T248. Werbos, Computers, and God

Project Save the World Podcast / Talk Show Episode Number: 248
Panellists: Paul Werbos
Host: Metta Spencer

Date Aired:  17 May 2021
Date Transcribed and Verified:  6 June 2021
Transcription: Otter.ai
Transcription Review and Edits: Adam Wynne

Synopsis: Paul Werbos’s dissertation 50 years ago is the basis for the “New AI.” He wavers between believing in Einstein and in David Deutsch’s formulas for the universe. 

Metta Spencer 

Hi, I’m Metta Spencer. Today, I am really gonna have fun. I don’t know about you, because this may be hard work for a lot of people, including me. I just spent a good part of my weekend watching videos by a guy who is maybe one of the most interesting people I have ever met. And I hardly understood a word that he was saying, but I couldn’t stop watching those videos. They were so interesting. And so important. Because this is a man who thinks a lot about a lot of different things. Trouble is he’s a mathematician. I am most decidedly am not. I wish I were. But it means there are real problems in communicating and understanding him. Now, this is Paul Werbos. Where are you, Paul? Are you in Philadelphia? 

Paul Werbos  

Actually, I’m in Arlington, Virginia.

Metta Spencer  

Right. Got it.

Paul Werbos  

I’m in a house that George Washington used to own. 

Metta Spencer  

No kidding. Really?

Paul Werbos  

 Yeah. It’s a great neighborhood.

Metta Spencer  

Wow. That’s a story right there. Okay. Now, the thing is that he is a mathematician. He’s famous for something called backpropagation. What is it, Paul?

Paul Werbos  

Backpropagation. 

Metta Spencer  

Backpropagation, which I promise you if you read it, you wouldn’t understand a word. I did. I printed it out and looked at it and couldn’t understand it.

Paul Werbos  

There are people in Washington who think they know what artificial intelligence is. And it makes me laugh, because even these heads of government agencies talk about the new AI and they don’t know what it is. A lot of what it is an advanced neural network technology, doing things like what brains do. And what they do with neural nets is all based on an algorithm I developed 50 years ago. And people in the field know about it.

Metta Spencer  

Okay, I’ve seen that and that is now the basis for artificial intelligence, more or less, right now? 

Paul Werbos  

New AI. They call it the new AI.

Metta Spencer  

Alright. Well, I don’t think we can talk about that, because I really couldn’t keep up with you there. But what happened is, when we were doing this setup, we started talking about God. Oh, naturally, what else would you talk about when you’re trying to set up your computer? So anyway, Paul has been thinking about God. And so have I and we talked about whether God really does play dice with the universe. And I said: Well, if he doesn’t… I mean if he does, he’s not God, because God’s supposed to know everything and if you got dice going around that you don’t know what it’s going to do, then you can’t possibly be God. On the other hand, my opinion is, I sort of want to believe that I have something like free will. And if I have free will, then that means God’s not in charge of everything, or doesn’t know what I’m going to do or something. Therefore, I don’t want God to know and be completely all powerful and all knowing. Therefore, I’m a little bit inconsistent. And whereupon Paul said: Yea, I’ve solved that problem. I’ve been thinking about that a long time. And I have the answer. And just then I had to set this thing up to get you guys in on it. So, I promised him that we would finish this conversation and you can listen in, and then we’ll start a different conversation about the things that probably most people might know a little bit about or think about more, Like climate. Okay. Okay, Paul, set me straight. In your 50 years of thinking about this subject, how did you solve it?

Paul Werbos  

It took a long time for me to figure out how these pieces fit together. I would say from 1967 until a year ago, I was really upset by the contradictions I saw in what I knew and in my life. In 1967, I really believed in a hard-core realistic view of how the universe works. I even met some friends of Albert Einstein when I was taking courses in Princeton when I was in high school. And what I read about Einstein sounded totally nutty. But then these guys in Princeton set me straight and they explained to me how Einstein’s view of the universe really works. And he doesn’t believe their dice. He believes it’s all governed by a certain kind of equation. And I believe that until 1967. Until 1967, I used to think that my Druid Catholic mother was a little crazy, because she would talk to the plants. 

Metta Spencer  

Druid Mother? You had a Druid Mother? 

Paul Werbos  

She had dreams about the future. I said: That can’t be right. I just refuse to believe it. She would show it to me. She would tell me. I would refuse to believe it. And then in 1967, something happened to me. And at first, I thought: Am I going crazy? And the next thought is: I can’t excuse these experiences. And from 1967 until a year ago, it’s a contradiction. I believed in one thing in physics., but in real life things seemed to be weirder and weirder and it’s hard to make sense of them. It took a long time for me to figure out how to resolve these inconsistencies. But the dice in the universe is not the biggest thing that I worried about. That actually was never such a big problem. Because Albert Einstein had this feeling, of course, the laws of the universe must be fixed. Equations must tell you exactly what will happen next. And he said to his friend, John von Neumann at Princeton. John von Neumann is sort of my hero. I grew up…. my mother practically worshipped John von Neumann. He was a really interesting guy. Was your mother mathematician also?  He was a mathematician.

Metta Spencer  

No, was your mother a mathematician?

Paul Werbos  

She was proud she won the school prize in mathematics. But you know, in those days, it wasn’t so easy to be a mathematician if you were a woman.

Metta Spencer  

Okay.

Paul Werbos  

Some people say von Neumann was the greatest mathematician of the 20th century. When I grew up, everybody said that. But a lot of mathematicians were embarrassed by von Neumann. Because he wasn’t pure. He ran around trying to save the world like you. He ran around talking to people at the Department of Defense and telling them how they should change their strategies. And he even went out to Los Alamos. He was one of the people who created the atomic bomb. And he wrote basic books on how does quantum mechanics work? How does life work? How does the mind work? He invented the field of neural nets. He was an unbelievable guy. But a lot of pure mathematicians said he’s not pure. And he kind of played with all these dirty people. And so, they decided that a different guy called Hilbert… they now cite Hilbert as the top mathematician of the 20th century. 

Metta Spencer  

Hilbert? H-I-L-B-E-R-T?

Paul Werbos  

Yeah. And he did great stuff. I mean, that’s why they call him number one. But Hilbert was, let’s say the top pure mathematician. Von Neumann was proving lots of theorems, but he connected to the real world. So, I really love von Neumann. I’ve read so many books about him. He was just fascinating guy.

Metta Spencer  

Except, I’m not sure I would love him for working at Los Alamos.

Paul Werbos  

There were people who hated him a lot. And there are a lot of old stories about what happened in Los Alamos. I even met a guy who worked there himself and had old home movies that he showed me of what really happened in Los Alamos that you never heard of. Those were fun home movies to watch. But, coming back, von Neumann and Einstein were more or less friends. They respected each other. They had offices in the same building in Princeton, where they told me about this, they had mathematics teas there. So, this is a very familiar place. And in that building, you can see where Einstein and von Neumann talked to each other, and a few other important people. And when Einstein told von Neumann God does not play dice with the universe, von Neumann listened. He paid a lot of attention. But he wouldn’t make up his mind until there was real data. And von Neumann would say: There might be randomness in the universe, there might not be. Let’s find out empirically. Let’s build theories. Let’s test them against reality. Maybe you’re right, my friend Einstein, but maybe you’re not right. Let’s find out. Let’s do the math. Let’s do the experiments. Let’s find out.

Metta Spencer  

See, I can’t imagine what conceivable experiment you could do that would prove one way or the other.

Paul Werbos  

Yeah. And in fact, there’s a famous guy named Karl Popper who used to say: You can’t prove that a theory is true. Popper is very famous in science.

Metta Spencer  

Listen, not only is he very famous, but he was my teacher.

Paul Werbos  

Really? Wow!

Metta Spencer  

Well, I can elaborate that and make it sound much more than it was, but I was his student in Berkeley for one term. I used to go to his office and talk to him.

Paul Werbos  

So, we may both believe there is something like magic in real human life. This is the kind of thing that made me wonder, because I too, I wasn’t such a student, but I did audit one of his classes. And the minute he saw me, he asked me to address the class, because he knew I could talk about a subject that I knew about that he wanted the class to hear about. That was in the London School of Economics. 

Metta Spencer  

Really?

Paul Werbos  

Yeah. 

Metta Spencer  

You were at LSE? 

Paul Werbos  

For the master’s degree.

Metta Spencer  

Yeah, I knew that and that puzzled me because I thought: How in the hell do you mix economics with optics? 

Paul Werbos  

Well, mathematics is the key to all of them.

Metta Spencer  

I believe you, but I cannot imagine it. But anyway, go on. 

Paul Werbos  

Actually, that’s important.

Metta Spencer  

Popper’s the most influential person, in my thinking, I would say. It’s like it was money in the bank. You know, the first class, I went to, the auditorium was full. The second-class session, it was half full. Then the third class it was finally like the first two rows. And of course, I stayed there. I can tell you a lot of stories about that course, it was wonderful. But the thing is that I realized why people dropped out is because it sounds so simple. He made it sound obvious and trivial and banal and, you know, like, anybody would understand this, it was plain. And, I mean, there wasn’t, you know, there wasn’t anything to argue about, it was so clear. But I had the feeling later, you know, it was like putting money in the bank every time in graduate school, that I would get into an intellectual problem that I couldn’t solve. I’d go to the bank and withdraw some from my Popper account, and that would solve it. 

Paul Werbos  

That’s a good way of putting it. 

Metta Spencer  

Yeah, he was so clear that it looked obvious, but of course it was obvious only when you saw the problem that was not obvious. 

Paul Werbos  

Yes. One of the reasons that I loved working for the old National Science Foundation is that you have got to have really close conversations with people in all different fields from all over the world. And that was a learning experience. But one thing I’ve learned is that even the most brilliant people on Earth screw up a few things. They become rigid and they get locked in a box. And more often than not, they could get out of the box if only they remembered something they thought was easy to understand back when they were in kindergarten or in a class like Popper’s. And when you say Popper said things that sounded simple, how many PhDs screw up because they forgot something simple? And when you go to the bank at the Popper account, you are remembering something simple. And I believe Popper said a lot of simple things that are very powerful if you would remember them at the right time. Because he’s not alone. People say that about Jesus, you know, he said some important things that people don’t always remember what he said.

Metta Spencer  

I never know what or how to use what he said. That’s another matter. 

Paul Werbos  

Yeah. 

Metta Spencer  

Okay, well, I deflected you. I’m sorry. Get me onto Popper and I ruin the whole conversation. Okay, back to where you were straightening out your confusion.

Paul Werbos  

Okay, so in 1967, I thought I had nice mathematical theories about how the brain works and then this personal experience became confusing. And then everybody told me Einstein’s stuff didn’t work anymore because of quantum mechanics. And I read some of the quantum stuff. And just as Schrodinger thought it didn’t make a lot of sense, Einstein thought it didn’t make a lot of sense, and I thought it didn’t make a lot of sense. What I really wanted to do for my PhD, was figure out how to do what Einstein wanted to do. That is, come up with a mathematical explanation for why quantum mechanics works as a statistical description of a universe that’s really governed by Einstein rules at the bottom. And that’s a hard mathematical problem. When I was in graduate school, I said: I really want to do that, but it’s too hard. I can’t do it in 3 years. So, I’ll do something easy instead for my PhD thesis. So, for my PhD thesis, I decided okay, I’ll explain how the human brain works. I already know that from the work that I already did. And when I showed it to the Harvard faculty, they said: Well, that’s interesting, but this is too complicated. Take a piece of it. And we’ll give you a PhD for a piece of it. So, I gave them a piece of it that was 400 pages long. And right now, that is the source of what they call backpropagation and the new AI. The new AI is Grade 1 of 12 grades. The new AI is much bigger than anything these guys know.

Metta Spencer  

What have you done with the other pieces?

Paul Werbos  

Ah, well, the beauty of working at the National Science Foundation, I got to work with industry people, frontline people, who are ready to start applying advanced technologies that are beyond what other people were working on. And I found out a week ago, this ResearchGate place, says that the paper that’s most cited this week, was not the backpropagation, but a slightly newer paper: Artificial Intelligence from Neural Intelligence to Biological Intelligence. It was an SPIE paper, it’s up there, they say it was the highly cited one. And it reviews 20 areas of real-world application of more advanced neural technologies that they don’t teach you in school even now. So for example: Ford was a great place. There was a time when Ford was the number one industry place in the whole world, for using real brain-like AI to solve real problems. And they used it for controlling cars. 

Metta Spencer  

When was that?

Paul Werbos  

In 1998, Businessweek published an article where the President of Ford said we can meet the new clean air standards, which the other guys say are impossible. They think it’s impossible. But Ford has a better idea, he said, in Businessweek. And he said: We know how to solve idle control, fuel-air mixtures, and NOx emissions. We have a way to solve these problems by using a new kind of recurrent neural network. And I knew, I worked with the people at Ford, who did that. I talked to them many times every year for many, many years until they retired. But yeah, Ford was already using advanced recurrent neural networks to improve air quality at a time when computer scientists were saying neural networks will never do anything. There were textbooks on AI by the leaders of computer science – I could tell you the names, but it wouldn’t be polite – leading textbooks on AI by the leaders of computer science, which say neural nets will never do XYZ. And they were already doing it in Ford cars. And they even published papers, but they didn’t advertise so heavily what they were doing. But they published papers explaining how they use neural nets to solve air quality problems. And that was more than 20 years ago that they did it. 

Metta Spencer  

Well, then they quit? They’re not doing it anymore.

Paul Werbos  

Well, the 2 key guys have a problem that you and I have. They got older. And you know, as you get older things happen. And these guys retired, of course, they not only got older, they also had an income so they could do what they wanted to do. They are still around so far as I know, I sure hope they’re still around.

Metta Spencer  

I think you should go talk to them. 

Paul Werbos  

Sometimes I do. Sometimes they even come talk to me. 

Metta Spencer  

Really? Okay We got off track. Where did this happen? Now you had some experience. that clashed with something else in your belief system and you couldn’t make them jive? 

Paul Werbos  

 Yeah.

Metta Spencer  

 I don’t know whether you want to tell me what that experience was or whether it makes any difference? I’m fascinated. But still, the question is: last year, you got to come together?

Paul Werbos  

Yes. And to be honest, there are bits and pieces of it. Some pieces came earlier, some pieces came later. There are a whole lot of inconsistencies between spiritual life to where you can even use the word God and not feel you’re being crazy; and hardcore Einstein view of the universe. That’s a big distance to try to bridge that distance. And there are a lot of little bridges you have to build to reconcile these 2 different ways of thinking.

Metta Spencer  

Well hold on. I mean, I don’t think there’s any incompatibilities between Einstein and God.

Paul Werbos  

It depends on what you mean by the word. God. Okay? So, Einstein’s view the word God. I saw a book by Michio Kaku. A famous popular book, but there’s this popular writer called Michio Kaku. People say he’s a string theorist. 

Metta Spencer  

The Japanese guy? 

Paul Werbos  

Yes. Japanese-American. He was teaching at MIT. He gave a talk about this great project, which my older daughter did when she was at MIT. But yeah, Kaku gives a lot of talks and his book says what I think Einstein would have said about God, frankly, because I think you and I take the word God more seriously now than Einstein did back in those days. When I was young, I didn’t take it seriously. When I was young, before 1967, I would have said, plain and simple, I am an atheist. This is total nonsense. I know why it’s total nonsense. It’s all imaginary garbage. That’s what I would have said before 1967. 

Metta Spencer  

Alright. And then you had this experience, which you’re not telling me about, which is okay. 

Paul Werbos  

It would bore people. 

Metta Spencer  

Oh no, no, no. You don’t bore people with that. Whatever it was, I’m with you. I’m all ears. But nevertheless, that may not be appropriate for this. Because what I want to hear is how you got the pieces to come together. Because I can’t make mine come together. 

Paul Werbos  

So, Einstein and modern quantum people had a big argument. There was a famous debate between Niels Bohr and Albert Einstein. People still talk about the Einstein-Bohr debate. And they weren’t debating about God. They were debating about how the universe works. But if you asked Einstein about God, I think he’s what people call a deist. Thomas Jefferson is also said to be a deist. And what the deist would say was: Yes, there is a God. God created a universe and then he left it entirely. And if you know the whole history of the universe, you won’t find a trace of God anywhere in this universe for your purposes, it’s as if he’s not there. But the point is, he put it together, somebody must have put it together, and we’re living in it. But there’s no sign of God in any human experience after that. All he did was make the universe.

Metta Spencer  

Do you think that Einstein believed that all along?

Paul Werbos  

Well, all of us humans wonder. And Einstein was capable of wondering. So, I suppose he must have had times of doubting what he believed. But I think he pretty much did not doubt it much. 

Metta Spencer  

I don’t know. There are things I read about what he was saying. He talked about God quite a bit, you know? 

Paul Werbos  

He used the word. 

Metta Spencer  

Well, but they sounded pretty good to me. 

Paul Werbos  

Yeah. That building at Princeton. I got to visit it again in 2015. God, it was beautiful homecoming kind of thing to that building in Princeton, with a whole bunch of people.

Metta Spencer  

The Institute for Advanced Study. Is that the building? 

Paul Werbos  

 No. So, I’m sure Einstein went to at least 3 different buildings in Princeton. I may be confusing or combining 1 building with another. But this is the building where, for example, von Neumann had his office. Whether Einstein used his office there, I don’t know. And there was this other guy. God, help me, the name should be immediately at the top of my head. W. This is weird. Usually, the name is at the top of my head. But at any rate, one of their friends who was a nuclear physicist and work with them also had an office in that building. And when a group of us went to visit that building a while ago, in 2015, over the fireplace is inscribed Einstein’s words. Over the fireplace, inscribed in German, it says something like: Subtle is the Lord, but he is not malicious. Okay? It’s hard to understand, but it’s not impossible. And people interpret that to mean: Yes, the laws of physics are hard to understand, but it’s not impossible. And so he used the name the Lord. But if you look at what it really means, this is like Popper. If you try to understand a possible theory of physics, how do you evaluate it? And this is an aesthetic judgment. How do you evaluate it? It’s not like you expect miracles. You use the word God to describe what you believe about the universe. And Kaku even says in his new book: He believes pretty much the kind of thing Einstein believed about God. He’s got other beliefs, but God is pretty much what Einstein believed. 

Metta Spencer  

Wait, hold on. You said something. Popper said that in his new book, is that what you said? 

Paul Werbos  

No. Kaku. Michio Kaku. Kaku has this new book. Yeah. And in Michio Kaku’s book, I think he expresses the same views of God that Einstein had. 

Metta Spencer  

Einsten was supposedly closer to Spinoza than anybody else. I don’t know whether that helps resolve this. 

Paul Werbos  

I don’t know Spinoza as well as I know these other guys.

Metta Spencer  

Okay. What happened last year?

Paul Werbos  

Oh, so I even have this YouTube discussion that I posted recently, it goes into detail. There’s the Einstein point of view and there is the best modern, mainstream quantum point of view. And before last year, I basically said, I don’t know what to believe. I know that life is weird, that’s for sure. But the weirdness doesn’t fit.

Metta Spencer  

It does not fit the quantum way of doing it, huh? 

Paul Werbos  

So, Einstein’s concept of the universe seems to say we live in 1 universe with 3 space dimensions and 1 time dimension. And he might play with 1 or 2, but you know, really 3 in 1. Everything is 3 to 1. And then in modern quantum field theory, the best quantum field theory out there today. If anybody’s interested, if you go to YouTube and search on “Deutsch multiverse” there’s a great living physicist and engineer called David Deutsch. Alive at Oxford University, a fantastic person. He’s the guy who invented quantum computing.

Metta Spencer  

Yeah, I know, because I watched that show all night.

Paul Werbos  

Great. So, you saw it. So, David Deutsch invented quantum computing and his theory of physics is that we live in something called a multiverse. And what that means is the Schrodinger cat is not the only creature that splits in half. So, here’s what he says about Schrodinger’s cat: Schrodinger made up this experiment, this imaginary thought experiment, to try to prove that Werner Heisenberg was stupid. He disagreed with Heisenberg. He said: Heisenberg, If your theory is true, we could do the following experiment. We put a cat in a closet or a small room. We take a gun, hook it up to a Geiger counter, point it at the cat with a timer. And it’s set up so there’s exactly a 50% probability – according to quantum mechanics – that the gun will fire and kill the cat. And the timer stops after an hour. According to Heisenberg, at the end of that hour, the cat is not alive or dead. According to Heisenberg’s theory, that cat is in what they call a mixed state. There are 2 copies of the cat. By building this experiment, you turn the cat into 2 cats. 1 is alive and 1 is dead. And it won’t make up its mind until you as a human being open the door and look at it. And by looking at you force it to be alive or dead. Schrodinger said: that’s what you predict Mr. Heisenberg and it’s crazy. I don’t believe it. So, what David Deutsch says is: He says, if you do that experiment, you can split the cat in half. And when you open the door, you split yourself in half. And there are 2 copies of you. And 1 of you sees a live cat and 1 of you sees a dead cat. The whole universe, the big cosmos, the multiverse has 2 copies of you and 1 copy of you seems to be in 1 place and another copy of you seems to be in a place with a dead cat. There are both places that exist in the larger universe, which he calls the multiverse. So you live in your universe and the other you lives in the other universe. I tend to think there are 2 copies of Donald Trump out there right now. And 1 of them actually was inaugurated as President and the other 1 is dead. But I think the President is kind of fading away because my theory of quantum mechanics is a little different. I think a copy can fade away with time. And I think the other Donald Trump is fading away. That is my opinion. But yeah, this is real stuff, this happens. 

Metta Spencer  

You know, I’m doing my best. 

Paul Werbos  

It’s a little weird, yeah?

Metta Spencer  

It’s not just sort of weird. I mean, where would you go beyond that? I mean, what could you possibly do if you accept – I presume you do accept that this is the way it really is – I don’t know how —

Paul Werbos  

There are 2 problems it raises. 1 is: What do you do? My wife would always ask that question. What do you do about this If it’s true? David Deutsch’s answer is you can use it to build better computers and you can. But the other question is: If you believe that Albert Einstein is true and there’s only 1 universe, how can there be 2 copies of you in 1 universe? That’s what was really bothering me. And it wasn’t until last year that I figured out that I really think I understand the answer to that, but it’s very tricky.

Metta Spencer  

Okay, can you do anything to help me with it? Because, you sure brought me to a funny place. 

Paul Werbos  

Well, should I talk about practical uses? Or should I talk about my logical contradictions?  Okay, well, let me start with the practical thing. Okay.  I put up like 3 YouTube videos this past month. And I’m not too good at organizing them. You can probably use it better than I can, because you know about these media in practice. I put them up and one of them was a discussion of a new quantum technology. The present quantum computing, that the US government is funding today is all based on this idea from David Deutsch. It is what he calls the quantum Turing machine. And he came up with that idea to prove that his idea of the universe can be put to work. So, when he talked about the multiverse, when he said: There are many copies of you, people would say, eh that’s some weird philosophers’ theory. So, what Deutsch did was he proved theorem saying: It’s not just a theory, if this theory of physics is true, it tells you how to build a new kind of computer, which in a way is a million times more powerful than those computers based on classical physics that you’re building now. You can build more powerful computers, if you take advantage of this multiverse. He said: what you do is that the cat is not the only thing that can be put in a mixed state. In fact, the cat is not the easiest one to put in a mixed state. You can take a computer chip and put the computer chip in a mixed state. And in fact, what if you have one computer chip, but actually, what if you have a million copies of that same computer chip in parallel universes in different states each doing different work? If you had one million computers working on one million different problems in parallel and then you combine the results, you get a computer which will do the work of one million computers in one. One million the work for the same physical object, if you know how to put the object in a quantum state. And when he started publishing papers proving that mathematically, a lot of people said that can’t be true. We can’t really do it. We can’t really build it. But then a computer scientists came along who said: Well, no, we could use this kind of computer. And then a guy at MIT came along and said; You know, I can build something like this. I can build a quantum superposition in a coffee cup. And there’s a guy in MIT named [Neil] Gershenfeld, who wrote a great paper in science that showed I can use quantum superposition the way David Deutsch said we can. We can use this. We can build more powerful computers in a coffee cup with molecules in a mixed state. And after that, there were people like NSA who said: Oh, my God, if people can build computers a million times as powerful – a trillion times as powerful as anything we have today – they could use these computers to crack our codes and break our communications. We better find out about this. And so for several years, they started investing something like a billion dollars a year trying to understand what can you do with a David Deutsch computer. And I was actually part of the interagency group that oversaw that kind of research. I got to see what they were doing at NSA and DOD and all these places. We funded some from the National Science Foundation. How do you build this kind of a computer? But my real message is: That’s all just Stage 1. That’s all based on that first idea that Deutsch had. But this multiverse idea is more powerful than his concept of a Turing machine. And that’s what my new talk was this past week – the same concept of physics can be used to solve other problems in computing that Deutsch doesn’t know about. And I know about some problems in computing that he doesn’t know because I have the background in AI and neural networks. I know ways that you can use Deutsch’s physics to solve problems in computing, which make those old quantum computers look like a horse and buggy. We have technologies that we could be implementing tomorrow, which are like a million times more powerful than what they’re getting out of these quantum computers.

Metta Spencer  

Both.  Whoa. 

Paul Werbos  

Yeah. 

Metta Spencer  

I don’t know whether to be happy or sad.

Paul Werbos  

That’s a good reaction. That is a sane human s speaking. So many people either say it’s great or it’s terrible. Will we live? Or will we die? And you are saying the truth, which is I don’t know. And the next step is: What will decide whether it’s good or bad? And what can we do to save the human race from the bad scary things, but also get the benefits of the good things? That’s the question we should be asking. 

Metta Spencer  

Absolutely, yeah. Do you have a good answer for that one? You have an answer for a lot of stuff, but that one… 

Paul Werbos  

Oh, yeah, yeah. I’m a terrible communicator, but I’ve run across a lot of stuff I’ve learned. 

Metta Spencer  

Alright, proceed. How are we going to… now come on… in a simple way? And what we want is to go back to Popper. You have got to find a way to falsify everything. Everything. I don’t know how to do everything. Can you do that?

Paul Werbos  

Well, there are two issues right here, right now related to what you just said. One issue is: Is this true? So, I gave this video on how to build a third-generation quantum computer, two generations beyond David Deutsch. So, question one is: Is it true, can you do it? And question two is: If you can, what do you do about it from a policy point of view? How do you make it good? And how do you figure out how dangerous it could be? So those are two different questions: the policy question and the proof question. 

Metta Spencer  

If you hear some funny noises, because there’s a man hanging on a rope outside my window, throwing pebbles at the window, and I’m not joking. Hi, guy.

Paul Werbos  

I said life is weird. My whole life has been one impossible, weird thing after another. Starting in 1967. Although some people would say my mother was already a little different from average too. 

Metta Spencer  

Okay. Alright, which of these questions do you want to tackle? We have 20 minutes.

Paul Werbos  

Maybe the policy one, because I do have this video on YouTube on this new quantum technology. And that’s a whole hour on what can you do and that includes how to prove it. You know, what are the stages you go through to build it and prove it? So, in a way, I have already said that, but the policy question is closer to your bailiwick anyway: how do we save the human race? And to me, the policy question about this new quantum technology is one part of a complex set of what I call existential risks, coming from all kinds of internet technology. Two years ago, the Japanese and the Koreans asked me to give talks to their top policy people on the future of the Internet and what are the threats? And what can we do to prevent the biggest risks to the human race? I have a good friend named Jerry Glenn, who runs the Millennium Project, this big international futures group. We talked about him a lot. He has said: All over the world, we need to create an Office of Existential Threats. Threats to the survival of the human species for the UN Security Council. They have a human security office there already, but we need a new office in that office to address the major threats that confront the whole human race. And I have argued —

Metta Spencer  

I’ve just set up the first office right here, you’re on it. And I’ll be glad to give it to the right person, if you can figure out who wants to take it over. I am not sure the US government is the outfit that should run the Office of Existential Risks, but te idea is a great one. 

Paul Werbos  

So, the proposal that my friend Jerry is making around the world is that this office should not be run by any one country. It should be as open and transparent as you can make it, because half the existential risks are due to people hiding stuff, they want to use to kill other people. Right? So, it’s very important, it should be open and transparent, and available to the whole world with feedback mechanisms so they can learn from the whole world, which is just as important. And the place where it should be – Jerry says the place where this new office should be – is under the Office of Human Security, which now exists in the United Nations Headquarters located in New York – but you know the UN, you can have satellite offices all over – but the central place, it should be reporting is directly to the Security Council in New York. And he’s trying to get support from the General Assembly on a feasibility study to start making that happen. And my claim is, if you want to do it, right, you could screw it up easily and the US government could screw up, of course, there are lots of ways to screw it up. If you want to do it right, this new office should start with the two most urgent new divisions. And one would be a division on climate risk – the climate risks that could kill the human species. And I’m not just talking about wildfires or warming, I’m talking about much worse stuff. And the world needs to know about the worst stuff. They think they know about climate risks. They don’t. There are risks in climate, which are much worse than what any of these policymakers begins to understand. So that’s one of the divisions. We need to both get that information out and we need to get the solutions out. But we need another office, just as important, which deals with the whole complex of the future of the Internet. And one part of that is the new quantum technology. 

Metta Spencer  

Okay. 

Paul Werbos  

And the new quantum technology should be developed on this open international basis. And that’s where I would put this new research that I’ve been proposing for more advanced quantum computing.

Metta Spencer  

Oh. So, the research for your quantum computing is going to be done under the control of the Office of Existential Risk?

Paul Werbos  

That should be the central focal point, not the only point, but that should be the focal point for the new research.

Metta Spencer  

Well, you know, that is as close to a good idea as I can think of, but I immediately think you’re going to put it under the Security Council? I don’t think so.

Paul Werbos  

We’ve had so much discussions of that Security Council this year. The United Nations…. well, there is no government on Earth that has been perfect. The UN has not been perfect. The US government hasn’t been perfect. Even Canada hasn’t been perfect. But you’ve got a lot of competition when it comes to people doing crazy things. All around the world.

Metta Spencer  

Okay, I’ve got qualifications on Security Council… I mostly think in terms of… it would be a help and a step in the right direction if we had a Parliamentary Assembly, because that could have more clout in a way than the General Assembly and the Security Council. It would not be accountable to and not be under the control of nation states, but be accountable – theoretically, in principle – to the human population. The trouble is, I don’t have any use for the human population anymore.

Paul Werbos  

Well, you know…

Metta Spencer  

I’m sorry. 

Paul Werbos  

You may think it sounds like a joke, but to me it brings back memories of scary, real people I have met. When you talk about real threats coming from the internet, they are much bigger than most people know. And I have met some of these threats. And believe me, I have met some very scary people with capabilities you wouldn’t believe.

Metta Spencer  

I think we both must have met the same guys. 

Paul Werbos  

Well, we’ve met Popper. 

Metta Spencer  

But he’s a good guy. He was really a good guy. 

Paul Werbos  

No, I know, but we seem to have some other things in common. But I wonder, could you have ever met somebody as scary as some of the people I have met?

Metta Spencer  

I tell you what…. No, that would have been a whole different conversation if I answered you. So I’m not going to answer the question. 

Paul Werbos  

The whole audience would be interested. Or maybe it would be too scary to tell them? 

Metta Spencer  

Oh, no. I mean, it’s not that these people are not so powerful. The scariest people I know are ordinary people who want to get along with each other, and therefore they will allow each other to tell lies and they won’t challenge anything that the other person says, because they want to get along. That’s what I consider scary. 

Paul Werbos  

No, I agree with you. Lies are a big part of what is threatening our survival. 

Metta Spencer  

They are the majority of the people now and they win elections. And so, I don’t have any answer for that. I mean, I used to believe in democracy and I’ve got to the point, I can’t believe in democracy anymore, because I know my friends. I don’t trust my best friends anymore. I can’t. We’re all crazy.

Paul Werbos  

I’ve met folks who don’t trust any human being. There’s a lot of scary people. And they can do it. 

Metta Spencer  

Okay, I mean, this is a whole different conversation, because it’s really where I’m at. I’m at the point of thinking, I have no solution to the problem that group dynamics is so powerful and people’s identities and, and affiliations and personal loyalties and things like that take precedence over any kind of rationality. And therefore, you can’t have a political system, you can’t have a democracy, where people base their judgments on complete falsehoods and know that they’re false. You can’t run the world that way. 

Paul Werbos  

Here’s part of the point. I even have a little web post that needs to be improved. You have something the UN could take over. I do too. If they create the Office, it should take over a lot of stuff. One of the things I have is called Sustainable Intelligent Internet. And this is a talk I gave in Korea and Japan. There is a way to try to design the future architecture of the internet to try to create what I would call networks of truth. A lot of the lies you see today – there are super lies today- the problem of lies is worse than it was 20 years ago. We didn’t invent the lie in computer science. But there are lies out there right now, which in many ways are more dangerous than the kind of lies we had back in the 1930s. And a lot of the reason why these lies have proliferated is because they’re proliferating on the internet. What if you had a tool for ordinary people, where ordinary people can connect in a way that doesn’t screw their heads? They have a choice. They can find out the truth. What if we build computer systems that are truly intelligent? That use high levels of intelligence, even quantum intelligence, to try to be able to balance it out to empower the humans by giving the humans knowledge and access which they don’t have in the world we are now building. So, we are building a dangerous world now, where humans have less and less power, where humans are oppressed more and more by whoever their boss is. They have less real information because of the lies. But what if we build an internet system to try to empower and inform the humans? A kind hybrid way of combining. So that’s my idea.

Metta Spencer  

Well, the problem is, what if they don’t want to hear you? You can tell them the truth and what if they don’t listen. What if they won’t? That’s what’s scary to me.

Paul Werbos  

Well, but see, if people don’t have a choice, what can they do? Now I know something about how human brains work. And I really believe that if humans had the right choices, and if they were really empowered – and they had the right choices – I think there’s a chance that humans could survive on this world. You’ll notice I’m not guaranteeing anything. But if we do the right thing, there’s a chance we could survive. 

Metta Spencer  

Say it again. 

Paul Werbos  

Okay, if we do the right things with the internet, if we design a system that is better at empowering people, than any of the things they’re working on today, I believe it is possible to empower humans and build a system which is more responsive to humans in such a way that together we’ll muddle through and we will survive. I think there is hope with that. And forgive me, if I say I think that God is a player in the game in the way I use the word God. I don’t use the word God the way the Pope does. I don’t use it the way I Einstein did.  have my own ideas of how it works?

Metta Spencer  

Okay, we have seven minutes now. You can do that. Explain God in 7 minutes. 

Paul Werbos  

Explain God in 7 minutes? Okay, you know, you’ve talked to the right guy. This has been a terrible year for a lot of people. I’ve had ups and downs like you wouldn’t believe. And I’ve had these contradictions. So, there are two books that I read a lot in the last year. One is a book by Carl Jung called The Red Book. And I discovered I could get a copy for $20 in a good Kindle edition from Amazon. The start of the book was very depressing. It was all these academic theorists with big words. And then begins the real book. Carl Jung talks about how the human mind and human life worked. And he lived in the 1920s in Europe. You think we’ve got problems? Well, they had a few problems in the 1920s, you know. Jung talked about something he calls the spirit of the times and the spirit of the deep. And I believe that the spirit of the deep and the spirit of the times actually connect two minds that I know about from my mathematics. I think there is a spirit of the times and a spirit of deep. And I think that the best use of the word God for me – the way I would like to use the word God – is to refer to what Jung calls the spirit of the deep, which is a real intelligence. And how does this intelligence work? Well, I can’t do that in one minute. But there is intelligence and purpose. And this is the funny thing, it’s in Einstein’s mathematics. We just have to understand the equations. I did a post on that on Blogspot this past week and I even copied it to Facebook.

Metta Spencer  

Okay, look, you’ve got to come back and we will do a whole hour on that blog. Okay? 

Paul Werbos  

Okay. Great. Thank you so much. This is fun. 

Metta Spencer  

Don’t go away. I don’t want to let go of you.  We’ve got another 5 minutes. 

Paul Werbos  

This gets to be at the far edges of what I understand. This is what I didn’t understand a year ago so it shouldn’t be easy to understand, unfortunately. But we know from physics, there are these Schrodinger cats. We know we can put systems in a state of quantum superposition even bigger than cats. The Chinese have done it with systems a thousand miles wide. You can build mixed states that big. You can build quantum You can build Schrodinger cats. But Einstein said we live in one universe. So, if there’s only one universe, how can there be a billion copies of me? And the answer takes some getting used to. The answer comes from something Plato said. Where Plato said: We are the self that we think we are most of the time. The self that we think we are most of the time, is like the shadows cast on the wall of the cave. And each true person can cast many shadows. And there’s only one true person, but there are many shadows. And these Schrodinger cats are really just shadows of what the universe eventually converges to. And believe it or not, I’ve worked out the math of it. It’s a paper I figured out 4 years ago and it took me 4 years to understand my own paper, because the math is very, very tricky

Metta Spencer  

I’m intrigued most by the notion of what the universe converges to. I don’t know what to ask about that. But converge to me sounds like something that’s going to happen in the future, like Teilhard de Chardin or something converging toward some final state. I don’t think that’s what you mean.

Paul Werbos  

Teilhard de Chardin, I want to treat with respect. Because there is so much truth in Teilhard de Chardin. It connects with his worldview, but that’s not the kind of convergence I’m talking about. To me Teilhard de Chardin, frankly, has more to do with the spirit of the times. The spirit of the times is in a way, the soul of our noosphere and of our solar system. And our solar system is so big, were just part of it. But it’s a small part of the whole cosmos. So, the spirit of the deep is as big as the cosmos. And the spirit of the times is just that little noosphere that we are part of. 

Metta Spencer  

Can you get guidance from it?

Paul Werbos  

As of this year, I think I have a better idea how to get guidance from the spirit of the deep. As of this year, I figured out how to make more conscious connection to the spirit of the deep. I have been able to make contact with the noosphere, with nature, and with life. I have known how to make contact with life and nature for many years, but the spirit of the deep, the big part of what changed for me this past year is to understand it and figure out better how to connect with it.

Metta Spencer  

Well, I would be happy to take a lesson or two.

Paul Werbos  

 I still have a lot of lessons I need to take. 

Metta Spencer  

This has been such fun. Listen, I don’t want to let you go, but times up. 

Paul Werbos  

Okay.

Metta Spencer  

Will you come back?

Paul Werbos  

 Oh, absolutely, absolutely. 

Metta Spencer  

Okay, to be continued then. It’ll be interesting to see what people think of this show. 

Paul Werbos  

I’m sure there will be lots of tomatoes and also some good little gifts.

Metta Spencer  

Yeah, well you know, we did not even get to what we were going to talk about, which is climate change. We’ll get to that one of these days, okay?   Okay. Bye. 

Select the Videos from Right

We produce several one-hour-long Zoom conversations each week about various aspects of six issues we address. You can watch them live and send a question to the speakers or watch the edited version later here or on our Youtube channel.