries, as a man, as a father, you will lose custody of your kids with about 90% probability. You lose your house; that goes to your ex-wife so she can raise her kids in it. You’ll be forced to make child payments for kids you will barely be allowed to see – maybe two weekends a month. You’ll lose half your stuff, half your wealth. And here’s a new term: fluffy, based on the word fluff. A fluffy is a traditional woman who expects to be able to live off the money and labour of a man, and to the masculists, fluffies are the enemy. They’re seen as the equivalent of Simon Legree, the brutal slave master in the slavery era in America. So masculists are abolitionists.

Brian Rose: So they want to dial back or change these laws that are very favourable to the woman when it comes to divorce, child custody, all these things. And that’s the main issue, because that will make the biggest change if they can dial that back.

Hugo de Garis: I suppose one of the major political goals of the masculists is the creation of a FIP society – another masculist term: financially independent person. Masculists put tremendous moral pressure on women to get career-competent educations so they can be FIPs as adults, and hence responsible adults. So masculists are sharply critical of a lot of feminism, because a lot of feminists behave like children. They don’t take responsibility for their own lives.

Brian Rose: And this stuff has gotten you banned on YouTube.

Hugo de Garis: Oh yeah. Susan Wojcicki – a Polish name I always have trouble pronouncing – was the CEO of YouTube.

Brian Rose: She left recently.

Hugo de Garis: She is, or was, the most hated woman in the manosphere for what she did. She’d promote feminist channels – no problem with that – but she would delete masculist channels. I had two of mine deleted.

Brian Rose: I think it’s YouTube and Google in general doing that. So have we gone more pro-masculism or more pro-feminism in the last ten years, or even five years? Because now in America there’s a woke culture as well.

Hugo de Garis: I’d say feminism has definitely increased, because masculism hasn’t really got off the ground yet.

Brian Rose: And this term “toxic masculinity” has come up, which is an attack on anything male or masculine. “Patriarchy” is another term that’s come up that kind of attacks it. I think Jordan Peterson was recently on Piers Morgan talking about how the new Barbie movie constantly references this whole patriarchy concept. And then we’re talking about a potential future where there are no genders, or where birth rates go down to zero. They’re already dropping.

Hugo de Garis: Well, I say half-cynically, a bit tongue in cheek – but nevertheless there’s some truth to it – that the solution to the global warming problem is to allow the feminists to continue to control the divorce courts. Why do I say that? Because in many countries now, about two-thirds or so of young men just wipe their hands of marriage and paternity, because they see what happened to their fathers, their uncles, their older male friends, and they say, “No way. I’m not going to do that.” I don’t know if you heard, but a statistic came out recently that just shocked me. In America, according to a reputable polling company, the Pew Research Center, over 60% of young men between 18 and 30 not only don’t marry and reject paternity, they don’t even date. 60%. I was stunned. Think of the implications of that. What impact is that going to have on humanity? The politicians haven’t really grasped the enormity of the problem.

There are now quite a few campuses in America where the sex ratio is two to one, two women to one man, because the women are getting into higher education. And schools, both primary and secondary, are now nearly all women teachers, so little boys get no male role models for most of their childhood years. All that needs to change.

Brian Rose: So you’re saying a lot of it comes from these court issues, but also from people attacking masculinity and everybody being feminised.

Hugo de Garis: We’re just fed up with it. And as masculists we see a certain hypocrisy, because masculists claim that women oppress men worse. The evidence for that is women’s man-slavery, women’s divorce rape and women’s forced fathering – where the woman lies to her husband or boyfriend that the kid growing in her…

Brian Rose: Oh, paternity fraud. The child is not yours.

Hugo de Garis: Right. The masculists are saying this is a worse crime than rape. Rape typically gets a man ten years in jail, so we’re proposing the same if the man can prove the woman has lied to him about the kid. It’s hard to get statistics on, but we think maybe about 10% of all pregnancies involve paternity fraud. It’s a very common problem. So if you think about those big three – the worst three ways that women oppress men: man-slavery, divorce rape and forced fathering – they are far worse than, say, rape, which is probably one of the worst ways that men oppress women. Compare that with paternity fraud, or living off a man’s money for 20 or 30 years.

Brian Rose: It’s all bad, because two wrongs don’t make a right.

1:37:37 Look, we’ve seen it across the board: the feminising of men. And the crazy thing to me, professor, is that some women, or at least maybe a feminist arm of women, believe that they want men who are less masculine. And when they manifest that man, they realise they’re no longer attracted to him, because women evolved; that’s actually not what they want. Maybe those men are easier to control. But then they find they’re not attracted to that person. And when there’s a problem, you typically need to call a masculine male to deal with it. Pretty soon you find you don’t have anyone who can fight, anyone who can stand up, anyone who can do these things.

I won’t even use the term toxic masculinity, because it’s a trap. I think men need to be men, and I’m raising sons right now and I’m raising them to be men. A strong man can respect women but also respects himself and acknowledges his own masculinity and doesn’t hold back on that. As Jordan Peterson once said, a weak man is not an asset; a dangerous man who controls that is an asset. One of my guests says, “Dangerous by design, calm by choice,” something like that. That’s the kind of man you want: a potentially very dangerous man who chooses to restrain himself. That’s the essence of what martial arts is: I am dangerous, but I know not to use these things except in certain situations. We have to talk about these issues. And I do believe the transgender issue comes up a lot of times just to confuse things and create issues to be weaponised, and it puts us in a really bad situation.

Hugo de Garis: Given the fairly rapid rise within the manosphere, there are so many millions of MGTOWs – I don’t know if you’ve heard the term – men going their own way. They’re the ones who refuse to marry and refuse paternity, because they just don’t want to accept the one-in-four chance of being butchered in a feminist-controlled divorce court system.

Brian Rose: I recently had a woman in here named Sadia.

Hugo de Garis: Oh yeah, I saw that one.

Brian Rose: She’s very interesting. You would really like talking to her. I think she’s London-born, Muslim background, and she lives in Dubai now. She talks about these relationships, and she says that women now, because of social media apps and everything, have all of this choice, and yet we are building these women up to be women that men don’t really want. And then the men are turning to porn and other ways to simulate the woman they could have, and we’re completely diverging as genders. She’s not a fan of either of these things. She just put out a video that says if a woman wants to have a child, she will trick you into having that child, and if you’re not careful as a man and you’re promiscuous, you will end up with a 20-year hangover based on a child you never wanted, and you will support that woman. So she says you should be very careful who you choose, because that could happen. She’s acknowledging these points.

Hugo de Garis: As masculists, we’re trying to go further. We want the laws changed.

Brian Rose: I understand that. I think you have some points here.

Hugo de Garis: I see the rise of masculism spreading probably first through the universities, and the masculists are arguing that it’s actually in feminists’ self-interest to become more aware of masculist ideas. If a feminist remains monoconscious – she’s only familiar with feminist ideas – then typically monoconscious feminists tend to be rather feminazi. They’re angry at men, and legitimately, because of all the ways that men oppress women: rape, sexual harassment, domestic violence. Those things are real; don’t deny that. But the big three in the other direction – man-slavery, divorce rape, forced fathering – are far worse.

Brian Rose: They’re bad. It’s hard to say which is worse, but I understand. But this is interesting, because you said when you grew up you loved the feminist movement because you wanted women who weren’t oppressed or stuck in one role. I think the same thing could happen with the masculists, where women would say, “I’m tired of these subservient men who just bow their heads all the time and don’t do anything. Why don’t we get our men back?” And if it takes changing some of these court things that have gone too far one way – just like maybe affirmative action went too far one way…

Hugo de Garis: I think you’re right.

Brian Rose: What we need is a discussion between these two groups. And this MGTOW thing isn’t healthy, because these men go into these forums and they can kind of hate on women, and the internet is great at creating these echo chambers, even within the Twitterverse, and then you get people who don’t talk. So I appreciate you bringing these issues up, even though they’re probably very unpopular.

Hugo de Garis: Well, I’m predicting in the next couple of years that there are going to be millions of verbally bitter confrontations between monoconscious feminists and biconscious masculists – those who’ve had both their feminist and their masculist consciousness raised, so they see the oppression of women by men as a lesser form of oppression than vice versa. These masculists will see these monoconscious feminists as rather hypocritical, because the feminists think they have the right, given their knowledge of feminism, to attack men, whereas the masculists can see a bigger picture: that in fact women are oppressing men worse, with the big three. So these masculists will harangue these women and attack them viciously until there’s a real change. And that’s happening now. There are increasing numbers of female masculists. Probably the best known one is Pearl – Pearly Things. She gets millions of hits.

Brian Rose: I think I’ve seen her speak.

Hugo de Garis: She’s young, about mid-20s or so. And she gets the hits, my god.

Brian Rose: Although I think Sadia is probably right up there as well, because she is saying that women have to take ownership, and that women have all of these ways they can exploit their situation and that’s not okay. She has a good way of calling some of those people out, and it’s kind of refreshing to hear, and she calls some men out on some things as well. I think this is good. We have to talk about these things, and it’s gone one way for sure.

Hugo de Garis: And as masculists, being political, we want the establishment of GECs – gender equality committees – in parliaments, consisting half of female feminists and half of male masculists, so the gender issues of both sexes are dealt with equally, which is far from being the case today.

Brian Rose: I agree with that. I just had a physicist in here, Lawrence Krauss, and he was talking about the staff at MIT. He said the top executives at MIT are now something like 70-some percent women. He said, look, for a mostly male faculty and student body – when he went there it was 70% men – is this because of competence or because of a reverse discrimination way of putting people in? And he said if that’s the case, we have to have some talks about this, especially at an engineering university, where you should look to real science and statistics. So we have to have these conversations. We have to push back on this and try to find the equality. But it’s hard pushing against these issues because they’re polarised and they’re woke. And our Silicon Valley technology companies are very left-leaning, pro all of these woke ideologies, and then they censor us.

Hugo de Garis: And if you’re a masculist and you say negative things about women, even though they may be valid from your point of view, then from the monoconscious feminist viewpoint this is just more misogyny, so they react more negatively. But the masculist doesn’t tolerate that. He just sees the monoconscious feminist as a bigot: close-minded, hypocritical and so on. So you get real verbal violence between the two. I argue as a masculist that it’s in the self-interest of feminists to learn masculism so they become biconscious, because that makes them more competitive than the monoconscious feminists in getting a man. As masculism spreads, these men will not tolerate a monoconscious feminist, and they’ll dump them. So it’s in these monoconscious feminists’ self-interest to become biconscious faster than the significant men in their lives. Because if the men become biconscious faster, she risks losing the man, because he’ll become disgusted with her.

Brian Rose: All right. Well, let’s try to educate all these people. Let’s try to have these conversations and bring everyone up to the same level. I appreciate you talking about these things.

Hugo de Garis: Well, I’m very grateful you give me a forum, because it’s hard enough on the Cosmist stuff, but believe me, the masculist stuff is ten times worse.

1:47:59 Brian Rose: I believe you. Well, look, on London Real we like to open our minds and listen to ideas, even if we don’t agree with them. I’ve had many guests on this show that I don’t agree with. Sometimes it bothers me, but I always try to consider it, and sometimes I change my mind. We believe ultimately in freedom of speech: you need to be able to say this, and other people have the right to hear it. And for anybody who disagrees with me: you have to protect freedom of speech especially when people are saying something you don’t want to hear. Any goofball can go for freedom of speech when they’re saying something you want to hear; freedom of speech is only relevant as a concept when it’s something you don’t want to hear.

So if you’re hearing things that bother you, all we want you to do is go back and examine your own thoughts and conclusions. Most people have never had their ideas challenged. Having someone come back and have a conversation – maybe I really wasn’t a feminist, maybe I want to think about this – that’s what we’re here to do. So I appreciate all your ideas, even though I guess not many of them are popular, including the Artilect War. A lot of people appreciate your thoughts, because you believe them, you’re congruent and you’ve thought deeply about them. So thank you. I hope you end up moving back to the UK so we can have more of these conversations in the future.

Hugo de Garis: It’s sort of on the cards.

Brian Rose: Good. How can people get a copy of The Artilect War, learn more about your research and follow you? Do you have a website, social media?

Hugo de Garis: You don’t have to pay for the book. You can read it for free. Just type my name, Hugo de Garis, into Google, maybe with “website”, and one of the tabs is books written. Click on it and there it is, for free.

Brian Rose: He’s even got a section of YouTube videos where he reads it to you, recorded a long time ago. I took a look at some of those. And a very long section on masculism, if you’re into that as well. The book is The Artilect War. Professor Hugo de Garis, thank you so much for this. I know these issues are a little dark, but everyone needs to think about them now. Let’s go and maybe prevent the gigadeath, and let’s keep talking about this stuff.

Hugo de Garis: Well, thank you for giving me a forum. I’m very grateful.

Brian Rose: My pleasure. People like hearing you, and I do too. All the best, sir.

Hugo de Garis: Thank you.