Full episode with Eric Weinstein (Apr 2020): https://www.youtube.com/watch?v=rIAZJNe7YtE
Clips channel (Lex Clips): https://www.youtube.com/lexclips
Main channel (Lex Fridman): https://www.youtube.com/lexfridman
(more links below)

Podcast full episodes playlist:
https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4

Podcasts clips playlist:
https://www.youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41

Podcast website:
https://lexfridman.com/ai

Podcast on Apple Podcasts (iTunes):
https://apple.co/2lwqZIr

Podcast on Spotify:
https://spoti.fi/2nEwCF8

Podcast RSS:
https://lexfridman.com/category/ai/feed/

Eric Weinstein is a mathematician with a bold and piercing intelligence, unafraid to explore the biggest questions in the universe and shine a light on the darkest corners of our society. He is the host of The Portal podcast, a part of which, he recently released his 2013 Oxford lecture on his theory of Geometric Unity that is at the center of his lifelong efforts in arriving at a theory of everything that unifies the fundamental laws of physics.

Subscribe to this YouTube channel or connect on:
- Twitter: https://twitter.com/lexfridman
- LinkedIn: https://www.linkedin.com/in/lexfridman
- Facebook: https://www.facebook.com/lexfridman
- Instagram: https://www.instagram.com/lexfridman
- Medium: https://medium.com/@lexfridman
- Support on Patreon: https://www.patreon.com/lexfridman

Without naming names, can you tell the story of your struggle during your time at Harvard?

Maybe in a way that tells the bigger story of the struggle of young, bright minds that are trying to come up with big, bold ideas within the institutions that we're talking about?

I mean, in part, it starts with coffee, with a couple of Croatians in the math department at MIT.

And we used to talk about music and dance and math and physics and love and all this kind of stuff as Eastern Europeans love to and I ate it up.

And my friend, Gordana, who was an instructor in the MIT math department when I was a graduate student at Harvard, said to me, and I'm probably gonna do a bad version of her accent.

Will I see you tomorrow at the secret seminar?

I said, I'm not used to this style of humor, Gordon.

Eric, the secret seminar that your advisor is running.

You know, your advisor is running a secret seminar on this aspect, I think it was like the Churn-Simons invariant.

Not sure what the topic was again, but she gave me the room number and the time and she was like not cracking a smile.

I've never known her to make this kind of a joke.

I didn't want an advisor, but people said you have to have 1, so I took 1.

And I went to this room like 15 minutes early and there was not a soul inside it.

It was outside of the math department and it was still in the same building, the science center at Harvard.

And I sat there and I let 5 minutes go by, I let 7 minutes go by, 10 minutes go by, there was nobody.

I thought, okay, so this was all an elaborate joke.

And then like 3 minutes to the hour, this graduate student walks in and like sees me and does a double take.

And then I start to see the professors in geometry and topology start to file in and everybody's like very disconcerted that I'm in this room.

And finally, the person who was supposed to be my advisor walks in to the seminar and sees me and goes white as a ghost.

And I realized that the secret seminar is true, that the department is conducting a secret seminar on the exact topic that I'm interested in, not telling me about it.

And that these are the reindeer games that the Rudolph's of the department are not invited to.

And so then I realized, okay, I did not understand it.

And that became the beginning of an incredible odyssey in which I came to understand that the game that I had been sold about publication, about blind refereeing, about openness and scientific transmission of information was all a lie.

I came to understand that at the very top, there's a second system that's about closed meetings and private communications and agreements about citation and publication that the rest of us don't understand.

And that in large measure, that is the thing that I won't submit to.

And so when you ask me questions like, well, why wouldn't you feel good about, you know, talking to your critics?

Or why wouldn't you feel, The answer is, oh, you don't know.

Like if you stay in a nice hotel, you don't realize that there's an entire second structure inside of that hotel where like there's usually a worker's cafe in a resort complex that isn't available to the people who are staying in the hotel.

And then there are private hallways, inside the same, hotel that are parallel structures.

So that's what I found, which was in essence, just the way you can stay hotels your whole life and not realize that inside of every hotel is a second structure that you're not supposed to see as the guest.

There is a second structure inside of academics that behaves totally differently with respect to how people get dinged, how people get their grants taken away, how this person comes to have that thing named after them.

And by pretending that we're not running a parallel structure, I have no patience for that anymore.

So I got a chance to see how the game, how hardball is really played at Harvard.

And I'm now eager to play hardball back with the same people who played hardball with me.

do you think it's possible, so I call those people assholes, that's the technical term.

Do you think it's possible that that's just not the entire system but a part of the system?

Sort of that there's, you can navigate, you can swim in the waters and find the groups of people who do aspire to

The guy who rescued my PhD was 1 of the people who filed in to the secret seminar.

No, but I'm trying to make this point, which is this isn't my failure to correctly map these people.

You have a simplification that isn't gonna work.

I think, okay, ask what was the wrong term, I would say lacking of character.

I don't understand the exact dynamics of a secret seminar, but I think the right thing to do is to, I mean, to see individuals like you.

There might be a reason to have a secret seminar, but they should detect that an individual like you, a brilliant mind, who's thinking about certain ideas could be damaged by this.

The idea is we're going to sneak food to the children we want to survive.

Yeah, so that's highly problematic, and there should be people within that room.

I'm trying to say, this is the thing, the ball that's thrown but won't be caught.

The problem is, they know that most of their children won't survive.

You mean that the fact that the whole system is underfunded, that they naturally have to pick favorites.

live in a world which reached steady state at some level, let's say, in the early 70s.

And in that world, before that time, you have a professor like Norman Steenrod and you'd have 20 children that is graduate students and all of them would go on to be professors and all of them would want to have 20 children.

So you start like taking higher and higher powers of 20 and you see that the system could not, it's not just about money.

So the way it's supposed to work now is that we should shut down the vast majority of PhD programs and we should let the small number of truly top places populate, mostly teaching and research departments that aren't PhD producing.

We don't want to do that because we use PhD students as a labor force.

So the whole thing has to do with growth, resources, dishonesty, and in that world, you see all of these adaptations to a ruthless world where the key question is where are we gonna bury this huge number of bodies of people who don't work out?

So my problem was I wasn't interested in dying.

So you clearly highlighted there's aspects of the system that are broken, but as an individual, is your role to exit the system or just acknowledge that it's a game and win it?

My role is to survive and thrive in the public eye.

In other words, when you have an escapee of the system.

And that person says, you know, I wasn't exactly finished.

Let me show you that the theory of telomeres never got reported properly.

Let me show you that all of marginal economics is supposed to be redone with a different version of the differential calculus.

Let me show you that you didn't understand the self-dual Yang-Mills equations correctly in topology and physics, because they're in fact much more broadly found, and it's only the mutations that happen in special dimensions.

There are lots of things to say, but this particular group of people, like if you just take, where are all the gen X and millennial university presidents?

They're all, they're all in a holding pattern.

Now where, why in this story, you know, was it a, of telomeres was it an older professor and a younger graduate student?

It's this issue of what would be called interference competition.

So for example, orcas try to drown minky whales by covering their blowholes so that they suffocate because the needed resource is air.

They try to make sure that you can't be viable, that you need them, that you need their grants, you need to be zinged with overhead charges or fringe rates or all of the games that the locals love to play?

Well my point is, okay, what's the cost of this?

How many people died as a result of these interference competition games?

You know, when you take somebody like Douglas Prasher who did green fluorescent protein, and he drives a shuttle bus, right?

Because his grant runs out, and he has to give away all of his research, and all of that research gets a Nobel Prize, and he gets to drive a shuttle bus for $35,000 a year.

Do you mean their career, their dreams, their

Yeah, as an academic, Doug Prasher was dead for a long period of time.

Okay, so as a person who's escaped the system, can't you at this, because you also have in your mind a powerful theory that may turn out to be useful, maybe not.

Can't you also play the game enough with the children, so like publish,

If you told me that this would work, really what I wanna do, you see, is I would love to revolutionize a field with an h-index of 0.

Like we have these proxies that count how many papers you've written, how cited are the papers you've written.

That's interesting, sorry, what do you mean by field with an H index?

H index counts somehow how many papers have you gotten that get so many citations.

Like for example I don't have an advisor for my PhD, but I have to have an advisor as far as something called the math genealogy project that tracks who advised who, who advised whom down the line.

So I am my own advisor, which sets up a loop, right?

How many students do I have an infinite number?

Or descendants, they don't want to have that story.

So I have to be, I have to have formal advisor, Raoul Bott, and my Wikipedia entry, for example, says that I was advised by Raoul Bott, which is not true.

So you get fit into a system that says, well, we have to know what your H index is.

We have to know, you know, where are you a professor if you want to apply for a grant?

What I'm trying to do is to impart, to show all of this is nonsense.

This is proxy BS that came up in the institutional setting.

And right now it's important for those of us who are still vital, like Elon, it would be great to have Elon as a professor of physics and engineering.

It seems ridiculous to say, but just as a shot, just as a shot in the arm, yeah.

You know, like be great to have Elon at Caltech even 1 day a week.

We need to know what you're gonna do before we take you on the show.

Well, I don't wanna tell you what I'm gonna do.

Do you think you need to be able to dance the dance a little bit?

Here's where the place that it goes South is, there's like a set of questions that get you into this more adversarial stuff.

And you've in fact asked some of those more adversarial questions, the setting.

And they're not things that are necessarily aggressive, but they're things that are making assumptions.

So when you have a question, it's like, Lex, are you avoiding your critics?

It's just like, okay, well, why did you frame that that way?

Or the next question would be, do you think that you should have a special exemption and that you should have the right to break rules and everyone else should have to follow them?

It doesn't really come out of anything meaningful.

It's just like we feel we're supposed to ask that of the other person to show that we're not captured by their madness?

That's not the real question you want to ask me.

If you want to get really excited about this, you want to ask, do you think this thing is right?

Do you think that it's going to be immediately seen to be right?

I think it's going to have an interesting fight and it's gonna have an interesting evolution.

And what do you hope to do with it in non-physical terms?

Gosh, I hope it revolutionizes our relationship well with people outside of the institutional framework and it re-inflicts us into the institutional framework where we can do the most good to bring the institutions back to health.

You know, it's like, these are positive uplifting questions.

If you had Frank Wilczek, you wouldn't say, Frank, let's be honest, you have done very little with your life after the original huge show that you used to break under the physics.

Like, we weirdly ask people different questions based upon how they sit down.

But you have to understand that, so here's the thing.

these days a large number of emails from people with the equivalent of a theory of everything for AGI.

And I use my own radar, BS radar, to detect unfairly perhaps whether they're full of shit or not.

I love where you're going with this, by the way.

And my concern I often think about is there's elements of brilliance in what people write to me.

And I'm trying to right now as you made it clear, the kind of judgments and assumptions we make, how am I supposed to deal with you who are an outsider of the system and think about what you're doing?

Because my radar is saying you're not full of shit.

You're- but I'm also not completely outside of the system.

You've actually got all the credibility that you're supposed to get, all the nice little stamps of approval.

I mean, it's hard to put into words exactly why you sound, whether your theory turns out to be good or not, you sound like a special human being.

So, but what am I supposed to do with that flood of emails from AGI folks?

And I would like to systemize that, I don't know.

Look, you know, when you're talking to people, you very quickly can surmise, am I claiming to be a physicist?

No, I say it every turn, I'm not a physicist, right?

When you say something about bundles, you say, well, can you explain it differently?

I'm pushing around on this area, that lever over there.

I'm trying to find something that we can play with and engage.

And another thing is that I'll say something at scale.

So if I was saying completely wrong things about bundles on the Joe Rogan program, you don't think that we wouldn't hear a crushing chorus?

And you know, same thing with geometric unity.

So I put up this video from this Oxford lecture.

I understand that it's not a standard lecture, but you haven't heard the most brilliant people in the field say, well, this is obviously nonsense.

And they're gonna hide behind, well he hasn't said enough detail, where's the paper?

I've seen the criticism, I've gotten the same kind of criticism, I've published a few things and like especially stuff related to Tesla, that we did studies on Tesla vehicles and the kind of criticism I've gotten was showed that they're completely.

Oh right, like the guy who had Elon Musk on

his program twice is gonna give us an accurate assessment.

Like without actually ever addressing the content.

You know, Lex, I think that in part, you're trying to solve a puzzle that isn't really your puzzle.

You don't know whether the theory is gonna work or not.

And you know that it's not coming out of somebody who's coming out of left field.

There's enough that's new and creative and different in other aspects where you can check me, that your real concern is, are you really telling me that when you start breaking the rules, you see the system for what it is and it's become really vicious and aggressive?

And I had to break the rules in part because of learning issues, because I came into this field with a totally different set of attributes.

My profile just doesn't look like anybody else's remotely.

But as a result, what that did is it showed me, is the system true to its own ideals or does it just follow these weird procedures and then when you take it off the rails, it behaves terribly.

And that's really what my story I think does is it just says, well, he completely takes the system into new territory where it's not expecting to have to deal with somebody with these confusing sets of attributes.

And I think what he's telling us is he believes it behaves terribly.

Now, if you take somebody with perfect standardized tests and a winner of math competitions and you put them in a PhD program, they're probably gonna be okay.

I'm not saying that the system breaks down for everybody under all circumstances.

I'm saying when you present the system with a novel situation, at the moment, it will almost certainly break down with probability approaching 100%.

But to me, the painful and the tragic thing is it, sorry to bring out my motherly instinct, but it feels like it's too much, it could be too much of a burden to exist outside the system.

first of all, I've got a podcast that I kind of like.

I have a life which has more interesting people passing through it than I know what to do with.

And they haven't managed to kill me off yet.


Speaker 0: Without naming names, can you tell the story of your struggle during your time at Harvard? Maybe in a way that tells the bigger story of the struggle of young, bright minds that are trying to come up with big, bold ideas within the institutions that we're talking about?

Speaker 1: You can start. I mean, in part, it starts with coffee, with a couple of Croatians in the math department at MIT. And we used to talk about music and dance and math and physics and love and all this kind of stuff as Eastern Europeans love to and I ate it up. And my friend, Gordana, who was an instructor in the MIT math department when I was a graduate student at Harvard, said to me, and I'm probably gonna do a bad version of her accent. There we go.

Will I see you tomorrow at the secret seminar? And I said, What secret seminar? Eric, don't joke. I said, I'm not used to this style of humor, Gordon. Eric, the secret seminar that your advisor is running.

I said, what are you talking about? Ha ha ha. You know, your advisor is running a secret seminar on this aspect, I think it was like the Churn-Simons invariant. Not sure what the topic was again, but she gave me the room number and the time and she was like not cracking a smile. I've never known her to make this kind of a joke.

And I thought this was crazy. And I was trying to have an advisor. I didn't want an advisor, but people said you have to have 1, so I took 1. And I went to this room like 15 minutes early and there was not a soul inside it. It was outside of the math department and it was still in the same building, the science center at Harvard.

And I sat there and I let 5 minutes go by, I let 7 minutes go by, 10 minutes go by, there was nobody. I thought, okay, so this was all an elaborate joke. And then like 3 minutes to the hour, this graduate student walks in and like sees me and does a double take. And then I start to see the professors in geometry and topology start to file in and everybody's like very disconcerted that I'm in this room. And finally, the person who was supposed to be my advisor walks in to the seminar and sees me and goes white as a ghost.

And I realized that the secret seminar is true, that the department is conducting a secret seminar on the exact topic that I'm interested in, not telling me about it. And that these are the reindeer games that the Rudolph's of the department are not invited to. And so then I realized, okay, I did not understand it. There's a parallel department. And that became the beginning of an incredible odyssey in which I came to understand that the game that I had been sold about publication, about blind refereeing, about openness and scientific transmission of information was all a lie.

I came to understand that at the very top, there's a second system that's about closed meetings and private communications and agreements about citation and publication that the rest of us don't understand. And that in large measure, that is the thing that I won't submit to. And so when you ask me questions like, well, why wouldn't you feel good about, you know, talking to your critics? Or why wouldn't you feel, The answer is, oh, you don't know. Like if you stay in a nice hotel, you don't realize that there's an entire second structure inside of that hotel where like there's usually a worker's cafe in a resort complex that isn't available to the people who are staying in the hotel.

And then there are private hallways, inside the same, hotel that are parallel structures. So that's what I found, which was in essence, just the way you can stay hotels your whole life and not realize that inside of every hotel is a second structure that you're not supposed to see as the guest. There is a second structure inside of academics that behaves totally differently with respect to how people get dinged, how people get their grants taken away, how this person comes to have that thing named after them. And by pretending that we're not running a parallel structure, I have no patience for that anymore. So I got a chance to see how the game, how hardball is really played at Harvard.

And I'm now eager to play hardball back with the same people who played hardball with me. Let me ask 2 questions on this. So 1,

Speaker 0: do you think it's possible, so I call those people assholes, that's the technical term. Do you think it's possible that that's just not the entire system but a part of the system? Sort of that there's, you can navigate, you can swim in the waters and find the groups of people who do aspire to

Speaker 1: the openness. The guy who rescued my PhD was 1 of the people who filed in to the secret seminar.

Speaker 0: Right, but are there people who are

Speaker 1: outside of this? Is he an asshole?

Speaker 0: Well, yes, I was a bad.

Speaker 1: No, but I'm trying to make this point, which is this isn't my failure to correctly map these people. It's yours. You have a simplification that isn't gonna work.

Speaker 0: I think, okay, ask what was the wrong term, I would say lacking of character.

Speaker 1: What would you have had these people do? Why did they do this? Why have a secret seminar?

Speaker 0: I don't understand the exact dynamics of a secret seminar, but I think the right thing to do is to, I mean, to see individuals like you. There might be a reason to have a secret seminar, but they should detect that an individual like you, a brilliant mind, who's thinking about certain ideas could be damaged by this.

Speaker 1: I don't think that they see it that way. The idea is we're going to sneak food to the children we want to survive.

Speaker 0: Yeah, so that's highly problematic, and there should be people within that room.

Speaker 1: I'm trying to say, this is the thing, the ball that's thrown but won't be caught. The problem is, they know that most of their children won't survive. And they can't say that.

Speaker 0: I see, sorry to interrupt. You mean that the fact that the whole system is underfunded, that they naturally have to pick favorites. They

Speaker 1: live in a world which reached steady state at some level, let's say, in the early 70s. And in that world, before that time, you have a professor like Norman Steenrod and you'd have 20 children that is graduate students and all of them would go on to be professors and all of them would want to have 20 children. Right? So you start like taking higher and higher powers of 20 and you see that the system could not, it's not just about money. The system couldn't survive.

So the way it's supposed to work now is that we should shut down the vast majority of PhD programs and we should let the small number of truly top places populate, mostly teaching and research departments that aren't PhD producing. We don't want to do that because we use PhD students as a labor force. So the whole thing has to do with growth, resources, dishonesty, and in that world, you see all of these adaptations to a ruthless world where the key question is where are we gonna bury this huge number of bodies of people who don't work out? So my problem was I wasn't interested in dying.

Speaker 0: So you clearly highlighted there's aspects of the system that are broken, but as an individual, is your role to exit the system or just acknowledge that it's a game and win it?

Speaker 1: My role is to survive and thrive in the public eye. In other words, when you have an escapee of the system. Like yourself. Such as. And that person says, you know, I wasn't exactly finished.

Let me show you a bunch of stuff. Let me show you that the theory of telomeres never got reported properly. Let me show you that all of marginal economics is supposed to be redone with a different version of the differential calculus. Let me show you that you didn't understand the self-dual Yang-Mills equations correctly in topology and physics, because they're in fact much more broadly found, and it's only the mutations that happen in special dimensions. There are lots of things to say, but this particular group of people, like if you just take, where are all the gen X and millennial university presidents?

Speaker 0: All right.

Speaker 1: Okay. They're all, they're all in a holding pattern. Now where, why in this story, you know, was it a, of telomeres was it an older professor and a younger graduate student? It's this issue of what would be called interference competition. So for example, orcas try to drown minky whales by covering their blowholes so that they suffocate because the needed resource is air.

Okay, well, what do the universities do? They try to make sure that you can't be viable, that you need them, that you need their grants, you need to be zinged with overhead charges or fringe rates or all of the games that the locals love to play? Well my point is, okay, what's the cost of this? How many people died as a result of these interference competition games? You know, when you take somebody like Douglas Prasher who did green fluorescent protein, and he drives a shuttle bus, right?

Because his grant runs out, and he has to give away all of his research, and all of that research gets a Nobel Prize, and he gets to drive a shuttle bus for $35,000 a year.

Speaker 0: What do you mean by died? Do you mean their career, their dreams, their

Speaker 1: passions? Yeah, as an academic, Doug Prasher was dead for a long period of time.

Speaker 0: Okay, so as a person who's escaped the system, can't you at this, because you also have in your mind a powerful theory that may turn out to be useful, maybe not. Can't you also play the game enough with the children, so like publish,

Speaker 1: but also. If you told me that this would work, really what I wanna do, you see, is I would love to revolutionize a field with an h-index of 0. Like we have these proxies that count how many papers you've written, how cited are the papers you've written. All this is nonsense.

Speaker 0: That's interesting, sorry, what do you mean by field with an H index? So a totally new field.

Speaker 1: H index counts somehow how many papers have you gotten that get so many citations. Let's say H index undefined. Like for example I don't have an advisor for my PhD, but I have to have an advisor as far as something called the math genealogy project that tracks who advised who, who advised whom down the line. So I am my own advisor, which sets up a loop, right? How many students do I have an infinite number?

Or descendants, they don't want to have that story. So I have to be, I have to have formal advisor, Raoul Bott, and my Wikipedia entry, for example, says that I was advised by Raoul Bott, which is not true. So you get fit into a system that says, well, we have to know what your H index is. We have to know, you know, where are you a professor if you want to apply for a grant? It makes all of these assumptions.

What I'm trying to do is to impart, to show all of this is nonsense. This is proxy BS that came up in the institutional setting. And right now it's important for those of us who are still vital, like Elon, it would be great to have Elon as a professor of physics and engineering. Yeah. Right.

It seems ridiculous to say, but just as a shot, just as a shot in the arm, yeah. You know, like be great to have Elon at Caltech even 1 day a week. Yeah. 1 day a month. Okay.

Well, why can't we be in there? It's the same reason. Well why can't you be on The View? Why can't you be on Bill Maher? We need to know what you're gonna do before we take you on the show.

Well, I don't wanna tell you what I'm gonna do.

Speaker 0: Do you think you need to be able to dance the dance a little bit?

Speaker 1: I can dance the dance fine.

Speaker 0: To be on the view.

Speaker 1: Oh, come on.

Speaker 0: So you can, yeah, you do.

Speaker 1: I can do that fine. Here's where the place that it goes South is, there's like a set of questions that get you into this more adversarial stuff. And you've in fact asked some of those more adversarial questions, the setting. And they're not things that are necessarily aggressive, but they're things that are making assumptions. So when you have a question, it's like, Lex, are you avoiding your critics?

It's just like, okay, well, why did you frame that that way? Or the next question would be, do you think that you should have a special exemption and that you should have the right to break rules and everyone else should have to follow them? Like that question I find enervating. It doesn't really come out of anything meaningful. It's just like we feel we're supposed to ask that of the other person to show that we're not captured by their madness?

That's not the real question you want to ask me. If you want to get really excited about this, you want to ask, do you think this thing is right? Yeah, weirdly I do. Do you think that it's going to be immediately seen to be right? I don't.

I think it's going to have an interesting fight and it's gonna have an interesting evolution. And what do you hope to do with it in non-physical terms? Gosh, I hope it revolutionizes our relationship well with people outside of the institutional framework and it re-inflicts us into the institutional framework where we can do the most good to bring the institutions back to health. You know, it's like, these are positive uplifting questions. If you had Frank Wilczek, you wouldn't say, Frank, let's be honest, you have done very little with your life after the original huge show that you used to break under the physics.

Like, we weirdly ask people different questions based upon how they sit down. Yeah, that's very strange, right? But you have to understand that, so here's the thing. I get

Speaker 0: these days a large number of emails from people with the equivalent of a theory of everything for AGI. Yeah. And I use my own radar, BS radar, to detect unfairly perhaps whether they're full of shit or not.

Speaker 1: Because. I love where you're going with this, by the way.

Speaker 0: And my concern I often think about is there's elements of brilliance in what people write to me. And I'm trying to right now as you made it clear, the kind of judgments and assumptions we make, how am I supposed to deal with you who are an outsider of the system and think about what you're doing? Because my radar is saying you're not full of shit. You're- but I'm also not completely outside of the system. That's right, you've danced beautifully.

You've actually got all the credibility that you're supposed to get, all the nice little stamps of approval. Not all, but a large enough amount. I mean, it's hard to put into words exactly why you sound, whether your theory turns out to be good or not, you sound like a special human being.

Speaker 1: I appreciate that and thank you for

Speaker 0: your wonderful saying. In a good way, right?

Speaker 1: No, no, right.

Speaker 0: So, but what am I supposed to do with that flood of emails from AGI folks? Why do

Speaker 1: I sound different?

Speaker 0: I don't know. And I would like to systemize that, I don't know.

Speaker 1: Look, you know, when you're talking to people, you very quickly can surmise, am I claiming to be a physicist? No, I say it every turn, I'm not a physicist, right? When you say something about bundles, you say, well, can you explain it differently? I'm pushing around on this area, that lever over there. I'm trying to find something that we can play with and engage.

And another thing is that I'll say something at scale. So if I was saying completely wrong things about bundles on the Joe Rogan program, you don't think that we wouldn't hear a crushing chorus? Yes. And you know, same thing with geometric unity. So I put up this video from this Oxford lecture.

I understand that it's not a standard lecture, but you haven't heard the most brilliant people in the field say, well, this is obviously nonsense. They don't know what to make of it. And they're gonna hide behind, well he hasn't said enough detail, where's the paper?

Speaker 0: And where's the paper? I've seen the criticism, I've gotten the same kind of criticism, I've published a few things and like especially stuff related to Tesla, that we did studies on Tesla vehicles and the kind of criticism I've gotten was showed that they're completely. Oh right, like the guy who had Elon Musk on

Speaker 1: his program twice is gonna give us an accurate assessment.

Speaker 0: Yeah, exactly, exactly.

Speaker 1: It's just very low level.

Speaker 0: Like without actually ever addressing the content.

Speaker 1: You know, Lex, I think that in part, you're trying to solve a puzzle that isn't really your puzzle. I think you know that I'm sincere. You don't know whether the theory is gonna work or not. And you know that it's not coming out of somebody who's coming out of left field. Like the story makes sense.

There's enough that's new and creative and different in other aspects where you can check me, that your real concern is, are you really telling me that when you start breaking the rules, you see the system for what it is and it's become really vicious and aggressive? And the answer is yes. And I had to break the rules in part because of learning issues, because I came into this field with a totally different set of attributes. My profile just doesn't look like anybody else's remotely. But as a result, what that did is it showed me, is the system true to its own ideals or does it just follow these weird procedures and then when you take it off the rails, it behaves terribly.

And that's really what my story I think does is it just says, well, he completely takes the system into new territory where it's not expecting to have to deal with somebody with these confusing sets of attributes. And I think what he's telling us is he believes it behaves terribly. Now, if you take somebody with perfect standardized tests and a winner of math competitions and you put them in a PhD program, they're probably gonna be okay. I'm not saying that the system breaks down for everybody under all circumstances. I'm saying when you present the system with a novel situation, at the moment, it will almost certainly break down with probability approaching 100%.

Speaker 0: But to me, the painful and the tragic thing is it, sorry to bring out my motherly instinct, but it feels like it's too much, it could be too much of a burden to exist outside the system. Maybe, but

Speaker 1: first of all, I've got a podcast that I kind of like. I've got amazing friends. I have a life which has more interesting people passing through it than I know what to do with. And they haven't managed to kill me off yet. So, so far so good.

See all Lex Fridman transcripts on Youtube

Eric Weinstein's Harvard Story - The System Breaks Down in Novel Situations | AI Podcast Clips