Bret Weinstein: Truth, Science, and Censorship in the Time of a Pandemic | Lex Fridman Podcast #194

Lex Fridman 0:00 The following is a conversation with Bret Weinstein, evolutionary biologist, author, co-host of the Dark Horse Podcast, and as he says, reluctant radical. Even though we've never met or spoken before this, we both felt like we've been friends for a long time. I don't agree on everything with Bret, but I'm sure as hell happy he exists in this weird and wonderful world of ours. Quick mention of our sponsors, Jordan Harbinger Show, ExpressVPN, Magic Spoon, and Four Sigmatic. Check them out in the description to support this podcast. As a side note, let me say a few words about COVID-19 and about science broadly. I think science is beautiful and powerful. It is the striving of the human mind to understand and to solve the problems of the world. But as an institution, it is susceptible to the flaws of human nature, to fear, to greed, power, and ego. 2020 is a story of all of these that has both scientific triumph and tragedy. We needed great leaders and we didn't get them. What we needed is leaders who communicate in an honest, transparent, and authentic way about the uncertainty of what we know and the large-scale scientific efforts to reduce that uncertainty and to develop solutions. I believe there are several candidates for solutions that could have all saved hundreds of billions of dollars and lessened or eliminated the suffering of millions of people. Let me mention five of the categories of solutions: masks, at-home testing, anonymized contact tracing, antiviral drugs, and vaccines. Within each of these categories, institutional leaders should have constantly asked and answered publicly, honestly, the following three questions. One, what data do we have on the solution and what studies are we running to get more and better data? Two, given the current data and uncertainty, how effective and how safe is the solution? Three, what is the timeline and cost involved with mass manufacture and distribution of the solution? In the service of these questions, no voices should have been silenced, no ideas left off the table. Open data, open science, open, honest scientific communication and debate was the way, not censorship. There are a lot of ideas out there that are bad, wrong, dangerous. But the moment we have the hubris to say we know which ideas those are is the moment we lose our ability to find the truth, to find solutions, the very things that make science beautiful and powerful in the face of all the dangers that threaten the well-being and the existence of humans on Earth. This conversation with Bret is less about the ideas we talk about. We agree on some, disagree on others. It is much more about the very freedom to talk, to think, to share ideas. This freedom is our only hope. Bret should never have been censored. I asked Bret to do this podcast to show solidarity and to show that I have hope for science and for humanity. This is the Lex Fridman Podcast, and here's my conversation with Bret Weinstein.

Lex Fridman 3:14 What to you is beautiful about the study of biology? The science, the engineering, the philosophy of it.

Bret Weinstein 3:21 It's a very interesting question. I must say, at one level, it's not a conscious thing. I can say a lot about why as an adult I find biology compelling, but as a kid, I was completely fascinated with animals. I loved to watch them and think about why they did what they did, and that developed into a very conscious passion as an adult. But I think, uh, in the same way that one is drawn to a person, I was drawn to the never-ending series of near-miracles that exist across biological nature.

Lex Fridman 4:02 When you see a living organism, do you see it from an evolutionary biology perspective of, like, this entire thing that moves around in this world? Or do you see, like, from an engineering perspective that are, like, uh, first principles almost down to the physics, like the little components that build up hierarchies, that you have cells, uh, the first proteins and cells and, uh, organs and all that kind of stuff? So do you see low level or do you see high level?

Bret Weinstein 4:30 Well, the human mind is a strange thing, and I think it's probably a bit like a time-sharing machine in which I have different modules. We don't know enough about biology for them to connect, right? So they exist in isolation, and I'm always aware that they do connect. But I basically have to step into a module in order to see the evolutionary dynamics of the creature and the lineage that it belongs to. I have to step into a different module to think of that lineage over a very long time scale, a different module still to understand what the mechanisms inside would have to look like to account for what we can see from the outside. And I think that probably sounds really complicated, but one of the things about being involved in a topic like biology and doing so for one, you know, really not even just my adult life, for my whole life, is that it becomes second nature. And, you know, when, when we see somebody do an amazing parkour routine or something like that, we think about what they must be doing in order to accomplish that. But of course, what they are doing is tapping into some kind of zone, right? They are in a zone in which they are in such, um, command of their center of gravity, for example, that they know how to hurl it around a landscape so that they always land on their feet. Um, and I would just say...... for anyone who hasn't found a topic on which they can develop that kind of facility, it is absolutely worthwhile. It's really something that human beings are capable of doing across a wide range of topics, many things our ancestors didn't even have access to. And that flexibility of humans, that ability to repurpose our machinery for topics that are novel means really, the world is your oyster. You can, you can figure out what your passion is and then figure out all of the angles that one would have to pursue to really deeply understand it, and it is, uh, it is well worth having at least one topic like that.

Lex Fridman 6:42 You mean embracing the full adaptability of the... both the- the body and the mind. So like, 'cause, 'cause I don't know what to attribute the parkour to. Like, biomechanics of how our bodies can move? Or is it the mind? Like, how much percent-wise? Is it the entirety of the hierarchies of biology that we've been talking about, or is it just all the mind?

Bret Weinstein 7:09 The way to think about creatures is that every creature is two things simultaneously. A creature is a machine of sorts, right? It's not a machine in, in the, you know, it's- it's a... I call it an aqueous machine, right? And it's run by an aqueous computer, right? So, it's not identical to our- our technological machines. But every creature is both a machine that does things in the world sufficient to accumulate enough resources to continue surviving, to reproduce. It is also a potential, so each creature is potentially, for example, the most recent common ancestor of some future clade of creatures that will look very different from it. And if a creature is very, very good at being a creature, but not very good in terms of the potential it has going forward, then that lineage will not last very long into the future because change will, uh, throw it challenges that it- its descendants will not be able to meet. So, the thing about humans is we are a generalist platform and we have the ability to swap out our software to exist in many, many different niches. And I was once watching a, uh, an interview with- with this, uh, British group of parkour experts who were being, you know, just... they were discussing what it is they do and how it works, and what they essentially said is, "Look, you're tapping into deep monkey stuff."

Lex Fridman 8:37 (laughs)

Bret Weinstein 8:38 Right? And I thought, "Yeah, that's about right." And, you know, anybody who is proficient at something like skiing or skateboarding, you know, has the experience of flying down the hill on skis, for example. Bouncing from the top of one mogul to the next. And if you really pay attention, you will discover that your conscious mind is actually a spectator. It's there-

Lex Fridman 9:06 Mm-hmm.

Bret Weinstein 9:06 ... it's involved in the experience, but it's not driving. Some part of you knows how to ski and it's not the part of you that knows how to think. And I would just say that the... what accounts for this flexibility in humans is the ability to bootstrap a new software program and then drive it into the unconscious layer where it can be applied very rapidly and, you know, I will be shocked if the exact thing doesn't exist in robotics, you know. If you- if you programmed a robot to deal with circumstances that were novel to it, how would you do it? It would have to look something like this.

Lex Fridman 9:43 There's a certain kind of magic, you're right, with the consciousness being an observer, when you play guitar for example, or piano for me, music, when you get truly lost in it, I don't know what the heck is responsible for the flow of the music, the kind of- the loudness of the music going up and down, the timing, the intricate like- even the mistakes, all those things. That doesn't seem to be the conscious mind, uh, it's- it is just observing and yet it's somehow intricately involved. More, like, 'cause you mentioned parkour, dance is like that too. When you start, I mean tango dancing, you- you... if- when you truly lose yourself in it then it's just like you're an observer, and how the hell is the body able to do that? And not only that, it's the physical motion is also creating the emotion, the... like, the "damn, it's good to be alive" feeling. (laughs) So, but then that's also intricately connected to the full biology stack that we're operating

Lex Fridman 10:47 in. I don't know how difficult it is to replicate that. We were talking offline about, uh, Boston Dynamics' robots, um, they've recently been- they did both parkour, they did flips, they've also done some dancing and it's something I think a lot about because what most people don't realize, because they don't look deep enough is, those robots are hardcoded to do those things. The- the robots didn't figure it out by themselves and yet the- the fundamental aspect of what it means to be human is that process of figuring out, of making mistakes, and then there's something about overcoming those challenges and the mistakes and like, figuring out how to lose yourself in the magic of the dancing, uh, or just movement is what it means to be human, that learning process. So, that's what I wanna do with the- with the, um, almost as a fun side thing with the- the Boston Dynamics' robots, is to have them learn and see what they figure out even if it- even if they make mistakes. I want to let Spot make mistakes and in so doing, discover what it means to be alive, discover beauty, because I think that's the central aspect of mistakes.... Boston Dynamics folks want Spot to be perfect, (laughs) 'cause they don't want Spot to ever make mistakes because it wants to operate in the factories, it wants to be, you know, very safe and so on. For me, if you construct the environment, if you construct a safe space for robots, and allow them to make mistakes, something beautiful might be discovered. But that requires a lot of brain power. So, Spot is currently very dumb, and I'm gonna add a, give it a brain. Uh, so first, make it see. It currently can't see, meaning computer vision. It has to understand its environment, it has to see all the humans, but then also has to be able to learn. Learn about its movement, learn how to use its body to communicate with others, all those kinds of things that dogs know how to do well, humans know how to do somewhat well. Um, I think that's a beautiful challenge, but first, you have to allow the robot to make mistakes.

Bret Weinstein 13:00 Well, I think, um, your objective is laudable, but you're gonna realize that the Boston Dyna- Dynamics folks are right the first time Spot poops on your rug. (laughs)

Lex Fridman 13:10 (laughs) Um, I hear the same thing about kids and so on.

Bret Weinstein 13:13 Yes.

Lex Fridman 13:13 Uh, but I still wanna have kids. (laughs)

Bret Weinstein 13:15 No, you, you should. It's, it's, it's a great experience. Um, so let me step back into what you said in a couple of different places. One, I have always believed that the missing element in, uh, robotics and artificial intelligence is a proper development, right? It is no accident, it is no mere coincidence that human beings are the most dominant species on planet Earth, and that we have the longest childhoods of any creature on Earth by far, right? The dev- development is the key to the flexibility. And so, uh, the capability of a human at adulthood is the mirror image, it's the flip side of our helplessness at birth. Um, so I'll be very interested to see what happens in your, uh, robot project if you do not end up, uh, reinventing childhood for robots, which, of course, is foreshadowed in 2001 quite brilliantly.

Lex Fridman 14:10 Mm-hmm.

Bret Weinstein 14:11 Um, but I also wanna point out, you can, you can see this issue of your conscious mind becoming a spectator very well if you compare tennis to table tennis.

Lex Fridman 14:24 Hmm.

Bret Weinstein 14:24 Right? If you watch a tennis game, you could imagine that the players are highly conscious as they play. You cannot imagine that if you've ever played ping pong decently. A volley in ping pong is so fast that your conscious mind, if it had... if your reactions had to go through your conscious mind, you wouldn't be able to play. So, you can detect that your conscious mind, while very much present, isn't there. And you can also detect where consciousness does usefully intrude. If you go up against an opponent in table tennis that knows a trick that you don't know how to respond to, you will suddenly detect that something about your game is not effective, and you will start thinking about what might be, how do you position yourself so that move that puts the ball just in that corner of the table, or something like that, doesn't catch you off guard.

Lex Fridman 15:15 Mm-hmm.

Bret Weinstein 15:15 And this, I believe, is, w- w- ... We highly conscious folks, those of us who try to think through things very deliberately and carefully-

Lex Fridman 15:25 Mm-hmm.

Bret Weinstein 15:26 ... mistake consciousness for, like, the highest kind of thinking, and I really think that this is an error. Consciousness is an intermediate level of thinking. What it does is it allows you... It's basically like uncompiled code, and it doesn't run very fast, it is capable of being adapted to new circumstances, but once the code is roughed in, right, it gets driven into the unconscious layer and you become highly effective at whatever it is, and then from that point, your conscious mind basically remains there to detect things that aren't anticipated by the code you've already written. And so I don't exactly know how one would establish this, how one would demonstrate it, but it must be the case that the human mind contains sandboxes in which things are tested, right? Maybe you can build a piece of code and run it in parallel next to your active code so you can see how it would have done comparatively. Um, but there's gotta be some way of writing new code and then swapping it in. And frankly, I think this has a lot to do with things like sleep cycles. Very often, you know, when I get good at something, I often don't get better at it while I'm doing it. I get better at it when I'm not doing it, especially if there's time to sleep and think on it. So, there's some sort of, you know, new program swapping in for old program phenomenon which, um, you know, will be a lot easier to see in machines.

Lex Fridman 16:50 (laughs)

Bret Weinstein 16:50 It's gonna be hard with the, the, the wetware.

Lex Fridman 16:54 I like... I mean, it, it is true 'cause as somebody that played, uh, played tennis for many years, I do still think the highest form of excellence in tennis is when the conscious mind is a spectator. So, it's... The compiled code is the highest form of being human, and then consciousness is just some, like, specific compiler. Y- you just have, like, Borland C++ compiler. You could just have different kind of compilers. Ultimately, the, the thing that by which we measure the, um, the power of life, the intelligence of life is the compiled code, and you can probably do that compilation all kinds of ways.

Bret Weinstein 17:34 Yep. Yeah, I'm not saying that tennis is played consciously and table tennis isn't. I'm saying that because tennis is slowed down by the, just the space on the court-

Lex Fridman 17:43 Right.

Bret Weinstein 17:43 ... you could, you could imagine that it was your conscious mind playing.

Lex Fridman 17:46 Mm-hmm.

Bret Weinstein 17:47 But when you shrink the court down-

Lex Fridman 17:48 It becomes obvious.

Bret Weinstein 17:49 ... it becomes obvious that your conscious mind is just present, rather than knowing where to put the paddle. And weirdly for me, um, I would say this probably isn't true in a podcast situation, but if I have to give a presentation, uh, especially if I have not overly prepared, I often find the same phenomenon when I'm giving the presentation. My conscious mind is there watching some other part of me present-

Lex Fridman 18:13 (laughs)

Bret Weinstein 18:13 ... which, uh, is a little jarring, I have to say.

Lex Fridman 18:17 Well, that, that means you've, you've gotten good at it.... not let the conscious mind get in the way of the flow of words. (laughs)

Bret Weinstein 18:24 Yeah, that's, that's the sensation, to be sure.

Lex Fridman 18:27 And, and that's the highest form of podcasting too. I mean, that's why you have ... Tha- that's what it looks like when a podcast is really in the pocket, like, like Joe Rogan just having fun and just losing themselves, and that's, that's something I aspire to as well, just losing yourself in the conversa- Somebody that has a lot of anxiety with people, like I'm such an introvert. I'm scared, I was scared before you showed up, I'm scared right now. Um, there's just anxiety, there's just, it's a giant mess. Uh, it's hard to lose yourself. It's hard to just get out of the way of your own mind.

Bret Weinstein 19:00 Yeah, actually, uh, trust i- is a big component of that. Your, your conscious mind, uh, retains control if you are very uncertain. Um, but when you do, when you do get into that zone when you're speaking, I realize it's different for you with English as a second language, although maybe you present in Russian and, and-

Lex Fridman 19:18 (laughs)

Bret Weinstein 19:19 ... and, you know, and it happens, but-

Lex Fridman 19:20 Yeah.

Bret Weinstein 19:20 ... do you ever hear yourself say something and you think, "Oh, that's really good." Right? Like, like you didn't come up with it, some other part of you that you don't exactly know came up with it?

Lex Fridman 19:31 I, I don't think I've ever he- heard myself in that way, because I have a much louder voice that's constantly yelling in my head at, "Why the hell did you say that?" There's a very self-critical voice-

Bret Weinstein 19:45 Yeesh.

Lex Fridman 19:45 ... that's much louder, so I'm very ... (laughs) Um, maybe I need to deal with that voice, but it's been like with a, with a, what is it called, like a megaphone just screaming, so I can't hear the other voice-

Bret Weinstein 19:55 Oh, no. (laughs)

Lex Fridman 19:55 ... that says, "Good job, you said that thing really nicely." So, I'm kind of focused right now on, on the megaphone person i- in the audience (laughs) versus the, the posi- But it's, that's definitely something to think about. It's been productive, but, you know, the, the place where I find gratitude and beauty and appreciation of life is in the quiet moments when I don't talk, when I listen to the world around me, when I listen to others, um, when I talk, I'm extremely self-critical in my mind. Uh, when I, when I produce anything out into the world that's, that originated with me, like any kind of creation, extremely self-critical. It's good for productivity, for like always striving to improve and so on. It might be bad for, (laughs) for like just appreciating the things you've created. I'm a little bit with Marvin Minsky on this, where he says, "The, the, the key to, um, to a productive life is to hate everything you've ever done in the past."

Bret Weinstein 20:58 I didn't know he said that. I must say, I resonate with it a bit. I-

Lex Fridman 21:01 (laughs)

Bret Weinstein 21:02 And, you know, I, uh, uh, unfortunately, my life currently has me putting a lot of stuff into the world, and I effectively watch almost none of it. I can't stand it. (laughs)

Lex Fridman 21:15 (laughs) Yeah. What, what do you make of that? I don't know. I just recent- I just yesterday read, um, Metamorphosis by K- re-read Metamorphosis by Kafka where he turns into a giant bug because of the stress that the world puts on him, his parents put on him to succeed. And, you know, I think that's, you have to find the balance, 'cause if, if you, if you allow the self-critical voice to become too heavy, the burden of the world, the pressure that, that the world puts on you to be the best version of yourself and so on, to strive, then you become a bug and that's a big problem. (laughs) And then, uh, and then ev- and then the world turns against you, because you're a bug. You become some kind of caricature of yourself. Uh, I, I don't know. (sighs) You become the worst version of yourself and then thereby end up destroying yourself and then the world moves on. That's the story.

Bret Weinstein 22:10 That's a lovely story.

Lex Fridman 22:11 (laughs)

Bret Weinstein 22:11 Um, I do think this is one of these places, and frankly you could map this onto all of modern human experience, but this is one of these places where our ancestral programming does not serve our modern selves.

Lex Fridman 22:25 Yeah.

Bret Weinstein 22:25 So I used to talk to students about the question of dwelling on things. You know, dwelling on things is famously understood to be bad, and it can't possibly be bad. It wouldn't exist, a tendency toward it wouldn't exist if it was bad. So, what is bad is dwelling on things past the point of utility, and, you know, that's obviously easier to say than to operationalize. But if you realize that your dwelling is the key, in fact, to upgrading your program for future wellbeing-

Lex Fridman 22:57 Mm-hmm.

Bret Weinstein 22:57 ... and that there's a point, you know, presumably from diminishing returns if not counter-productivity, there is a point at which you should stop because that is what is in your best interest, then knowing that you're looking for that point is useful, right? This is the point at which it is no longer useful for me to dwell on this error I have made, right? That is, that's what you're looking for and, um, it also gives you license, right? If, if some part of you feels like i- it, you know, is punishing you rather than searching, then that also has a point at which it's no longer valuable and there's some liberty in realizing, yep, even the part of me that was punishing me knows it's time to stop.

Lex Fridman 23:37 (laughs) So, if we, if we, uh, map that onto compiled code discussion, as a computer science person I find that very compelling, you know, there's a, when you compile code, you get warnings sometimes and, um, usually if you're a good software engineer, you're going to make sure there's no, you know, you treat warnings as errors so you, you make sure the compilation produces no warnings. But at a certain point when you have a large enough system, you just let the warnings go. It's fine. Like, I don't know where that warning came from, but, you know, (laughs) it's just, uh, ultimately you need to compile the code and, and run with it and, uh, hope nothing terrible happens.

Bret Weinstein 24:19 Well, I think what you will find, and believe me, I think what you're, what you're talking about with respect to robots unlearning, uh-... is gonna end up having to go to a deep developmental state and a helplessness that evolves into hypercompetence and all of that. But, um, I- I live, I noticed that I live by something that I, for lack of a better descriptor, call the theory of close calls.

Lex Fridman 24:47 (laughs)

Bret Weinstein 24:47 And the theory of close calls says that people-

Lex Fridman 24:50 (laughs)

Bret Weinstein 24:50 ... typically miscategorize the events in their life where something almost went wrong. And, you know, for example, if you... I have a friend who, um... I was walking down the street with my- my college friends, and one of my friends stepped into the street thinking it was clear and was nearly hit by a car going 45 miles an hour. It would have been an absolute disaster. Might have killed her. Certainly would have permanently injured her. But she didn't... You know, the car didn't touch her, right? Now, you could walk away from that and think nothing of it because, well, what is there to think? Nothing happened.

Lex Fridman 25:27 Mm-hmm.

Bret Weinstein 25:28 Or you could think, "Well, what is the difference between what did happen and my death?" The difference is luck. I never want that to be true, right? I never want the difference between what did happen and my death to be luck. Therefore, I should count this as very close to death, and I should prioritize coding so it doesn't happen again at a very high level. So anyway, my- my basic point is, the accidents and disasters and misfortune describe a distribution that tells you what's really likely to get you in the end. And so, um, personally, you can use them to figure out where the dangers are so that you can afford to take great risks because you have a really good sense of how they're gonna go wrong, but I would also point out, civilization has this problem. Civilization is now producing these events that are major disasters, but they're not existential scale yet, right? They're very serious errors that we can see, and I would argue that the pattern is, you discover that we are involved in some industrial process at the point it has gone wrong, right? So, I'm now always asking the question, okay, in light of the Fukushima triple meltdown, the financial collapse of 2008, the Deepwater Horizon, uh, blowout, COVID-19 and its probable origins in the Wuhan lab, what processes do I not know the name of yet that I will discover at the point that some gigantic accident has happened? And can we talk about the wisdom or lack thereof of engaging in that process before the accident, right? That's what a wise civilization would be doing, and yet we don't.

Lex Fridman 27:13 I- I just want to mention something that happened a couple of days ago. Uh, I don't know if- if you know who J.B. Straubel is. He's the co-founder of Tesla, CTO of Tesla for many, many years. His wife just died. She was riding a bicycle, and in the same, in the same thin line between death and life that many of us have been in where you walk into an intersection and there's this close call, every once in a while, it... You get the- the short straw. I wonder how much of our own individual lives and the entirety of the human civilization rests on this little roll of the dice.

Bret Weinstein 28:00 Well, this is sort of my point about the close calls, is that there's a, there's a level at which we can't control it, right? The gigantic asteroid that comes from deep space that you don't have time to do anything about. There's not a lot we can do to hedge that out, or at least not short term. But there are lots of other things, you know. Obviously, the financial collapse of 2008 didn't break down the entire world economy. It threatened to, but a Herculean effort managed to pull us back from the brink. The triple meltdown at Fukushima was awful, but every one of the seven fuel pools held. There wasn't a major fire that made it impossible to manage the disaster going forward. We got lucky. Um, you know, we could say the same thing about the, uh, the blowout at the Deepwater Horizon where a hole in the ocean floor large enough that we couldn't have plugged it could have opened up. All of these things could have been much, much worse, right? And I think we can say the same thing about COVID. As terrible as it is, and I, you know... We cannot say for sure that it came from the Wuhan lab, but there's a strong likelihood that it did, and it also could be much, much worse. So, in each of these cases, something is telling us we have a process that is unfolding that keeps creating risks where it is luck that is the difference between us and some scale of disaster that is unimaginable, and that wisdom... You know, you can be highly intelligent and cause these disasters. To be wise is to stop causing them, right? And that would require a process of restraint, a process that I don't see a lot of evidence of yet. So, I think we have to generate it and somehow, we... You know, at the moment, we don't have a political structure that would be capable of taking a protective algorithm and actually deploying it, right? Because it would have important economic consequences, and so it would almost certainly be shot down. But we can obviously also say, you know, we paid a huge price for all of the disasters that, uh, I- I've mentioned, and we have to factor that into the equation. Something can be very productive short term and very destructive long term.

Lex Fridman 30:15 A- and also, the question is, how many disasters we avoided because of the ingenuity, uh, of humans or just the integrity and character of humans. That's sort of an open question. Uh, we may be more...... intelligent, then lucky. (laughs) That's the hope. Because the optimistic message here that you're getting at is maybe the process that we should be, that maybe we can overcome luck with ingenuity. Meaning, I guess you're suggesting the process is we should be listing all the ways a human civilization can destroy itself, uh, assigning likelihood to it, and thinking through how can we avoid that? And being very honest that, with the data out there about the close calls, and using those close calls to, uh, to then create sort of mechanism by which we minimize the probability of those close calls, and just being honest and, uh, transparent, uh, with the data that's out there.

Bret Weinstein 31:23 Well, I think w- we need to do a couple things for it to work. Um, so I've been an advocate for the idea that sustainability is actually, it's difficult to operationalize, but it is an objective that we have to meet if we're to be around long-term. And I've realized that we also need to have reversibility of all of our processes, because processes very frequently when they start do not appear dangerous, and then they, when they scale, they become very dangerous. So for example, if you imagine the first internal combustion engine in a vehicle driving down the street, and you imagine somebody running after them saying, "Hey, if you do enough of that, you're gonna alter the atmosphere and it's gonna change the temperature of the planet," it's preposterous, right?

Lex Fridman 32:07 Mm-hmm.

Bret Weinstein 32:07 Why would you stop the person who's invented this marvelous new contraption? But of course, eventually you do get to the place where you're doing enough of this that you do start changing the temperature of the planet. So if we built the capacity, if we basically said, "Look, you can't involve yourself in any process that you couldn't reverse if you had to," then progress would be slowed, but our safety would go up dramatically. And I, I think, I think in some sense, if we are to be around long-term, we have to begin thinking that way. We're just involved in too many very dangerous processes.

Lex Fridman 32:43 So let's talk about one of the things that, um, if not threatened human civilization, certainly hurt it, uh, at a deep level, which is COVID-19.

Lex Fridman 32:56 What percent probability would you currently place on the hypothesis that COVID-19 leaked from the Wuhan Institute of Virology?

Bret Weinstein 33:06 So I maintain a flowchart of all the possible explanations, and it doesn't break down exactly that way. The likelihood that it emerged from a lab is very, very high. If it emerged from a lab, the likelihood that the lab was the Wuhan Institute is very, very high. The, there are multiple different kinds of evidence that point to the lab, and there are, is literally no evidence that points to nature. Either the evidence points nowhere or it points to the lab, and the lab could mean any lab, but geographically obviously, the labs in Wuhan are the most likely, and the lab that was most directly involved with research on viruses that look like COVID, uh, that look like SARS-CoV-2 is obviously the place, uh, that one would start. But I would say the likelihood that this virus came from a lab is well above 95%. We can talk about the question of could a virus have been brought into the lab and escaped from there without being modified. That's also possible, but it doesn't explain any of the anomalies in the genome of SARS-CoV-2. Um, could it have been delivered from another lab? Could Wuhan be a distraction, um, in order that p- we would connect the dots in the wrong way? That's conceivable. I currently have that below 1% on my flowchart, but I think-

Lex Fridman 34:30 A very dark thought (laughs) that somebody would, would do that almost as a, a political attack on China.

Bret Weinstein 34:37 Well, it depends. Uh, I don't even think... That's one possibility. Sometimes when Erik and I talk about these issues, we will generate a scenario just to prove that something could, could live in that space, right? It's a placeholder for whatever may actually have happened. And so it doesn't have to have been an attack on China. That's certainly one possibility, but I would point out, if you can predict the future in some unusual way better than others, you can print money, right? That's what markets that allow you to bet for or against, uh, virtually any sector allow you to do. So you can imagine a simply amoral person or entity-

Lex Fridman 35:24 Yeah.

Bret Weinstein 35:25 ... generating a pandemic, attempting to cover their tracks because it would allow them to bet against things like, you know, cruise ships, air travel, whatever it is, and bet in favor of, I don't know, uh, sanitizing gel and w- whatever else you would do. So am I saying that I think somebody did that? No, I really don't think it happened. We've seen zero evidence that this was intentionally released. However, were it to have been intentionally released by somebody who did not know, did not want it known where it had come from, releasing it in Wuhan would be one way to cover their tracks, so we have to leave the possibility formally open-

Lex Fridman 36:05 Mm-hmm.

Bret Weinstein 36:05 ... but acknowledge there's no evidence, so.

Lex Fridman 36:07 And the probability therefore is low. I tend to believe, maybe this is the optimistic nature that I have, that people who are competent enough to do the kind of thing we just described are not going to do that because th- uh, it requires a certain kind of, I don't wanna use the word evil, but whatever word you wanna use to describe the kinda, uh, disregard for human life required to do that.You're just n- that's like, that's just not going to be coupled with competence. I f- I feel like there's a trade-off chart, where competence on one axis and evil's on the other. And the more evil you become, the- the- the- the crappier you are at doing great engineering, scientific work required to deliver weapons of different kinds, whether it's bioweapons or nuclear weapons and all those kinds of things. That seems to be the lessons I- I take from history, but that doesn't necessarily mean that's what's going to be happening in the future. But to stick on the lab leak idea, the- the- 'cause the flowchart is probably huge here 'cause there's a lot of fascinating possibilities. One question I wanna ask is, um, what would evidence for natural origins look like? So, one piece of evidence for natural origins is, um, that it has happened in the past, that- that viruses have jumped.

Bret Weinstein 37:33 Oh, they do jump.

Lex Fridman 37:34 They- they... So, like, that's one, like, that's possible to have happened, you know. (laughs) So, that- that's a sort of like a historical evidence, like, okay, well, it's possible that it hap- it's possible-

Bret Weinstein 37:45 It's not, it's not evidence of the kind you think it is. It's a justification for a presumption.

Lex Fridman 37:51 Right.

Bret Weinstein 37:51 Right? So, the presumption upon discovering a new virus circulating is certainly that it came from nature, right? The problem is the presumption evaporates in the face of evidence, or at least it logically should.

Lex Fridman 38:03 Right.

Bret Weinstein 38:04 And it didn't in this case. It was maintained by people who-

Lex Fridman 38:07 Yes.

Bret Weinstein 38:07 ... privately in their emails acknowledged that they had grave doubts about the natural origin of- of this virus.

Lex Fridman 38:14 Mm-hmm. Is there some other piece of evidence that we could s- look for and see that would say, "Hmm, this increases the probability that it's natural origins?"

Bret Weinstein 38:24 Yeah. Um, in fact, there is evidence. You know, I always worry that somebody is going to make up some evidence in order to reverse the flow.

Lex Fridman 38:34 Oh, boy.

Bret Weinstein 38:35 Well, let's say I am part-

Lex Fridman 38:36 There's a lot of incentive for that, actually.

Bret Weinstein 38:38 There's a huge amount of incentive. On the other hand, why didn't the powers that be, the powers that lied to us about weapons of mass destruction in Iraq, why didn't they ever fake weapons of mass destruction in Iraq? Whatever force it is, I hope that force is here too, and so whatever evidence we find is real.

Lex Fridman 38:55 It's- it's the competence thing I'm talking about, but okay, go ahead, sorry. (laughs)

Bret Weinstein 38:58 Okay. Well, we can get back to that.

Lex Fridman 39:00 Yeah.

Bret Weinstein 39:00 But I would say, yeah, the- the- the giant piece of evidence that will shift the probabilities in the other direction is the discovery of either a human population in which the virus circulated prior to showing up in Wuhan that would explain where the virus learned all of the tricks that it knew instantly upon spreading from Wuhan. So, that would do it, or an animal population in which an ancestor epidemic can be found in which the virus learned this before jumping to humans. But I'd point out in that second case, you would certainly expect to see a great deal of evolution in the early epidemic, which we don't see. So, there almost has to be a human population somewhere else that had the virus circulating or an ancestor of the virus that we first saw in Wuhan circulating, and it has to have gotten very sophisticated in that prior epidemic before hitting Wuhan in order to explain the total lack of evolution and extremely effective virus that emerged, uh, at the end of 2019.

Lex Fridman 40:01 So, you don't believe in the magic of evolution to spring up with all the tricks already there? Like, everybody who doesn't have the tricks, they die quickly, and then you just have this beautiful virus that comes in with the spike protein and the- and the- through mutation, uh, and selection, just do... Like, the ones that succeed and succeed big are the ones that are going to just spring into life with the tricks.

Bret Weinstein 40:26 Well, no.

Lex Fridman 40:28 (laughs)

Bret Weinstein 40:28 That's- that's called a hopeful monster, and-

Lex Fridman 40:30 Okay.

Bret Weinstein 40:31 ... hopeful monsters don't work. It's too... The job of becoming a new pandemic virus is too difficult. It involves two very difficult steps, and they both have to work. One is the ability to infect a person and spread in their tissues, uh, sufficient to make an infection, and the other is to jump between individuals at a sufficient rate that it doesn't go extinct for one reason or another.

Lex Fridman 40:53 Right.

Bret Weinstein 40:53 Those are both very difficult jobs. They require, as you describe, selection. And the point is selection would leave a mark. We would see evidence that it was taking place.

Lex Fridman 41:02 In animals or humans, what would see?

Bret Weinstein 41:04 Both, right? And-

Lex Fridman 41:05 We would see this evolutionary trace of, um, of the virus gathering the tricks up?

Bret Weinstein 41:10 Yeah. You would see the virus, uh, you would see the clumsy virus get better and better. And yes, I am a full believer in the power of that process. In fact, I believe it, uh... What I know from studying the process is that it is much more powerful than most people imagine, that what we teach in the Evolution 101 textbook is too clumsy a process to do what we see it doing, and that actually people should increase their expectation of the rapidity with which that process can produce, um, just jaw-dropping adaptations. That said, we just don't see evidence that it happened here, which doesn't mean it doesn't exist. But it means in spite of immense pressure to find it somewhere, there's been no hint, which probably means it took place inside of a laboratory.

Lex Fridman 41:55 So, inside the laboratory, gain-of-function research on viruses, and I believe most of that kind of research is doing this exact thing that you're referring to, which is accelerated evolution and just watching evolution do its thing on a bunch of viruses and seeing what kind of, uh, tricks get developed. The other method is, uh, engineering viruses, so manually adding on the tricks.

Bret Weinstein 42:24 (laughs)

Lex Fridman 42:26 Uh, what, which do you think we should be thinking about here?

Bret Weinstein 42:31 So, mind you, I learned what I know in the aftermath of this pandemic emerging. I- I started studying the question. And I would say-... based on the content of the genome and other evidence in publications from the various labs that were involved in, in, uh, generating this technology, a couple of things seem likely. This SARS-CoV-2 does not appear to be entirely the result of either a splicing process or serial passaging. It, it, it appears to have both, uh, things in its past, or it's at least highly likely that it does. So, for example, the furin cleavage site looks very much like it, uh, was added in to the virus, and it was known that that would increase its infectivity in humans and increase its tropism. The, uh, virus appears to be excellent at spreading in humans and minks and ferrets. Now, minks and ferrets are very closely related to each other, and ferrets are very likely to have been used in a serial passage experiment. The reason being that they have an ACE2 receptor that looks very much like the human ACE2 receptor. And so, were you going to passage the virus, or its ancestor, through an animal in order to increase its infectivity in humans, which would have been necessary, uh, ferrets would have been very likely. It is also quite likely that humanized mice were, uh, utilized, and it is possible that human airway tissue was utilized. I think it is vital that we find out what the protocols were. If this came from the Wuhan Institute, we need to know it, and we need to know what the protocols were exactly, because they will actually give us some tools that would be useful in fighting SARS-CoV-2 and hopefully driving it to extinction, which ought to be our priority. It is a, it is a priority that does not, it is not apparent from our behavior, but it really is, uh, it should be our objective. If we understood where our interests lie, w- we would be much more focused on it. Um, but those protocols would tell us a great deal. If it wasn't the Wuhan Institute, we need to know that. If it was Nature, we need to know that. And if it was some other laboratory, we need to figure out w- who, what, and where, so that we can determine what is, uh, what we can determine about, about what was done.

Lex Fridman 44:53 You're opening up my mind about why we should investigate, why we should know the truth of the origins of this virus. So, for me, personally, let me just tell the story of my own kind of journey. Uh, when I first started looking into the lab leak hypothesis, what became terrifying to me and important to unders- understand and obvious, is the sort of like Sam Harris way of thinking, which is, it's obvious that a lab leak of a deadly virus will eventually happen. My mind was, it doesn't even matter if it happened in this case, it's obvious that it's going to be, happen in the future, so why the hell are we not freaking out about this? And COVID-19 is not even that deadly relative to the possible future viruses. It's this, uh, the way, I disagree with Sam on this, but he thinks about this way about AGI as well, about a- artificial intelligence. It's a different discussion, I think, but with viruses, it seems like something that could happen on the scale of years, maybe a few decades. AGI is a little bit farther out for me. But it seemed, the terrifying thing, it seemed obvious that this will happen very soon for a much deadlier virus, as we get better and better at both engineering viruses and doing this kind of evolutionary-driven research, gain-of-function research. Okay, but then you started speaking out about this as well, but also started to say, "No, no, no, we should hurry up and figure out the origins now, because it will help us figure out how to actually respond, uh, to this particular virus, how to treat this particular virus, what is, in terms of vaccines, in terms of antiviral drugs, in terms of just, uh, all the, the h- the number of responses we should have." Okay. I s- still am much more freaking out (laughs) about the future. Um, maybe you can break that apart a little bit. Which are you f- most, um, focused on now? Which are you most freaking out about now in terms of the importance of figuring out the origins of this virus?

Bret Weinstein 47:10 I am most freaking out about both of them-

Lex Fridman 47:13 (laughs)

Bret Weinstein 47:13 ... (laughs) because they're both really important. And, you know, we can put bounds on this. Let me say first that this is a perfect test case for the theory of close calls, because as much as-

Lex Fridman 47:23 Yeah.

Bret Weinstein 47:23 ... COVID is a disaster, it is also a close call from which we can learn much.

Lex Fridman 47:28 Yeah.

Bret Weinstein 47:28 You are absolutely right. If we keep playing this game in the lab, if we are not, uh, if we are, especially if we do it under pressure, and when we are told that a virus is going to leap from nature any day, and that the more we know, the better we'll be able to fight it, we're gonna create the disaster all, all the sooner. So yes, that should be an absolute focus. The fact that there were people saying that this was dangerous back in 2015, uh, ought to tell us something. The fact that the system bypassed a ban and offshored the work to China ought to tell us this is not a Chinese failure. This is a failure of something l- larger and harder to see. Um, but I also think that there's a, there's a clock ticking with respect to SARS-CoV-2 and COVID, the disease that it creates, and that has to do with whether or not we are stuck with it permanently. So, if you think about the cost to humanity of being stuck with influenza, it's an immense cost year after year, and we just stop thinking about it because it's there. Some years you get the flu, most years you don't. Maybe you get the vaccine to prevent it. Maybe the vaccine isn't particularly well-targeted. But imagine just simply doubling that cost.Imagine we get stuck with SARS-CoV-2 and its descendants going forward, and that it just settles in and becomes a fact of modern human life. That would be a disaster, right? The number of people we will ultimately lose is incalculable. The amount of suffering that will be caused is incalculable. The loss of wellbeing and wealth, incalculable. So, that ought to be a very high priority, driving this extinct before it becomes permanent, and the ability to drive it extinct goes down the longer we delay effective responses. To the extent that we let it have this very large canvas, large numbers of people who have the disease in which mutation and selection can result in adaptation that we will not be able to counter, the greater its ability to figure out features of our immune system and use them, uh, to its advantage. So, I'm feeling the pressure of driving it extinct. I believe we could have driven it extinct six months ago, and we didn't do it because of very mundane concerns among a small number of people. And I'm not alleging that they were brazen about, um... or that they were callous about deaths that would be caused. I have the sense that they were working from a kind of autopilot in which you, you know, let's say you're in some kind of a corporation, a pharmaceutical corporation. You have a, a portfolio of therapies that in the context of a pandemic might be very lucrative. Those therapies have competitors. You, of course, want to position your product so that it succeeds and the competitors don't. And lo and behold, at some point, through means that I think those of us on the outside can't really, uh, intuit, you end up saying things about competing therapies that work better and much more safely than the ones you're selling that aren't true and do cause people to die in large numbers. Um, but it, you know, it- it's some kind of autopilot, at least part of it is.

Lex Fridman 50:43 So, the... there's a complicated coupling of the autopilot of, uh, institutions, companies, governments, and then there's also the geopolitical game theory thing going on where you wanna keep secrets. It's the Chernobyl thing where if you messed up, there is a big incentive, I think, to hide the fact that you messed up. So, how do we fix this? And what's more important to fix, the autopilot, which is the response that we often criticize about our institutions, es- especially the leaders in those institutions, Anthony Fauci and so on, uh, some of the members of the scientific community? And the second part is the, um, the game with China, of hiding the information, uh, in terms of on the fight between nations.

Bret Weinstein 51:40 Well, in our live streams on Darkhorse, Heather and I have been talking from the beginning about the fact that although, yes, what happens began in China, it very much looks like a failure of the international scientific community. And that's frightening, but it's also hopeful in the sense that actually if we did the right thing now, we're not navigating a puzzle about Chinese responsibility. We're navigating a question of collective responsibility for something that has been terribly costly to all of us. So, that's not a very happy process. But as you point out, what's at stake is in large measure, at the very least, the strong possibility this will happen again, and that at some point it will be far worse. So, you know, just as a person that does not learn the lessons of their own errors doesn't get smarter and they remain in danger, we collectively, humanity, have to say, "Well, there sure is a lot of evidence that suggests that this is a self-inflicted wound." When you have done something that has caused a massive self-inflected wound, self-inflicted wound, it makes sense to dwell on it exactly to the point that you have learned the lesson that makes it very, very unlikely that something similar will happen again.

Lex Fridman 53:01 I think this is a good place to kind of ask you to do almost like a thought experiment, or, um, to steel man the argument against the lab leak hypothesis. So, if you were to argue, uh, you know, you said 95% chance that it, uh, the virus leaked from a lab. There's a bunch of ways that I think you can argue that even talking about it is, uh, bad for the world. S- so, I, if I just d- put something on the table, it's to say that, um, for one, it would be racism versus Chinese people, that talking about that it leaked from a lab, there's a kind of immediate kind of blame, and it can spiral down into this idea that s- somehow the people are responsible for the virus and this, this kind of thing. Is it possible for you to come up with other steel man arguments against talking or against the possibility of the lab leak hypothesis?

Bret Weinstein 54:13 Well, so I think steel manning is a tool that is extremely valuable, but it's also possible to abuse it.

Lex Fridman 54:22 Hm.

Bret Weinstein 54:22 I think that you can only steel man a good faith argument, and the problem is, we now know that we have not been engaged in opponents who were wielding good faith arguments because privately, their emails reflect their own doubts, and what they were doing publicly was actually a punishment, a public punishment for those of us who spoke up with, I think, the purpose of...... either backing us down, or more likely, warning others not to engage in the same kind of behavior. And obviously for people like you and me who, uh, regard science as our likely best hope for navigating difficult waters, shutting down people who are using those tools honorably is itself dishonorable. So, I, I don't really, I don't feel that it is, I don't feel that there's anything to steel man. And I also think that, you know, immediately at the point that the world suddenly, with no new evidence on the table, switched gears with respect to the lab leak, you know, at the point that Nicholas Wade had published his article and suddenly the world was going to admit that this was at least a possibility, if not a likelihood, we got to see something of the rationalization process that had taken place inside the institutional world. And it very definitely involved the claim that what was being avoided was the targeting of, uh, Chinese scientists. And my point would be, I don't want to see the targeting of anyone. I don't want to see racism of any kind. On the other hand, once you create license to lie in order to protect individuals when the world has a stake in knowing what happened, then it is inevitable that that process, that license to lie will be used by the thing that captures institutions for its own purposes. So, my sense is, it may be very unfortunate if the story of what happened here can be used against, uh, Chinese people. That would be very unfortunate and a- as I think I mentioned, Heather and I have taken great pains to point out that this doesn't look like a Chinese failure, it looks like a failure of the international scientific community. So, I think it is important to broadcast that message along with the analysis of the evidence. But no matter what happened, we have a right to know and, um, I- I frankly do not take the institutional layer at its word that its motivations are honorable and that it was protecting, you know, good-hearted scientists at the expense of the world. That, that explanation does not add up.

Lex Fridman 57:00 Well, this is a very interesting question about whether it's ever okay to lie at the institutional layer to protect the populous. I think both you and I are probably on the same... have the same sense that it's a slippery slope. Even if it's an effective mechanism in the short-term, in the long-term, it's going to be destructive. This happened with masks. This happened with, with other things. If you look at just history of pandemics, there is, um, there's an idea that panic is destructive amongst the populous, so you, you want to construct a narrative, whether it's a lie or not, to, uh, minimize panic. But you're suggesting that almost in all cases, and I think that was the lesson from the, the pandemic in the early 20th century, that lying creates distrust, and distrust in the institutions is ultimately destructive. That's your sense? That lying is not okay?

Bret Weinstein 58:10 Well, okay. You know, there are obviously places where complete transparency is not a good idea, right?

Lex Fridman 58:17 Okay.

Bret Weinstein 58:17 To the extent that you broadcast a technology that allows one individual to hold the world hostage, right? Obviously, you've, you've got a something to be navigated. But, in general, I don't believe that the scientific system should be lying to us. In the case of this particular lie, the idea that the wellbeing of Chinese scientists outweighs the wellbeing of the world is preposterous.

Lex Fridman 58:49 Yeah.

Bret Weinstein 58:50 Right? As you, as you point out, one thing that rests on this question is whether we continue to do this kind of research going forward, and the scientists in question, all of them, American, Chinese, all of them were pushing the idea that the risk of a zoonotic spillover event causing a major and highly destructive pandemic was so great that we had to risk this. Now, if they themselves have caused it, and if they are wrong, as I believe they are, about the likelihood of a major world pandemic spilling out of nature in the way that they, uh, wrote into their grant applications, then the danger is, you know, the call is coming from inside the house. And, uh, we have to look at that. And yes, whatever we have to do to protect scientists from retribution, we should do. But, we cannot... Protecting them by lying to the world, and even worse, by demonizing people like me, like, uh, Josh Rogin, like Yuri Deigin, the entire Drastic group on Twitter, by demonizing us for simply following the evidence is to set a terrible precedent, right? You're demonizing people for using the scientific method to evaluate evidence that is available to us in the world. What a terrible crime it is to teach that lesson, right? Thou shalt not use scientific tools. No, I'm sorry. Whatever your license to lie is, it doesn't extend to that.

Lex Fridman 1:00:22 Yeah. I've seen the, the attacks on you, the pressure on you. It has a very important effect on thousands of world-class biologists, actually. So, at MIT, colleagues of mine, people I know, there's a, there's a slight pressure to not be allowed to, uh, one, speak publicly, and two, actually think. (laughs) Like, to even think about these ideas, it, it sounds kind of ridiculous, but just the, in the, in the privacy of your own home, to, to read things, to, to think, it's many people, many world-class biologists that I know will just avoid looking at the data. There's not even that many people that are publicly opposing gain-of-function research. They, they're, they're also like, "Ugh, it's not worth it." (laughs) "It's not worth the battle." And there's many people that kind of argue that those battles should be fought in private, in, um, uh, you know, with colleagues in the, um, in the privacy of the scientific community, that the public is somehow not maybe intelligent enough to be able to deal with the, the, the complexities of this kind of discussion. I don't know. But the final result is, combined with the bullying of you and, uh, all the different pressures in the academic institutions is that it's just people are self-censoring and, and silencing themselves, and silencing the most important thing, which is the power of their brains. (laughs) Like, these people are brilliant and the fact that they're not utilizing their brain to come up with solutions, uh, outside of the conformist line of thinking is tragic.

Bret Weinstein 1:02:11 Well, it is. I also think that it, we have to look at it and understand it for what it is.

Lex Fridman 1:02:17 Yeah.

Bret Weinstein 1:02:17 I mean, for, for one thing, it's kind of a cryptic totalitarianism.

Lex Fridman 1:02:20 Yeah.

Bret Weinstein 1:02:20 Somehow people's sense of what they're allowed to think about, talk about, discuss is causing them to self-censor, and I can tell you, it's causing many of them to rationalize, which is even worse. They are blinding themselves to what they can see. Um, but, uh, it is also the case, I believe, that what you're describing about what people said, and a great many people understood that the lab leak hypothesis could not be taken off the table, but they didn't say so publicly, and I think that their discussions with each other about why they did not say what they understood, that's what capture sounds like on the inside. I don't know exactly what force captured the institutions. I don't think, uh, anybody knows for sure out here in public. I don't even know that it wasn't just simply a process, but you have these institutions. They are behaving towards a kind of somatic obligation. They have lost sight of what they were built to accomplish and on the inside, the way they avoid going back to their original mission is to say things to themselves like, "The public can't have this discussion. It can't be trusted with it. Yes, we need to be able to talk about this, but it has to be private." Whatever it is they say to themselves, that is what capture sounds like on the inside. It's a institutional ras- rationalization mechanism, and it's very, very deadly. And at the point you go from lab leak to repurposed drugs, you can see that it's very deadly in a very direct way.

Lex Fridman 1:03:54 Yeah, I see this on, in my field of, with, um, things like autonomous weapon systems. People in AI do not talk about the use of AI in weapon systems. They, they kind of avoid the idea that AI is used in, in the military. It, it's kind of funny. There, there's this, like, kind of discomfort and they're like, they all hurry, like, um, you know, like something scary happens and a bunch of sheep kind of like run a- run away. That's what it, it looks like. (laughs) And it, I don't even know what to do about it 'cause... And then I feel this n- natural pull every time I bring up autonomous weapon systems to go along with the sheep. There's a natural kind of pull towards that direction 'cause it's like, what can I do as one person? Now, there's currently nothing destructive happening with autonomous weapon systems, so we're in, like, in the early days of this race that in 10, 20 years might become a real problem. Now, we're, uh, the discussion we're having now, we're, we're now facing the result of that in the space of viruses, uh, like, f- for many years avoiding the conversations here. (sighs) I don't know what to do that in the early days, um, but I think we have to, I guess, create institutions where people can stand out. People can stand out and, like, uh, basically be individual thinkers and break out into all kinds of spaces of ideas that allow us to think freely. Freedom of thought. And maybe that requires the decentralization of institutions.

Bret Weinstein 1:05:22 Well, years ago, I came up with a concept called cultivated insecurity, and the idea is, let's just take the example of, uh, the average Joe, right? The average Joe has a job somewhere and their mortgage, their medical insurance, their retirement, their connection with the economy is, to one degree or another, dependent on their relationship with the employer. That means that there is a strong incentive, especially in any industry where it's not easy to move from one employer to the next, there's a strong incentive to, uh, stay in your employer's good graces, right? So, it creates a very top-down dynamic, not only in terms of who gets to tell, uh, uh, other people what to do, but it really comes down to who gets to tell other people how to think. So, that's extremely dangerous. The way out of it is to cultivate security. To the extent that somebody is in a position to go against the grain and have it not be a catastrophe for their family and their ability to earn, you will see that behavior a lot more. So, I, I would argue that some of what you're talking about is just a simple, uh, predictable consequence of the concentration of the, uh, the sources of well-being and that this is a solvable problem.

Lex Fridman 1:06:51 You got a chance to talk with Joe Rogan yesterday.

Bret Weinstein 1:06:55 Yes, I did.

Lex Fridman 1:06:56 And I just saw the episode was released, and, uh, ivermectin is trending on Twitter. Joe told me it was an incredible conversation, and I look forward to listening again today. Many people have probably, by the time this is released, have already listened to it. I think it would be interesting to discuss, uh, a postmortem. How do you feel how that conversation went? And maybe broadly, how do you see the story as it's unfolding of ivermectin, from the origins, from before COVID-19, through 2020, to today?

Bret Weinstein 1:07:35 I very much enjoyed talking to Joe, and I'm, uh, undescribably grateful that he would take the risk of such a discussion, that he would, as he described it, do an emergency podcast on the subject. Which I, I think that was not an exaggeration, this needed to happen for various reasons, that he, uh, you know, he took us down the road of talking about the censorship campaign against ivermectin, which I find utterly shocking, and talking about the drug itself. And I should say, we talked ... Uh, we had Pierre Kory available. Uh, he, he came on the podcast as well. He is, of course, the face of the FL triple C, the Front Line COVID-19 Critical Care Alliance. These are doctors who have innovated, uh, ways of treating COVID patients, and they happened on ivermectin, and have, um, been using it, and ... I hesitate to use the word advocating for it, because that's not really the role of doctors or scientists, but they are advocating for it in the sense that there is this, uh, pressure not to talk about its effectiveness, for reasons that we can go into.

Lex Fridman 1:08:44 So, maybe step back and say what is ivermectin? And, uh, how much studies have been done to show its effectiveness?

Bret Weinstein 1:08:54 So, ivermectin is an interesting drug. It was discovered in the '70s by a Japanese scientist named Satoshi Omura, and he found it in soil near a Japanese golf course. So, I would just point out in passing that if we were to stop self-silencing over, uh, the possibility that Asians will be demonized over the possible lab leak in Wuhan, and to recognize that actually the natural course of the story has a likely lab leak in China, it has a unlikely hero in Japan. Um, the story is naturally not a simple one. Um, but in any case, uh, Omura discovered this molecule. He, uh, sent it to a friend who was at Merck, a scientist named Campbell. They won a Nobel Prize for the discovery of the ivermectin molecule in 2015. Its initial use was in treating parasitic infections. It's very effective in treating, uh, the, um, the worm that causes river blindness, uh, the pathogen that causes elephantiasis, scabies. It's a very effective anti-parasite drug. It's extremely safe. It's on the WHO's list of essential medications. It's safe for children. It has been administered something like four billion times in the last four decades. It has been given away in the millions of doses by Merck in Africa. People have been on it for long periods of time, and in fact one of the reasons that Africa may have had less severe impacts from COVID-19 is that ivermectin is widely used there to prevent parasites, and the drug appears to have a long-lasting impact. So, it's an interesting molecule. It was discovered some time ago, apparently, that it has antiviral properties, and so it was tested in, early in the COVID-19 pandemic to see if it might work, uh, to treat humans with COVID. It turned out to have very promising, uh, evidence that it did treat humans. It was tested in tissues. It was tested at a very high dosage, which confuses people. They think that those of us who believe that ivermectin might be useful in confronting this disease are advocating those high doses, which is not the case. Um, but in any case, there have been quite a number of studies. A wonderful meta-analysis was finally released. We had seen it in preprint version, but it was finally peer-reviewed and published this last week. Um, it reveals that the drug, as clinicians have been telling us, those who have been using it, it's highly effective at treating people with the disease, especially if you get to them early. And it showed an 86% effectiveness as a prophylactic to prevent people from contracting COVID. And that number, 86%, is high enough to drive SARS-CoV-2 to extinction if we wished to deploy it.

Lex Fridman 1:11:56 First of all, the, the, the meta-analysis, is this the, uh, ivermectin for COVID-19 real-time meta-analysis of 60 studies? Or, there's a bunch of meta-analysis there. 'Cause I, I was really impressed by the real-time meta-analysis that keeps getting updated. I don't know if it's the same kind of thing.

Bret Weinstein 1:12:13 So, the one at, uh, ivmmeta.com?

Lex Fridman 1:12:18 Well, I saw it at, uh, c19ivermectin.com.

Bret Weinstein 1:12:20 Yeah, yeah. Um, no, this is not that meta-analysis. That, so that is, as you say, a living meta-analysis-

Lex Fridman 1:12:26 Yeah.

Bret Weinstein 1:12:26 ... where you can watch as evidence

Narrator 1:12:27 rolls in.

Lex Fridman 1:12:27 Which is super cool, by the way.

Bret Weinstein 1:12:29 It's really cool, and they've got some really nice graphics that allow you to understand, well, what is the evidence? You know?

Lex Fridman 1:12:34 Yeah.

Bret Weinstein 1:12:34 Oh, it's concentrated around this level of effectiveness, et cetera. So anyway, it's a great site. Well worth paying attention to. No, this was a meta-analysis, um, I don't know any of the authors but one. Second author is Tess Lory of the Bird Group, Bird being a, uh, a-... group of analysts and doctors in Britain that is playing a role similar to the FL Triple C here in the US. Um, so anyway, this is a meta-analysis that, that, uh, that, uh, Tess Lory and others did of all of the available evidence, and, um, it's quite compelling. People if... People can look for it on my Twitter. I will, I will put it up and people can find it there.

Lex Fridman 1:13:15 So, what about dose here, th- uh, in terms of safety, uh, h- uh, what do we understand about the, the kind of dose required to, to have that level of effectiveness, and what do we understand about its, the safety of that kind of dose?

Bret Weinstein 1:13:30 So let me just say, I'm not a medical doctor. I'm a biologist. Uh, I am on ivermectin, uh, in lieu of vaccination. In terms of dosage, there is one reason for concern, which is that the most effective dose for prophylaxis involves something like weekly administration. And that that is... Because that is not a historical pattern of use for the drug, um, it is possible that there is some long-term implication of being on it for... Uh, being on it weekly for a long period of time. There's not a strong indication of that. The safety signal that we have, uh, over people using the drug over many years and using it in high doses... In fact, Dr. Corey told me yesterday, um, that there are cases in which people have made, uh, calculation errors and taken a massive overdose of the drug and had no ill effect. So anyway, there's lots of reasons to think the drug is comparatively safe, but no drug is perfectly safe. And I do worry about the long-term implications of taking it. I also think it's very likely, um, because the drug is administered in, uh, y- a dose something like, let's say, 15 milligrams for somebody my size once a week after you've gone through the initial, um, the initial double dose that you take 48 hours apart, it is apparent that if the amount of drug in your system is sufficient to be protective at the end of the week, then it was probably far too high at the beginning of the week. So there's a question about whether or not you could flatten out the intake so that yo- the amount of ivermectin goes down, but the protection remains. I have little doubt that that would be discovered if we looked for it. Um, but that said, it does seem to be quite safe, highly effective at preventing COVID. The 86% number is plenty high enough for us to drive, uh, SARS-CoV-2 to extinction in light of its R naught number of slightly more than two. Um, and so why we are not using it is a bit of a, a mystery.

Lex Fridman 1:15:36 So even if everything you said now turns out to be not correct, it is nevertheless obvious that it's sufficiently promising, it always has been, in order to merit rigorous scientific exploration, investigation, doing a lot of studies, and certainly not censoring the science or the discussion of it.

Lex Fridman 1:16:00 So before we talk about the various vaccines for COVID-19, I'd like to talk to you about censorship.

Bret Weinstein 1:16:07 Mm-hmm.

Lex Fridman 1:16:08 Given everything you're saying, why did YouTube and other places censor discussion of, um, ivermectin?

Bret Weinstein 1:16:21 Well, there's a question about why they say they didn't, and there's a question about why they actually did it. Now, it is worth mentioning that YouTube is part of a consortium. It is partnered with Twitter, Facebook, Reuters, AP, Financial Times, Washington Post, some other notable organizations. And that this group has appointed itself the arbiter of truth. In effect, they have decided to control discussion ostensibly to prevent the distribution of misinformation.

Lex Fridman 1:17:00 Mm-hmm.

Bret Weinstein 1:17:00 Now, how have they chosen to do that? In this case, they have chosen to simply utilize the recommendations of the WHO and the CDC and apply them as if they are synonymous with scientific truth. Problem: even at their best, the WHO and CDC are not scientific entities. They are entities that are about public health. And public health has this, whether it's right or not, and I believe I disagree with it, but it has this, uh, self-assigned right to lie that comes from the fact that there is game theory that works against, for example, a successful vaccination campaign. That if everybody else takes a vaccine and therefore the herd becomes immune through vaccination, and you decide not to take a vaccine, then you benefit from the immunity of the herd without having taken the risk. So people who do best are the people who opt out. That's a hazard. And the WHO and CDC as public health entities effectively, um, oversimplify stories in order that that game theory does not cause a predictable tragedy of the commons. With that said, once that right to lie exists, then it finds out ser- i- it turns out to serve the interests of, for example, pharmaceutical companies, which have emergency use authorizations that require that there not be a safe and effective treatment and have immunity from liability for harms caused by their product. So that's a recipe for disaster, right? You, you don't need to be a sophisticated, uh, thinker about complex systems to see the hazard of immunizing, uh, a company from the harm of its own product at the same time that that product can only exist in the market if some other product that works better, uh, somehow fails to be noticed. So somehow, YouTube is doing the bidding of Merck and others...... whether it knows that that's what it's doing, I have no idea. I think this may, um, be another case of an autopilot that thinks it's doing the right thing because it's parroting the, uh, corrupt wisdom of The Who and the CDC. But The Who and the CDC have been wrong again and again in this pandemic. And the irony here is that with YouTube coming after me, well, my channel has been right where The Who and CDC have been wrong consistently over the whole pandemic. So, how is it that YouTube is censoring us because The Who and CDC disagree with us when, in fact, in past disagreements, we've been right and they've been wrong?

Lex Fridman 1:19:38 There's so much to talk about here. So, (sighs) I've heard this many times actually on the inside of YouTube and with colleagues that I've talked about, talked with is they kind of, in a very casual way, say their job is simply to, uh, slow or prevent the spread of misinformation. And they say it like that's an easy thing to do.

Bret Weinstein 1:20:06 (laughs)

Lex Fridman 1:20:07 Like, to know what is true or not is an easy thing to do. And so from the YouTube perspective, I think they basically outsource of w- of the, the task of knowing what is true or not to public institutions that on a basic Google search claim to be the arbiters of truth. So, if you were YouTube who are exceptionally profitable and exceptionally powerful in terms of, uh, controlling what people get to see or not, what would you do? Would you take a stand, a public stand against the WHO, Who, CDC? Or would you instead say, "You know what? Let's open the dam and let any video on anything fly?" H- w- what do you do here? If you, say, you were put ... If Bret Weinstein was put in charge (laughs) of YouTube for a month in this most critical of times where YouTube actually has incredible amounts of power to educate the populous, uh, to give power of knowledge to the populous such that they can reform institutions, what would you do? How would you run YouTube?

Bret Weinstein 1:21:26 Well, unfortunately-

Lex Fridman 1:21:28 (laughs)

Bret Weinstein 1:21:28 ... or fortunately this is actually quite simple. The founders, the American founders settled on a counterintuitive formulation that people should be free to say anything. They should be free from the government blocking them from doing so. They did not imagine that in formulating that right that most of what was said would be of high quality, nor did they imagine it would be free of harmful things. What they correctly reasoned was that the benefit of leaving everything so it can be said exceeds the cost which everyone understands to be substantial. What I would say is they could not have anticipated the impact, the centrality of platforms like YouTube, Facebook, Twitter, et cetera. If they had, they would not have limited the First Amendment as they did. They clearly understood that the power of the federal government was so great that it needed to be limited by granting, explicitly, the right of citizens to say anything. In fact, YouTube, Twitter, Facebook may be more powerful in this moment than the federal government of their worst nightmares could have been. The power that these entities have to control thought and to shift civilization is so great that we need to have those same protections. It doesn't mean that harmful things won't be said, but it means that nothing has changed about the cost-benefit analysis of building the right to censor. So, if I were running YouTube, the limit of what should be allowed is the limit of the law. All right? If what you are doing is legal, then it should not be YouTube's place to limit what gets said or who gets to hear it. That is between speakers and audience. Will harm come from that? Of course it will. But will net harm come from it? No, I don't believe it will. I believe that allowing everything to be said does allow a process in which better ideas do come to the fore and win out.

Lex Fridman 1:23:31 So, you believe that in the end when there's complete freedom to share ideas that truth will win out? So, what I've noticed just, uh, uh, as a brief side comment, that certain things become viral irregardless of their truth. I, I've noticed that things that are dramatic and/or funny, like, things that become memes are not, don't have to be grounded in truth. And so that, what worries me there is that we basically m- maximize for drama versus maximize for truth in a system where everything is f- free. And that's worrying in the time of emergency.

Bret Weinstein 1:24:16 Well, yes, it's all worrying in time of emergency-

Lex Fridman 1:24:18 (laughs)

Bret Weinstein 1:24:18 ... to be sure. But I want you to notice that what you've happened on is actually an analog for a much deeper and older problem. Human beings are the ... We are not a blank slate, but we are the blankest slate that nature has ever devised. And there's a reason for that, right? It's where our flexibility comes from. We have effectively, we are robots in which the, a large fraction of the cognitive capacity has been, or of the behavioral capacity has been offloaded to the software layer which gets written and rewritten over evolutionary time. That means, effectively, that much of what we are, in fact, the important part of what we are, is housed in the, in the cultural layer and the conscious layer and not in the hardware hard coding layer.... that layer is prone to make errors, right? And anybody who's watched a child grow up knows that children make absurd errors all the time, right? That's part of the process, as we were discussing earlier. It is also true that as you look across a field of people discussing things, a lot of what is said is pure nonsense. It's garbage. But the tendency of garbage to emerge and even to spread in the short term does not say that over the long term what sticks is not the valuable ideas. So, there is a high tendency for, uh, novelty to be created in the cultural space, but there's also a high tendency for it to go extinct. And you have to keep that in mind. It's not like the genome, right? Everything is happening at a much higher rate. Things are being created, they're being destroyed. And I can't say that, you know, I mean, obviously we've seen totalitarianism arise many times, and it's very destructive each time it does. So it's not like, hey, freedom to come up with any idea you want hasn't produced a whole lot of carnage. But the question is over time, does it produce-

Lex Fridman 1:26:19 Mm-hmm.

Bret Weinstein 1:26:19 ... more open, fairer, more decent societies? And I believe that it does. I can't prove it, but that does seem to be the pattern.

Lex Fridman 1:26:27 I, I believe so as well. The thing is, in the short term, freedom of speech, absolute freedom of speech can be quite destructive. But you, nevertheless, have to hold onto that because in the long term, I, I think you and I guess are optimistic in a sense that, um, good ideas will win out.

Bret Weinstein 1:26:51 I don't know how strongly I believe that it will work, but I will say I haven't heard a better idea. (laughs)

Lex Fridman 1:26:58 (laughs) Yeah.

Bret Weinstein 1:27:00 Um, I would also point out that there's something very significant in this question of the hubris involved in imagining that you're going to improve the discussion by censoring, which is the majority of concepts at the fringe are nonsense. That's automatic. But the heterodoxy at the fringe, which is indistinguishable at the beginning from the nonsense ideas, is the key to progress. So if you decide, "Hey, the fringe is 99% garbage, let's just get rid of it," right? "Hey, that's a strong win. We're getting rid of 99% garbage for 1% something or other."

Lex Fridman 1:27:40 Mm-hmm.

Bret Weinstein 1:27:40 And the point is, yeah, but that 1% something or other is the key. You're throwing out the key. And so that's what YouTube is doing. Frankly, I think at the point that it started censoring my channel, you know, in the immediate aftermath of this major reversal of la- for Lab Leak, it should've looked at itself and said, "Well, what the hell are we doing? Who are we censoring? We, we're censoring somebody who was just right, right? In a conflict with the very same people on whose behalf we are now censoring," right? That should've caused them to wake up.

Lex Fridman 1:28:09 So you said one approach if you were on YouTube is just basically let all videos go that do not violate the law.

Bret Weinstein 1:28:16 Well, I should fix that.

Lex Fridman 1:28:17 Yeah.

Bret Weinstein 1:28:18 Okay? I believe that that is the basic principle.

Lex Fridman 1:28:20 Yeah.

Bret Weinstein 1:28:20 Eric makes an excellent point about the distinction between ideas and personal attacks, doxing-

Lex Fridman 1:28:27 Right.

Bret Weinstein 1:28:27 ... these other things. So I agree, there is no value in allowing people to destroy each other's lives even if there's a technical, uh, legal defense for it. Now, how you draw that line? I don't know. But, you know, what I'm talking about is, yes, people should be free to traffic in bad ideas and they should be free to expose that the ideas are bad, and hopefully that process results in better ideas winning out.

Lex Fridman 1:28:50 Yeah, there's an interesting line between, you know, uh, like, uh, ideas, like the Earth is flat, which I believe you should not censor, and then, like, s- you start to encroach on, like, personal attacks. So not, not, you know, doxing, yes, but, like, not even getting to that. Like, there's a certain point where it's like that's no longer ideas. That's more... that's somehow not productive even if it's w- (laughs) wr- it feels like, uh, believing the Earth is flat is somehow productive because-

Bret Weinstein 1:29:22 Hmm.

Lex Fridman 1:29:22 ... maybe there's tiny percent chance it is? (laughs) You know, like, the, it just feels like personal attacks, it doesn't, um... well, you know, it's, I'm, I'm torn on this because there's assholes in this world, there's fraudulent people in this world, so sometimes personal attacks are useful to reveal that. Um, but there's a line you can cross.

Bret Weinstein 1:29:44 Mm-hmm.

Lex Fridman 1:29:44 Like, there's a comedy where people make fun of others. I think that's amazing, that's very powerful, and that's very useful even if it's painful. But then there's, like, once it gets to be, um... yeah, there's a certain line. It's a gray area where you cross where it's no longer p- uh, in any possible world productive. And that's a really weird gray area for YouTube to operate in, and it feels like it should be a crowdsource thing where people vote on it. But then again, do you trust the majority to vote on what is crossing the line and not? I mean, this is where this is really interesting on this particular, like, the scientific aspect of this. Do you think YouTube should take more of a stance, not censoring, but to actually have scientists within YouTube having these kinds of discussions and then be able to almost speak out in a transparent way, "This is what, we're going to let this video stand, but here's all these other opinions." Almost like take a more active role in, uh, its recommendation system in trying to present a full picture to you? Right now they're not. They're n- they're, the recommender systems are not human fine-tuned. They're all based on w- how you click and there's this clustering algorithms. They're not taking an active role on giving you the full spectrum of ideas in, in, in the space of science. They just censor or not.

Bret Weinstein 1:31:13 Well-At the moment, it's gonna be pretty hard to compel