Rage Against The Machines: Yuval Noah Harari On Surviving AI, The History Of Information, And The Future Of Humanity

29:57It's to understand that even very primitive AIs, and we were talking about, you know, like eight years ago. Not things like Chagipiti. Still, the decisions made by these algorithms to promote certain content had far-reaching and terrible consequences. In Myanmar, they were not just producing conspiracy theories. They were producing, there are millions of users producing, you know, cooking lessons and biology lessons and sermons on compassion from Buddhist monks and conspiracy theories. And the algorithms made the decision to promote the conspiracy theories. And this is just kind of a warning of look what happens with even very primitive AIs.

30:40And the AIs of today, which are far more sophisticated than in 2016, they too are still just the very early stages of the AI evolutionary process. And we can think about it like the evolution of animals. Until you get to humans, you have 4 billion years of evolution. You start with microorganisms like amoebas, and it took billions of years of evolution to get to dinosaurs and mammals and humans. Now, AIs are at present at the beginning of a parallel process. The Chagipiti and so forth, they are the amoebas of the AI world. But AI evolution is not organic. It's inorganic. It's digital. And it's millions of times faster.

31:27So where it took billions of years to get from amoebas to dinosaurs, it might take just 10 or 20 years to get from the AI amoebas of today to AI T-Rex in 2040 or 2050. Maybe even less. Maybe even less. We're talking about, I don't think our brains are organized properly to really comprehend the accelerated speed at which this is self-learning and iterating and improving upon itself. Like, it's a compounding thing that is astronomical. Meanwhile, trillions of dollars are being spent to build these server farms with these NVIDIA chips. And there's so much power required to keep these things going.

32:08They're talking about nuclear power. I mean, this is like, this is a whole new world. And yet in talking about it, it still feels somewhat like an academic exercise because for myself or somebody who might be watching or listening, their experience with AI comes in the form of chat GPT or some of these helpful tools. Like I like my algorithm. It shows me the kind of products that I want to buy without having to search for it. And a simple example would be preparing for this podcast. Like I listened to your book on audio book and I'm doing what I usually do, pulling up a bunch of tabs and, you know, like just collating a bunch of information on you and the book and the message that you're putting out.

32:50But I did something I had never done before, which is I got a PDF of Nexus and I uploaded it to a tool called Notebook LM. And that tool then synopsized the entire book and created a chatbot where I could ask it questions about your book and ask it to elaborate on certain concepts. And it will even create a podcast conversation between two people about the subject matter of the book. So even this conversation is at risk, right? It's now irrelevant, yeah. And I'm like, wow, that's kind of a remarkably helpful tool. And it's easy to just not really appreciate or connect with the downside risk and power of these tools and where they're leading us.

33:36So I think what I'm saying is, I guess the point I'm trying to make is consumers like all of us, we're being lured into a trust of something so powerful we can't comprehend and are ill-equipped to be able to kind of cast our gaze into the future and imagine where this is leading us. Absolutely. I mean, part of it is that there is enormous positive potential in AI. It's not like it's all doom and gloom. There is really enormous positive potential if you think about the implications for healthcare. that, you know, AI doctors available 24 hours a day that know our entire medical history and have read every medical paper that was ever published and can tailor their advice, their treatment to our specific life history and our blood pressure, our genetics.

34:24It can be the biggest revolution in healthcare ever. If you think about self-driving vehicles. So every year, more than a million people die all over the world in car accidents. Most of them are caused by human error, like people drinking and then driving or falling asleep at the wheel or whatever. Self-driving vehicles are likely to save about a million lives every year. This is amazing. You think about climate change. So yes, developing the AIs will consume a lot of energy, but they could also find new sources of energy, new ways to harness energy that could be our best shot at preventing ecological collapse.

35:01So there is enormous positive potential. We shouldn't deny that. We should be aware of it. And on the other hand, it's very difficult to appreciate the dangers because the dangers, again, they are kind of alien. Like if you think about nuclear energy, yeah, it also had positive potential, nuclear, cheap nuclear energy. But people had a very good grasp of the danger, nuclear war. Anybody can understand the danger of that. With AI, it's much more complex because the danger is not straightforward. The danger is really, I mean, we've seen the Hollywood science fiction scenarios of the big robot rebellion, that one day a big computer or the AI decides to take over the world and kill us or enslave us.

35:44And this is extremely unlikely to happen anytime soon, because the AIs are still a kind of very narrow intelligence. Like the AI that can summarize a book, it doesn't know how to act in the physical world outside. You have AIs that can fold proteins, you have AIs that can play chess, but we don't have this kind of general AI that can just find its way around the world and build a robot army and whatever. So people, it's hard to understand what's so dangerous about something which is so kind of narrow in its abilities. And I would say that the danger doesn't come from the big robot rebellion. It comes from the AI bureaucracies.

36:23Already today, and more and more, we will have not one big AI trying to take over the world. We will have millions and billions of AIs constantly making decisions about us everywhere. You apply to a bank to get a loan, it's an AI deciding whether to give you a loan. You apply to get a job, it's an AI deciding whether to give you a job. You're in court, or you're found guilty of some crime, the AI will decide whether you go for six months or three years or whatever. Even in armies, we already see now in the war in Gaza and with the war in Ukraine, AI make the decision about what to bomb. And in the Hollywood scenario, you have the killer robots shooting people.

37:04In real life, it's the humans pulling the trigger, but the AI is choosing the targets. This is much more complex than the standard scenario.

37:21more than 900 episodes of this podcast, I have noticed a pattern. And that pattern is that the highest performers don't buy into the latest trendy hacks. Instead, they obsess on what actually works, which is always the unassuming basics. And there is nothing more basic than hydration. But here's the kicker. Your body can't hold onto water without the right minerals. Without them, water is just like this temporary visitor. But Element has cracked the code on this, which is why I've been using it religiously for years. Zero sugar, no artificial junk, just sodium, potassium, and magnesium in the ratios that actually work.

38:01And look, I'm not exactly crushing ultras right now, healing from this surgery, but in some ways I need it even more. In order to properly recover, I need to treat my body even better than ever so it can heal properly and expeditiously while also maintaining my focus and my energy levels to rock out all of these podcasts, write a book, be a husband and a dad. And I gotta say, Element keeps my brain firing in a way that water alone can't. Their new sample pack features their most popular flavors. Citrus salt, raspberry salt, watermelon salt, that's my favorite, and orange salt. Eight stick packs total, perfect for finding your favorite or sharing with a friend.

38:44Get a free eight count sample pack of Element's most popular drink mix flavors with any purchase at drinklmnt.com slash richroll. Find your favorite element flavor or share it with a friend. We're brought to you today by the wonderful folks at Go Brewing. Let me tell you a story. A few years ago, this guy, Joe Chura, rings me up out of the blue and he asked if I'll fly out to Illinois and speak at this event that he was hosting called Go, which ended up being this really incredible weekend oriented around taking inspired action. Joe and I hit it off, but, you know, that was kind of that. And it wasn't until I ran into him a couple years ago at Jesse Itzler's Running Man event that I realized that he had taken inspired action himself by creating this new enterprise that was also called Go.

39:32Go Brewing, in fact, which from Go has grown into what it is today, one of the most exciting revolutions in craft brewing. One of the many things that makes Go Brewing extraordinary is that they don't outsource like most companies. They handcraft everything from scratch in small batches. In fact, this commitment to quality has fueled their growth into one of America's fastest growing breweries, now in over 5 ,000 locations across 20 states and available online. The Salty AF Chilada earned the untapped number one non-alcoholic lager in the United States, and they're constantly creating bold new flavors almost every month that push the boundaries of what non-alcoholic beer can be.

40:13Double IPAs, mouth-watering sours, all with zero added sugars and none of the junk. Hear that? Incredible stuff. The non-alcoholic revolution is here, people. I am proud to help champion it alongside Joe. So get on board by getting with Go by going to gobrewing.com, where you're going to use the code RICHROLL for 15 % off your first purchase. Go.

40:45When it comes to health, sleep is a big deal. And there's just so much science out there to back up the role that it plays in every facet of well-being, from heart health to mental health, recovery, cognition, and just being able to show up as your best self. Getting a quality eight hours per night is a personal, non-negotiable that I go to great lengths to ensure. It's sort of a commitment not only to myself, but to my career and to those that I love that all begins with what I'm sleeping on. Now, I've tried many mattresses, but the one that's really won my heart is Birch. And there's many reasons for this, all of which boil down to the simple fact that Birch just does things right.

41:25In addition to being incredibly comfortable and cool, helping you regulate temperatures at night, which are essential for quality rest, Birch mattresses are also firm in all the right places, and all their materials are sustainably sourced, including organic fair trade cotton. And also, everything is hypoallergenic and toxin-free, meaning no harmful off-gassing typical of most mattresses. Birch is so confident in their product, they're offering both a 100-night risk-free trial as well as a 25-year warranty. Plus, they provide free delivery right to your door in a conveniently sized box-up. And so I wanna offer all of my listeners the chance to enjoy a deep and restful night's sleep with a new mattress from Birch.

42:12So right now you can get 27 % off site-wide plus two free EcoRest pillows with mattress purchase. Just go to birchliving.com slash richroll.

42:30Every point of connection with bureaucracy then becomes turned over to an algorithm that makes decisions in a black box without the opportunity for rebuttal or conversation, right? So we're outsourcing all of these decisions and creating like an autocratic diaspora of decision makers. And that in turn, like you can imagine over time, like what emerges from that is like a godhead or a pantheon of gods where there's an authoritarian regime that's dispersed across this in which we are relenting our agency over to these machines and trusting that they're making the right decisions, but not knowing how those decisions are being made.

43:14Even the engineers who are creating the algorithms don't know. And there's something, you know, kind of innately terrifying about that. Again, it's not authoritarian in the sense that there is a single human being that is pulling all the levers. No, it's the AIs. Like the bank has this AI that decides who is qualified to get a loan. And if they tell you, we decided not to give you a loan, and you ask the bank why not, and the bank says, we don't know. I mean, computer says no. I mean, the algorithm says no. We don't understand why the algorithm says no, but we trust the algorithm. and this is likely to spread to more and more places.

43:50The key thing is it's not that the bank is hiding something from you. It's really that the AIs make decisions in a very different way than human beings on a basis of a lot more data. So if the bank really wanted to explain to you why they refuse to give you a loan, like let's say there is a law, the government passes a law of a right to an explanation. If the bank refused to give you a loan, you can apply. They must give you an explanation. So the explanation, well, people fear that it will be kind of, I don't know, racist bias or homophobic bias, like in the old days, that the algorithms saw that you're black or you're Jewish or you're gay, and this is why I refuse to give you a loan.

44:32It won't be like that. I mean, the bank will send you an entire encyclopedia and millions of pages saying, this is why the computer refused to give you a loan. The computer took into account thousands and thousands of data points about you, each one based on statistics on millions of previous cases. And now you can go over these millions of pages if you like. And if you want to challenge, OK. But it's not the kind of old style racism or whatever. Sure. A new version of the terms and conditions that we just click on without reading, right? except extrapolated a hundredfold. In addition to that, with all of these data points, I can't help but think that these machines, the veracity of the information that these machines provide us with is only as reliable as the data sets that it has been provided with.

45:32And right now, we're tiptoeing into a situation where the internet is being rapidly degraded because it's being populated more and more by AI content. Now, when you go to Google and you search, the first thing you see is a, is sort of an AI kind of summary of your query as opposed to links. And this in turn is undermining the business model of legacy media and all forms of media, right? So as those continue to die on the vine, more and more of the internet will be a result of AI generated content. And then it becomes a recursive thing in which it's feeding upon its own inputs to make decisions.

46:15And, you know, with that, like you can imagine a degradation of the data set upon which it is making those decisions. Exactly. Even if you think about something like music. So AI that now creates music, it basically ate the whole of human music. For thousands of years, humans produced music or art or theater or whatever. Within a year, the current AI just ate the whole of it and digested it and start now creating new music or new texts or new images. And the first kind of generation of AI texts or music, this is based on previous human culture. But with each passing year, the AIs will be eating their own products.

47:02Because as you know, the human share in music production or the human share in text production or image production will go lower and lower. Most images, most music will be produced at least in part by AI. And this will be the new food that the AI eats. And then you have exactly what you described, this recursive pattern. And where it will lead us, we have no idea. I mean, another way to think about it, this is the first time that we are basically about to enter a non-human culture. Like, humans are our cultural entities. We live cocooned inside culture. Like, all this music and art and also finance and also religion, this is all part of culture.

47:46And for tens of thousands of years, the only entities that produced culture were other humans. So all the songs you ever heard were produced by humans. All the religious mythologies you ever heard came from the human imagination. Now there is an alien intelligence, a non-human intelligence, that will increasingly produce songs and music, mythology, financial strategies, political ideas. Even before we rush to decide, is it good? Is it bad? Just stop and think about the meaning of living in a non-human culture or a culture which is, I don't know, 40 % or 70 % non-human. It's not like going to China and seeing a different human culture.

48:30It's like really alien culture here on earth. Yeah, my human mind bristles at that. I start thinking about like this bias I have around the originality of human thought and of motion and this kind of assumption that AI will never be able to fully mimic the human experience, right? There's something indelible about what it means to be human that the machines will never be able to fully replicate. And when you talk about, you know, information, the purpose of information being to create connection, a big piece there is intimacy, like intimacy between human beings. So information is meant to create connection, But now we have so much information and we're feeling very disconnected.

49:14So there's something broken in this system. And I think it's driving this loneliness epidemic. But on the other side, it's making us value like intimacy, maybe a little bit more than we were previously. And so I'm curious about where intimacy kind of fits into this, you know, post-human world in which culture is being dictated by machines. I mean, human beings are wired for that kind of intimacy. And I think our radar or our kind of ability to identify it when we see it is part of what makes us human to begin with. Maybe the most important part. I think the key distinction here that is often lost is the distinction between intelligence and consciousness.

49:59That intelligence is the ability to pursue goals and to overcome problems and obstacles on the way to the goal. The goal could be a self-driving vehicle trying to get from here to San Francisco. The goal could be increasing user engagement. And an intelligent agent knows how to overcome the problems on the way to the goal. This is intelligence. And this is something that AI is definitely acquiring. In at least certain fields, AI is now much more intelligent than us. Like in playing chess, much more intelligent than human beings. But consciousness is a different thing than intelligence. Consciousness is the ability to feel things, pain, pleasure, love, hate.

50:47When the AI wins a game of chess, it's not joyful. If there is a tense moment in the game, it's not clear who is going to win. The AI is not tense. It's only the human player which is tense or frightened or anxious. The AI doesn't feel anything. Now, there is a big confusion because in humans and also in other mammals, in other animals, in dogs and pigs and horses and whatever, intelligence and consciousness go together. We solve problems based on our feelings. Our feelings are not something that kind of evolution is decoration. it's the core system through which mammals make decisions and solve problems is based on our feelings so we tend to think that consciousness and intelligence must go together and in all these science fiction movies you see that as the computer or robot becomes more intelligent then at some point it also gains consciousness it falls in love with the human or whatever And we have no reason to think like that.

51:53Yeah, consciousness is not a mere extrapolation of intelligence. Absolutely not. It's a qualitatively different thing. Yeah, and again, if you think in terms of evolution, so yes, the evolution of mammals took a certain path, a certain road in which you develop intelligence based on consciousness. But so far, what we see as computers, they took a different route. their road develops intelligence without consciousness. I mean, computers have been developing, you know, for 60, 70 years now. They are not very intelligent, at least in some fields, and still zero consciousness. Now, this could continue indefinitely.

52:34Maybe they are just on a different path. Maybe eventually they will be far more intelligent than us in everything and still will have zero consciousness, will not feel pain or pleasure or love or hate. You know, the same way that if you think about birds and airplanes. So airplanes did not become like birds. Airplanes don't fly using feathers and so forth. They fly in a completely different way. It's not like that at a certain point when the airplane flies fast enough, suddenly the feathers will appear. No. And it could be the same with intelligence and consciousness, that it will be more and more intelligent without feelings ever appearing.

53:15Now, what adds to the problem is that there is nevertheless a very strong commercial and political incentive to develop AIs that mimic feelings, to develop AIs that can create intimate relations with human beings, that can cause human beings to be emotionally attached to the AIs. Even if the AI have no feelings of themselves, they could be trained, they are already trained to make us feel that they have feelings and to start developing relationships with them. Why is there such an incentive? Because intimacy is, on the one hand, maybe the most cherished thing that the human can have. you know, just on the way here, we were listening to Barbara Streisand singing, people who need people are the luckiest people in the world.

54:14That intimacy is not a liability. It's not something bad that, oh, I need this. No, it's the greatest thing in the world. But it's also potentially the most powerful weapon in the world. If you want to convince somebody to buy a product, if you want to convince somebody to vote for a certain politician or party, Intimacy is like the ultimate weapon. I mean, so far in history, there was a big battle for attention. How to grab human attention. Also, we talked about earlier in social media, how to get human attention. And there were ways, like, I don't know, in Nazi Germany, Hitler could force everybody to listen to his speech on radio.

54:55So he had command of attention, but not of intimacy. There was no technology for Hitler or Stalin or anybody else to mass produce intimacy. Now with AIs, it is possible technically to mass produce intimacy. You can create all these AIs that will interact with us and they will understand our feelings because feelings are also patterns. You can predict a person's feelings by watching them for weeks and months and learning their patterns and facial expression and tone of voice and so forth. And then if it's in the wrong hands, it could be used to manipulate us like never before. Sure. It's our ultimate vulnerability.

55:38This beautiful thing that makes us human becomes this great weakness that we have because as these AIs continue to self-iterate, their capacity to mimic consciousness and human intimacy will reach such a degree of fidelity that it will be indistinguishable to the human brain. And then humans become like these unbelievably easy to hack machines who can be directed wherever the AI chooses to direct them. Yeah, it's not a prophecy. We can take actions today to prevent this. We can have regulations about it. We can, for instance, have a regulation that AIs are welcome to interact with humans, but on condition that they disclose that they are AIs.

56:27If you talk with an AI doctor, that's good, but the AI should not pretend to be a human being. You know, I'm talking with an AI. I mean, it's not that there is no possibility that AI will develop consciousness. We don't know. I mean, there could be that AIs will really develop consciousness. It's mimicking it to such a degree of fidelity. Does it even, in terms of like how human beings interact with it, does it matter? For the human beings, no. I mean, this is the problem. I mean, because we don't know if they really have consciousness or they're only very, very good at mimicking consciousness.

57:01So the key question is ultimately political and ethical. If they have consciousness, if they can feel pain and pleasure and love and hate, this means that they are ethical and political subjects. They have rights that you should not inflict pain. on an AI the same way you should not inflict pain on a human being, that what they like, what they love might be as important as what human beings desire. So they should also vote in elections and they could be the majority because you can have a country, 100 million humans and 500 million AIs. So do they choose the government in this situation? Now, you know, Now, in the United States, interestingly enough, there is actually an open legal path for AIs to gain rights.

57:50It's one of the only countries in the world where this is the case. Because in the United States, corporations are recognized as legal persons with rights. Until today, this was a kind of legal fiction. Like according to U.S. law, Google is a person. It's not just a corporate, it's a person. And as a person, it also has freedom of speech. This is the Supreme Court ruling for 2010 of Citizen United. Now, until today, this was just legal fiction because every decision made by Google was actually made by some human being, an executive, a lawyer, an accountant. Google could not make a decision independent of the humans.

58:29But now you have AIs. So imagine the situation when you incorporate an AI. Now this AI is a corporation. and as a corporation, US law recognizes it as a person with certain rights like freedom of speech. Now, it can earn money. It can go online, for instance, and offer its services to people and earn money. Then it can open a bank account and invest its money in the stock exchange. And if it's very smart and very intelligent, it could become the richest person in the US. Now, imagine the richest person in the U.S. is not a human. It's an AI. And according to U.S. law, one of the rights of this person is to make political contributions, donations.

59:14This was the main reason behind Citizen United in 2010. So this AI now makes billions of dollars of contributions to politicians in exchange for expanding AI rights. And the legal path in the U.S. is completely open. You don't need any new law to make this happen. That's like a plot of a movie. Yeah, I mean, you know, we're in L.A. Yeah, I mean, wow, that's so wild to contemplate. What are the differences in the ways in which the advent of this powerful technology is impacting democratic systems and authoritarian systems? So both systems have a lot to gain and have a lot to lose. Again, the AI, it's the most powerful technology ever created.

1:00:06It's not a tool, it's an agent. So you have millions and billions of new agents, are very intelligent, very capable, that can be used to create the best healthcare system in the world, but also the most lethal army in the world, or the worst secret police in the world. If you think about authoritarian regimes, So throughout history, they always wanted to monitor their citizens around the clock. But this was technically impossible. Even in the Soviet Union, you know, you have 200 million Soviet citizens. You can't follow them all the time because the KGB didn't have 200 million agents. And even if the KGB somehow got 200 million agents, that's not enough.

1:00:49Because, you know, in the Soviet Union, it's still basically paper bureaucracy. the secret police, if a secret agent followed you around 24 hours a day, at the end of the day, they write a paper report about you and send it to KGB headquarters in Moscow. So imagine every day KGB headquarters is flooded with 200 million paper reports. Now, to be useful for anything, somebody needs to read and analyze them. They can't do it. They don't have the analysts. Therefore, even in the Soviet Union, some level of privacy was still the default for most people. For technical reasons. Now, for the first time in history, it is technically possible to annihilate privacy.

1:01:36A totalitarian regime today doesn't need millions of human agents if he wants to follow everybody around. You have the smartphones and cameras and drones and microphones everywhere. and you don't need millions of human analysts to analyze this ocean of information, you have AI. And this is already beginning to happen. This is not a future prediction. In many places around the world, you begin to see the formation of this totalitarian surveillance regime. It's happening in my country, in Israel. Israel is building this kind of surveillance regime in the occupied Palestinian territories to follow everybody around all the time.

1:02:15And also in our region in Iran, since the Islamic revolution in 1979, they had the hijab laws, which says that every woman, when she goes out walking or even driving in her private car, she must wear the hijab, the headscarf. And until today, the regime had difficulty enforcing the hijab laws because they didn't have, you know, millions of police officers that you can place on every street, a police officer, if a woman drives without a headscarf, immediately she's arrested and fine or whatever. In the last few years, they switched to relying on an AI system. Iran is now crisscrossed by surveillance cameras with facial recognition software, which recognizes automatically if in the car that just passed by the camera, the facial recognition software can identify that this is a woman, not a man, and she's not wearing the hijab, and identify her identity, find her phone number, and within half a second, they send her an SMS message saying, you broke the hijab law, your car is impounded, your car is confiscated, stop the car by the side of the world.

1:03:34This is daily occurrence today in Tehran and Isfahan and other parts of Iran, and this is based on AI. And it's not like there is a report that goes to the court and some human judge goes over the data and decides what to do. The AI like immediately decides, okay, the car is confiscated. And this can happen in more and more places around the world, like even in the US. You know, if you think about all the debate about abortion. Without going into the debate itself, the people who think, rightly or wrongly, but they think that abortion is murder, they have a very strong incentive to build a similar surveillance surveillance.

1:04:18system for American women, you know, to stop murder. Like you can build this surveillance system that can identify yesterday you're pregnant, today or not, what happened in between. So it's not just a problem, you know, for Iran or for the Palestinians or the Chinese. This can come to the U.S. as well. And to prevent them from crossing state lines, things like that. Yeah. Yeah. Like, OK, you went from, I don't know, Texas to California. You were pregnant. You came back. You're not pregnant. What happened in California? So it feels like AI is this incredible tool to consolidate power around authoritarian regimes, but it also has its pitfalls, too.

1:05:02It's not the perfect tool. It also frightens the autocrats because the one thing that human dictators always feared most was not a democratic revolution. The one thing they feared most is a powerful subordinate that they can't control and that might manipulate them or take power from them. If you can look at the Roman Empire, not a single Roman emperor was ever toppled by a democratic revolution. Never happened. But many of them lost their life or their power to subordinate, you know, a general that rebelled against them, a provisional governor, their brother, their wife that took power from them.

1:05:44This is the greatest fear of every dictator also today. And so if you think about AI, so if you're a human dictator and you now give this immense power to an AI system, where is the guarantee that this system will not turn against you and either eliminate you or just turn you into a puppet? I mean, what we also know about dictators, it's relatively easy to manipulate these people if you can whisper in their ear. Right. Because they are very paranoid. And the easiest people to manipulate are the paranoid people. And we have our AI corporation in the United States that can deploy billions of dollars towards bots and whatever else to, you know, create that paranoia or enhance it.

1:06:30You really just need to hack one person, you know, to for an AI to take power in the US, very complicated. It's such a distributed system. Like, okay, the AI can learn to manipulate the president, but it also needs to manipulate the senators and the Congress members and the state governors and the Supreme Court. Like, what would the AI do with the Senate filibuster? It's difficult. But if you want to take power in a dictatorship, you just need to learn to manipulate a single person. So the dictators are not all happy about the AIs. and we're already beginning to see it, for instance, with chatbots, that they are very concerned because, you know, you can design a chatbot which will be completely loyal to the regime.

1:07:17But once you release it to the internet to start interacting with people in real life, it changes. I mean, remember what we talked earlier, that AI is defined by the ability to learn and change by itself. So even if Putin creates like the Putinist chatbot that always says that Putin is great and Putin is right and Russia is great and so forth, but then you release it to the real world, it starts observing things in the real world. For instance, it notices that, you know, in Russia, the invasion of Ukraine is officially not a war. It's called a special military operation. And if you say that it's a war, you go to prison for up to, I think, three years or something like that, because it's not a war.

1:08:03It's a special military operation. Now, what do you do if a very intelligent chatbot that you released, you know, connects the dots and says, no, it's not a special military operation. It's a war. Would you send a chatbot to prison? What can you do? And, you know, democracies, of course, also have a problem with chatbot saying things we don't like. They can be racist. They can be homophobic, whatever. But the thing about democracy, it has a relatively wide margin of tolerance, even for anti-democratic speech. Dictatorships have zero margin for dissenting views. so they have a much bigger problem with how to control these unpredictable chatbots.

1:08:58I own a bunch of spectacles and I made the grave error the other day of donning a normal non-Roka pair on my indoor trainer when I was riding my bike indoors and I gotta tell you it as a disaster. Every three to five seconds, I had to take my hands off the handlebars and push my glasses back up my nose until I got so frustrated I just tossed them aside. This is the dilemma of every active but optically impaired person I know. And as someone who has relied upon eyewear every single day since I was five years old, it is also the source of endless aggravation. Thankfully, now eradicated thanks to Broca, the stylish performance eyewear company founded by two former Stanford swimming teammates of mine who have gifted everyone like me and quite frankly, the world with their fashionable line of super lightweight prescription glasses and sunglasses with patented no-slip nose and temple pads that are just impervious to sweat.

1:10:02And no matter what you do, remain locked on your mug no matter how intense your workout. Without the dork factor, these things go everywhere with me from the trail to the dinner party. Put them on, feel the difference, and wear without limits. Unlock 20 % off your order with the code RICHROLL at ROKA.com. That's R-O-K-A dot com.

1:10:32How are you interpreting the current moment, given that we're on the cusp of an election here in the United States? And, you know, there's a lot of discourse around the existential threat to democracy that we may be facing. What role is AI playing in this? What should we understand about the impact of this technology on us as citizens and voters? At present, I don't think that AI has, again, social media has, of course, a huge impact on the political discourse and thereby on the results of the elections. But I don't see AI really kind of changing or manipulating the elections in November. It's too close.

1:11:16The big question is, whoever wins the elections, maybe the most important decisions that person has to make will be about AI. Because of the extremely rapid pace that this technology is developing. You know, you look at where ChatGPT was a year ago. You look at what things are now in 2024. What will be the state of AI in 2027, 2028? So, you know, I watched the presidential debate. Most people, their main takeaway was about the cats and the dogs. It's the most memorable thing from the debate. I mean, you know, whoever wins, maybe will have to make some of the most important decisions in history about the relations.

1:12:02If you're worried about immigration, it's not the immigrants that will, you know, replace the taxi drivers. It's the immigrants that will replace the bankers that you should be worried about. And it's the AIs, not somebody coming from south of the border. And who do you trust to make these momentous decisions? Now, and if you think specifically about the threats to democracy, so one thing we learned from history is that democracies always, since, again, ancient Athens, they always had this one single big problem or weakness. that democracy is basically a kind of a deal that you give power to somebody for a limited time period, for four years, on condition they give it back.

1:12:50And then you can make a different choice. Like we tried this, it didn't work, let's try something else. This ability to say, let's try something else, this is democracy. And it's based on that you give power and you expect to get it back after four years. What happens if you give power to somebody who then doesn't give it back? They now have the power. They have the power to also stay in power. That was always the biggest danger in democracy. So for me, the issue in the U.S. elections, you can discuss the economic policies, the foreign policies. You like this, you like that. There is discussion to be had.

1:13:31But you have your one person, Donald Trump, and that has, you know, you have a record. For the previous time, this person doesn't want to give power back and is willing to go a long way, including potentially inciting violence, to avoid giving power back. And you want to give him so much power? That doesn't sound like a very good idea. So for me, this is kind of the number one issue in the elections. Everything else is of marginal importance in comparison. Yeah. I mean, I think it challenges our predilections around the stability of democracy and is forcing us to really embrace the fact that it is a delicate dynamic that is informed by collective action by the people.

1:14:21And in reflecting upon this technology also, the story of technology is one in which our ability to legislate around it and regulate it always falls way behind the pace of advancement. And now we're in a situation where the pace of advancement is like nothing we've ever seen before, which calls into question our ability to not only, you know, kind of put guardrails around it, but to even understand what is actually happening. The history of information systems is one of collective human cooperation. And yet we're in a situation right now where it feels like cooperation is being challenged, not only nationally here in the United States, but internationally.

1:15:10And so as we kind of begin to talk about how we're going to triage this or find solutions, like, where do you land in terms of our capacity to collectively come together as a global community to figure out solutions and then put them into motion so that we don't tiptoe into some kind of dystopia? So there is a lot to unpack here. So first of all, when we think about cooperation, as we said earlier, this was always our biggest advantage as a species, that we cooperate better than anybody else. We can construct these even global networks of trade that no other animal even understands. Like if you think about, I don't know, horses.

1:15:54So horses never figured out money. They were bought and sold, but they never understood what are these things that the humans are exchanging. And this is why horses could never unite against us or could never manipulate us because they never figured out how the system works. That one person is giving me to another person in exchange for a few shiny metal things or some pieces of paper. AI is different. It understands money better than most people. Like most people don't understand how the financial system really works. And financial AIs in fintech, they already surpass most human beings, not all human beings, but most human beings in their understanding of money.

1:16:39So we are now confronting, again, millions and billions of new agents that potentially can use our own systems against us. that computers can now collaborate using, for instance, the financial system more efficiently than humans can. So the whole issue of cooperation is changing. And computers also learn how to use the communication systems to manipulate us, like in social media. So they are cooperating where we are losing the ability to cooperate. And that should raise the alarm. Now, and the thing is it's very difficult to understand what is happening. If we want humans around the world to cooperate on this, to build guardrails, to regulate the development of AI, first of all, you need humans to understand what is happening.

1:17:32Secondly, you need the humans to trust each other. And most people around the world are still not aware of what is happening on the AI front. you have a very small number of people in just a few countries, mostly the US and China and a few others, who understand. Most people in Brazil, in Nigeria, in India, they don't understand. And this is very dangerous because it means that a few people, many of them are not even elected by the US citizen. They are just, you know, private companies. They will make the most important decisions. And the even bigger problem is that even if people start to understand, they don't trust each other.

1:18:15Like I had the opportunity to talk to some of the people who are leading the AI revolution, which is still led by humans. It is still humans in charge. I don't know for how many more years, but as of 2024, it's still humans in charge. And you meet with these, you know, entrepreneurs and business tycoons and politicians also in the US, in China, in Europe, and they all tell you the same thing, basically. They all say, we know that this thing is very, very dangerous, but we can't trust the other humans. If we slow down, how do we know that our competitors will also slow down? Whether our business competitors, let's say here in the U.S.

1:19:01or our Chinese competitors across the ocean. And you go and talk with the competitors, they say the same thing. We know it's dangerous. We would like to slow down to give us more time to understand, to assess the dangers, to debate regulations, but we can't. We have to rush even faster because we can't trust the other corporation, the other country. And if they get it before we get it, it will be a disaster. And so you have this kind of paradoxical situation where the humans can't trust each other, but they think they can trust the AIs. Because when you talk with the same people and you tell them, okay, I understand you can't trust the Chinese or you can't trust open AI.

1:19:45So you need to move faster developing this super AI. How do you know you could trust the AI. And then they tell you, oh, I think that will be okay. I think we've figured out how to make sure that the AI will be trustworthy and under our control. So we have this very paradoxical situation when we can't trust our fellow humans, but we think we can trust. And later on top of that is an incentive structure, of course, that further engenders distrust in this arms race, right? Like the prize goes to the breakthrough developers and those will be rewarded and remunerated in ways that are, you know, perhaps unprecedented, right?

1:20:26So the breakthroughs and what's on the other side of that is so enticing that any discourse around regulation or anything else that might slow it down becomes not only a national security threat, but also an entrepreneurial threat, right? So everything is motivating rapid acceleration at the cost of transparency and regulation and all these other things, all these checks and balances that we really need right now. And I don't know how you're feeling about this, but it leaves me a little cold and pessimistic. You're a historian. The story of humankind is all gas, no brakes. You know, like, let's just, we're plowing forward and we'll deal with the consequences when they come.

1:21:14Like, we're not wired adequately to really appreciate the long-term consequences of our behavior. We're kind of, you know, looking right in front of us and making decisions based on how it's going to impact us in the immediate future and very little else. Yeah, I mean, throughout history, the problem is people are very good at solving problems, but they tend to solve the wrong problems. Like they spend very little time deciding what problem we need to solve. Like 5 % of the effort goes on choosing the problem. Then 95 % of the effort goes in solving the problem we focus on. And then we realize, oh, we actually solved the wrong problem.

1:21:54And it just creates new problems down the road that we now need to. And then we do it the same again. And, you know, wisdom often comes from silence, from taking time, from slowing down. Let's really understand the situation before we rush to make a decision. And, you know, it starts on the individual level that so many people, for instance, think, oh, my main problem is in life is that I don't have enough money. And then they spend the next 50 years making lots of money. And even if they succeed, they wake up at a certain point and said, oops, I think I chose the wrong problem. I think it wasn't, yeah, I need some money, but it wasn't my main problem in life.

1:22:37And we are perhaps doing it collectively as a species, the same thing. You know, you go back to something like the agriculture revolution. So people thought, okay, we don't have enough food. Let's produce more food with agriculture. We'll domesticate wheat and rice and potatoes. We'll have lots more food. Life will be great. And then they domesticate these plants and also some animals, cows, chickens, pigs, whatever. And they have lots of food and they start building these huge agricultural societies with towns and cities. And then they discover a lot of new problems they did not anticipate. For instance, epidemics.

1:23:16Hunter-gatherers did not suffer almost any infectious diseases. Because most infectious diseases came to humans from domesticated animals. and they spread in the dense towns and cities. Now, if you live in a hunter-gatherer band, you don't hold any chickens or pigs, so it's very unlikely some virus will jump from a wild chicken to you. And even if you got some new virus, you have just like 20 other people in your band, and you move around all the time. Maybe you infect five others, and like three die, and that's the end of it. But once you have these big agricultural cities, then you get the epidemic.

1:23:55people thought they were building paradise for humans. Turned out they were building paradise for germs. And human life expectancy and human living conditions for most humans actually goes down. If you're a king or a high priest, it's okay. But for the average person, it was actually a bad move. And the same thing happens again and again throughout history. And it can happen now on a very, very big scale with AI. In a way, it goes back to this issue of organic and inorganic, that organic systems are slow. They need time. And this AI is an inorganic system which accelerates beyond anything we can deal with.

1:24:37And the big question is whether we will force it to slow down or it will force us to speed up until the moment we collapse and die. I mean, if you force an organic entity to be on all the time and to move faster and faster and faster, eventually it collapses and dies. One of the things I heard you say that really struck me was this. It's a quote. If something ultimately destroys us, it will be our own delusions. So can you elaborate on that a little bit and how that applies to what we've been talking about? Yeah, I mean, the AIs, at least of the present day, they cannot escape our control and they cannot destroy us unless we allow them or unless we kind of order them to do that.

1:25:27We are still in control. But because of our, you know, political and mythological delusions, we cannot trust the other humans. and we think we need to develop these AIs and faster and faster and give them more and more power because we have to compete with the other humans. And this is the thing that could really destroy us. And, you know, it's very unfortunate because we do have a track record of actually being quite successful of building trust between humans. It just takes time. I mean, if you think about, again, the long arc of human history, so these hunter-gatherer bands tens of thousands of years ago, they were tiny, a couple of dozen individuals.

1:26:13And even though the next steps like agriculture, they had their downside, again, like epidemics, people did learn over time how to build much larger societies which are based on trust. If you now live in the United States or in some other country, You're part of a system of hundreds of millions of people who trust each other in many ways which were really unimaginable in the Stone Age. Like, you don't know 99.99 % of the other people in the country. And still, you trust them with so much. I mean, the food you eat, mostly, you did not go to the forest to hunt and gather it by yourself. You rely on strangers to provide the food for you.

1:27:04Most of the tool you use are coming from strangers. Your security. You rely on police officers, on soldiers that you never met in your life. They are not your cousins. They are not your next door neighbors. And still they protect your life. So yes, if you now go to the global level, okay, we still don't know how to trust the Chinese. And the Israelis still don't know how to trust the Iranians and vice versa. But it's not like we are stuck while we were in the Stone Age. We've made immense progress in building human trust and we are rushing to throw it all away because it just, again, it takes time.

1:27:43It will not happen tomorrow. Yeah. I mean, I think it's urgent that we find a way back to repairing some institutional trust, right? Like that has been degraded in recent times. And I think without that, we stand very little chance as a democratic republic of surviving and solving these kinds of problems. Absolutely. If you ask in brief, what is the key to building trust between millions of strangers, the key is institutions. Because you can't build a personal intimate relationship with millions of people. So it's only institutions, whether it's courts or police forces or newspapers or universities or healthcare systems that build trust between people.

1:28:31And unfortunately, we now see this, again, another epidemic of distrust in institutions on both the right and the left. It is fueled by a very cynical worldview, which basically says that the only reality is power, and humans only want power, and all human interactions are power struggles. So whenever somebody tells you something, you need to ask whose privileges are being served, whose interests are being advanced. And any institution is just an elite conspiracy to take power from us. So journalists, they are not really interested in knowing the truth about anything. They just want power. And the same for the scientists and the same for the judges.

1:29:17And if this goes on, then all trust in institutions collapses and then society collapses. And the only thing that can still function in that situation is a dictatorship. Because dictatorships don't need trust. They are based on terror. So people who attack institutions, they often think, oh, we are liberating the people from these authoritarian institutions. They are actually paving the way for a dictatorship. And the thing is that this view is not just very cynical. It's also wrong. Humans are not these power crazy demons. All of us want power to some extent. That's true, but that's not the whole truth about us.

1:29:58Humans are really interested in knowing the truth about ourselves, about our lives, about the world on a very deep level because you can never be happy if you don't know the truth about your life. Because you will not know what are the sources of misery. Again, you will focus on your life. If you don't know the truth, you waste all your life trying to solve the wrong problems. And this is true also of journalists and judges and scientists. Yes, there is corruption in every institution. This is why we need a lot of institutions to keep one another in check. But if you destroy all trust in institutions, what you get is either anarchy or a dictatorship.

1:30:44And again, it's a good exercise every now and then to stop and think about how every day we are protected by all kinds of institutions. Like when people talk with me about the deep state, you know, this conspiracy about the deep state, I immediately think about the sewage system. The sewage system is the deep state. It's a deep system of tunnels and pipes and pumps, which is a state built, under our houses and streets and neighborhoods, and saves our life every day because it keeps our sewage separate from our drinking water. You know, you go to the toilet, you do your thing, it goes down into the deep state, which keeps it separate from the drinking water.

1:31:31If I can tell one historical anecdote, Where did it come from? So, you know, after an agricultural revolution, you have big cities. They are paradise for germs, hotbeds for epidemics. This continues really until the 19th century. London in the 19th century was the biggest city in the world and one of the most dirty and polluted and a hotbed for epidemics. And in the middle of the 19th century, there is a cholera epidemic and people in London are dying from cholera. And then you have this bureaucrat, medical bureaucrat, Jon Snow, not the guy from Game of Thrones, a real Jon Snow, who did not fight dragons and zombies, but actually did save millions of lives.

1:32:12Because he went around London with lists and he interviewed all the people who got sick or died. If somebody died from Colorado, he would interview their family. Tell me, where did this person get their drinking water from? and he made these long lists of hundreds and thousands of people. And by analyzing these lists, he pinpointed a certain well on Broad Street in Soho in London, where everybody, almost everybody who got sick on cholera, they had a zip of water from that well at a certain stage. And he convinces the municipality to disable the pump of the well and the epidemic stops. And then they investigate, they discover that the well was dug about a meter away from a cesspit.

1:32:59And water, sewage water from the cesspit got into the drinking water. And today, if you want to dig a well or a cesspit in London or in Los Angeles, you have to fill so many forms and to get all these bureaucratic permits and it saves our lives. And how does that relate to this idea of the deep state? I'm trying to tether those two notions together. Again, the people who believe the conspiracy theories about the deep state, they say that all these state bureaucracies, they are elite conspiracies against the common people trying to take over power, trying to destroy us. And in most cases, no, the people in these, you know, to manage a sewage system, you need plumbers, you also need bureaucrats.

1:33:44Again, you need to apply for a license to dig a well. And it is managed by all these kind of state bureaucrats. And it's a very good thing because, again, there is corruption in these places sometimes. This is why we keep also courts. You can go to court. This is why we keep newspapers so they can expose corruption in the cities, in the municipalities, sewage department. But most of the time, most of these people are honest people who are working very hard every day to keep our sewage separate from our drinking water and to keep us alive. And by extrapolation, there are all of these bureaucracies that are working in our interest in invisible ways that we take for granted.

1:34:27Exactly. Basically, right. You've often said clarity is power. And I think your superpower is your ability to kind of stand at 10 ,000 feet and look down on humanity and the planet and identify what's most important in these macro trends. that help us make sense of what's happening now. And I'd like to kind of end this with some thoughts on how you cultivate that clarity through meditation and your very kind of like profound practice of mindfulness and information deprivation, I should say, right? Yeah, information fast. Yeah. Starting maybe with the idea of an information fast. So I think this is important today for every person to go on an information diet.

1:35:17That this idea that more information is always good for us is like thinking that more food is always good for us. It's not true. And the same way that the world is full of junk food that we'd better avoid, the world is also full of junk information that we have better avoid. information which is artificially filled with greed and hate and fear, information is the food of the mind. And we should be as mindful as what we put into our minds as of what we put into our mouths. But it's not just about limiting consumption. It's also about digesting. It's also about detoxifying. Like, we go throughout our life and we take in a lot of junk, whether we like it or not.

1:36:03that fills our mind. And I meditate two hours every day, so I can tell you there is a lot of junk in there, a lot of hate and fear and greed that I picked up over the years. And it's important to take time to simply digest the information and to also detoxify, to kind of let go of all this hatred and anger and fear and greed, which is in our minds. So I began, when I was doing my PhD in Oxford, a friend recommended that I go on a meditation retreat or a Vipassana meditation. And for a year, he kind of nagged me to go on a meditation. And I said, no, this is kind of mystical mumbo-jumbo. I don't want to...

1:36:48And eventually I went. And it was amazing because it was the most remote thing for mysticism that I could imagine. Because it was a 10 days retreat, and on the very first evening of the retreat, the teacher, Esen Goenka, the only instruction he gave, he didn't tell me to kind of visualize some goddess or do this man, nothing. He just said, what is really happening right now? Bring your attention to your nostrils, to your nose, and just feel whether the breath is going in or whether the breath is going out. That's the only exercise. Like a pure observation of reality. What amazed me was my inability to do it.

1:37:33Like I would bring my attention to the nose and try to feel, is it going in, is it going out? And after about five seconds, some thought, some memory, some fantasy would arise in the mind and would just hijack my attention. and for the next two or three minutes, I would be rolling in this fantasy or memory until I realized, hey, I actually need to observe my breath. And I would come back to the breath, again, five seconds, maybe 10 seconds, I will be able to, oh, now it's coming in, it's coming in, oh, now it's going out, it's going out. And again, some memory would come and hijack me. And I realized first that I know almost nothing about my mind.

1:38:10I have no control of my mind, and my mind is just like this factory that constantly produces fantasies and illusions and delusions that come between me and reality. Like if I can't observe the breath going in and out of my nostrils because some fantasy comes up, what hope do I have of understanding AI or understanding the conflict in the Middle least without some mind-made illusion or fantasy coming between me and reality. And for the last 24 years, I have this daily exercise of I devote two hours every day to just what is really happening right now. I sit with closed eyes and just try and focus, let go of all the mind-made stories and feel what is happening to the breath, what is happening to my body, the reality of the present moment.

1:39:08I also go for a long meditation retreat usually every year of between 30 days and 60 days of meditation. Because again, one of the things you realize there is so much noise in the mind that just to calm it down to the level that you can really start meditating seriously, it takes three or four days of continuous meditation. Just so much noise. So long retreats, they enable to have this really deep observation of reality, which is impossible. Most of life we spend like detached from reality. Two hours a day. That's a commitment. Even in the midst of all the book promotion craziness. Yeah. Before I came here, I usually do one in the morning, one in the afternoon or evening.

1:39:57What a beautiful thing. And obviously your ability to think clearly and write so articulately about these ideas is very much a product of this practice. Absolutely. I mean, without the practice, I would not be able to write such books and I would not be able to deal with the kind of all the publicity and all the interviews and, you know, this roller coaster of positive and negative feedback from the world all the time. I would say one important thing, this is not necessarily for everybody. Because I meditate and I have meditator friends and so forth, I mean, different things work for different people.

1:40:36There are many people that I wouldn't recommend to meditate two hours a day or to go for a 10 days meditation retreat because they are different. Their body, their minds are different. For them, perhaps going on a 10 days hike in the mountains would be better. For them, perhaps devoting two hours a day to music, to say playing or to creating or going to psychotherapy would have better results. Humans are really different in many ways from one another. There is no one size fits all. So if you never try meditation, absolutely try it out and give it a real chance. It's not like you go for like a few hours and it doesn't work.

1:41:16Okay, I give it up. Like give it a real chance, but keep in mind that again, different minds are different. So find out what really works for you. And whatever it is, that's the important part, whatever it is, invest in it. I have to release you back to your life, but maybe we can end this with just a concise thought about what it is that you want people to take away from this book. Like what is most vital and crucial for people to understand about what you're trying to communicate? But information isn't truth. Truth is a costly, a rare and precious thing. It is the foundation of knowledge and wisdom and of benign beneficial societies.

1:42:02You can build terrible societies without the truth. But if you want to build a good society and you want to build a good personal life, you must have a strong basis in the truth. And it's difficult, again, because most information is not the truth. and invest in it. It's worthwhile to have a practice, whatever it is, that gets you connected with reality, that gets you connected with the truth. Thank you for coming here today. I really appreciate you taking the time to share your wisdom and experience. I think Nexus, your latest book, is, as I said at the outset, a crucial, vital book that everybody should read.

1:42:43We're entering into a very interesting time and we are well advised to be as best prepared as we possibly can. And I appreciate the work that you do. And thank you again, Yuval. Thank you. I only graced the surface of the outline that I created. So hopefully you can come back because I got a million more questions. I could have talked to you for hours. Next time I'm in LA, I'll be happy to. Thanks, man. Appreciate it. Cheers. Peace.

1:43:16Thank you.

1:43:46change in the Plant Power way, as well as the Plant Power Meal Planner at meals.richroll.com. If you'd like to support the podcast, the easiest and most impactful thing you can do is to subscribe to the show on Apple Podcasts, on Spotify, and on YouTube and leave a review and or comment. This show just wouldn't be possible without the help of our amazing sponsors who keep this podcast running wild and free. To check out all their amazing offers, head to richroll.com slash sponsors. And sharing the show or your favorite episode with friends or on social media is of course awesome and very helpful.

1:44:25And finally, for podcast updates, special offers on books, the meal planner and other subjects, please subscribe to our newsletter, which you can find on the footer of any page at richroll.com. Today's show was produced and engineered by Jason Camiolo. The video edition of the podcast was created by Blake Curtis with assistance by our creative director, Dan Drake. Portraits by Davey Greenberg, graphic and social media assets courtesy of Daniel Solis. And thank you, Georgia Whaley, for copywriting and website management. And of course, our theme music was created by Tyler Pyatt, Trapper Pyatt, and Harry Mathis.

1:45:01Appreciate the love, love the support. See you back here soon. Peace. Plants. Namaste.

1:45:13Thank you.