StarTalk Radio - Emergence Explained with David Krakauer

Episode Date: August 8, 2025

What is life? What is intelligence? What is… complexity? Neil deGrasse Tyson, Chuck Nice, and Gary O’Reilly learn how complexity science, chaos theory, and emergence help us understand our place in the universe with David Krakauer, president of the Santa Fe Institute.NOTE: StarTalk+ Patrons can listen to this entire episode commercial-free here:https://startalkmedia.com/show/emergence-explained-with-david-krakauer/Thanks to our Patrons teonie, Dixie Gamoning, Greg Meyer, Mike Bilodeau, Mitchell Keesler, john hutt, Karen Buss, The Merry Widow, Casandra Martin, Swaraj Jaiswal, Hoang Nguyen, Knooble Gooble, Panainte Victor, Peter Jensen, Rajesh Bhaidasna, Victor Pomales, George Mulder, Life Space and the Lot, RandomBrian423, blitzgrub, Travis Bridges, Sreya Kumpatla, Erik Scheirer, Natalie Tabor, SwaZam!, KILOCREAMYY, Lisa Peldiak, Tosin Awofeso, Joe Buzz, daevon pearson, Amie Christy, Simone Adair, Philippe, Logan Davis, Ted Parsons, Macs Ton, Ben, Quentin Ferguson, Ash De Zylva, Evalena Marie, Nancy Bijok, Jacob Garcia, The Preschool Doctor, Amber Shaw, Erin, ilya, Kevin Nguyen, Austin Weets, and Alan G for supporting us this week. Subscribe to SiriusXM Podcasts+ to listen to new episodes of StarTalk Radio ad-free and a whole week early.Start a free trial now on Apple Podcasts or by visiting siriusxm.com/podcastsplus.

Complexity and emergence, two terms that I think in modern times should be on everyone's tongue. They will be off to this show instead of what we have right now, which is stupidity and ignorance. Yeah. Coming up on StarTalk. Welcome to StarTalk. Your place in the universe where science and pop culture collide. StarTalk begins right now. This is StarTalk, special edition, which means I got Gary O'Reilly sitting right next to me.

Hi, Neil. Gary, former soccer pro. Yes. Chuck, nice. Good to have you, man. Who knows nothing about soccer? Nothing. That's because you're American.

That's right. American, baby. So, you guys always come up with fun topics. And today is no exception. Yeah. Complexity. Oh, my gosh. Complexity.

Yes. And I'm amazed me if... This is our first time we've ever handled this subject. It's been one we've been waiting to find the right guest to enter the arena. Okay. And guess what? We have.

We have. All right. Take us there. Let me frame it this way. As we come to grips with the reality of the lives we live, I ask myself, is there anyone out there considering the complexity of everything, right? Asking the big questions, how did intelligence evolve in the universe? Does intelligence have limits?

How do ideas evolve? Are there laws of life? And then here, the biggie, what is life? Oh, yeah. So, happy to say there isn't just one individual but an institution that's set to these tasks. Neil, if you would kindly introduce our guest. I would be delighted to. We've got with us, David Krakauer.

David, welcome to StarTalk. Fantastic to be with you. Yeah, and you are president of the Santa Fe Institute in New Mexico. This is a world-famous place where deep thinkers go. And the question is... And die. Do they ever come out is the question. So you are the William H. Miller professional.

of complex systems. I'm impressed that that's even a title you can have. That's a great, yes. Professor of complex systems. Imagine how big he's door is just to get that name on it. I want to be professor of simple systems. Well, there are those. You have a background in evolutionary theory, good,

with also a background in computer science and math. So it sounds like you've got just the right pedigree for this. And I think if we have time, we'll get to to that you're founder of the interplanetary project at the Santa Fe Institute. I only just now first heard of that, but we wouldn't get to the bottom of that as well. So let me just lead off here. We've had a couple of guests on our show who have either spent time at the Santa Fe Institute or were on the faculty there, if that's the right way to say it. Can you just remind everybody what the mission statement is of the Santa Fe Institute? And what distinguishes it, and what distinguishes

it from any other place that believes they're having deep thoughts about the world. And can you simplify that into one sentence? No, he's a complexity professor. Oh, that's right, that's right. Well, I'll tell you the mission, which is one sentence. It won't help, but it's searching for order in the complexity of evolving worlds. That's the mission statement, and I guess this whole conversation is, what does that even mean? Which has its own bias, because you're presuming there is order within the complexity to begin with.

city to begin with. Absolutely. And I assume the fact that we're here having this conversation is some kind of evidence of that fact. Okay. Okay. All right. Yeah, so essentially the Institute was founded in 1984 in the mountains of New Mexico. There's a reason we're here because of Los Alamos and we can discuss that history. It's sort of interesting, actually. I think of us as the sort of more generative, optimistic, fissile material in this part of the landscape. Just to remind people, Los Alamos, I mean, it's a national lab. Okay, and where nuclear reserves are kept and managed and overseen. Very important place.

And it's not an accident that it's in the middle of freaking nowhere. Right. Okay. You're not going to put that in a city. Just in case of an accident, it's in the middle of freaking nowhere. Okay, so you have sort of genetic history, overlapping history with Los Alamos. Los Alamos. Absolutely.

In fact, it's sort of interesting that there's an interesting... cultural history here, the founding president, George Cowan, he was a child prodigy, and he had worked very early with two people from your world. He'd worked with Eugene Vigner, who won the Nobel Prize for his work on symmetries applied to physics, and with Enrico Fermi. Famous for the Fermi paradox and among other things, yes. And among many other things. And so in the, he then became eventually the director of research at Los Alamos. In the 19th, 50s, he was asked to give a talk at the Aspen Institute, which is kind of just down the road

in Colorado, and he gave a talk on social entropy. So the notion of entropy, we're familiar with, you know, from physics, that systems tend towards disorder. And he thought, well, maybe that's true of society. He gives the talk and he flames out. No one understood what he was talking about. Is this a metaphor? Do you actually have something in mind mathematically?

That was in the early 50s. 30 years elapsed, you're thinking there's something wrong with the world, where the social sciences and the natural sciences, the mathematical sciences, are not communicating as they should be. And that's really the origin story in the 80s, a group of rather illustrious people came together, several Nobel laureates, said, what would it take to build an institute where we don't start with the divisions of departments and the divisions of knowledge, but we actually just kind of short-circuit them and have people compressed in a very high density in one place in the high desert. And that's sort of the premise. I'm interested. Interplanetary project.

I mean, there's a niche that needs scratching right now. Please, yes. Okay. So given our perspective on how the world works, how the universe works, our belief, right, that life is probably universal. There's no evidence, but I think it's a reasonable assumption to make. Right. What would it mean to have a theory of the universe or a festival of the universe that included not just physics, but economics and sociology and poetry and art and music, right? In other words, astrobiology, as Neil well knows, is dominated by physics. But if it's true that there's life elsewhere, there might be extraordinary sport, extraordinary music to be discovered on another planet.

So we wanted to expand the range of thinking about life in the universe to encompass all disciplines. And that's, again, that's the kind of spirit of SFI. But are these really disciplines that are created by life? Or are they our perception of our preeminence in life? Yeah, I mean, that's a really deep question that relates to this concept. I didn't understand it. Really? No, so here's what I'm saying.

If you find aliens and they play music, is that what you're saying or what? Yeah, my point is this. We think that that would happen because we do it. Oh, okay. Do you understand? So that's an interesting bias. We have a, we have a perception of preeminence in the universe itself. And so even in the movie, why will we make the supposition that those things exist?

Right. They played musical notes back to us. Right. Yeah. Look at our other bias. Let me just turn this around on you a bit, which is that no, everyone would say, look, maybe chemistry is universal, right? So we're going to find DNA or RNA.

That's universal. But mathematics, well, we could describe things mathematically. But is mathematics can be universal? Is music? I mean, what is the, if you like, set of beliefs we have that make the most fundamental constituents universal and everything else somehow anthropomorphic? And I would like to question that premise.

Okay. I'm down with that. And that flows in and through your thinking about an interplanetary project. Is that correct? Yeah. I mean, in fact, the slogan of the interplanetary project was changing the world, one planet at a time. That's actually very good. And I think, right, you guys want to change this world.

I don't want to particularly change others. And the question is, could a more expansive, humanistic, decent perspective on things allow us to make progress on this? planet. Well, we did discover a planet that's completely inhabited by robots. It's called Mars. Oh. How long have you been set on that, Joe? Waiting to put it in, out there. So now we got the institute understood. Now tell us about complex complexity. Yeah, right. So, I mean, this is, there are several ways to do this and you just tell me which you like and don't like. And I'll just throw out options.

because I'm going to take different slices through this idea, right? Let's start with roots. What are the roots of complexity science? Okay, aha. Well, okay. So, all right. So I just wrote a whole series of books on this.

So it's a very good question for me. So essentially, one way to say this is the roots are the study of machines. Machines that were made in the Industrial Revolution, like steam engines, or machines that evolved, like organisms. Just to clarify, you're not referring to the five basic machines of physics here. No. You're referring to industrial machines that we have made manufactured in our civilization. Exactly, like the steam engine, like the centrifugal governor, or like an organism.

So, exactly. So mechanisms in the natural world that do work of one kind or another. Gotcha. And, right, so sometimes we call that problem-solving matter. matter in contrast to the regular matter that's studied by physicists and chemists. I like that distinction. Yeah, because we, in physics class, there's a frictionless pulley on a, pulling a mass on an inclined plane.

Right. We're not thinking that has consciousness or anything else. No. It's not doing anything interesting other than fitting into my problem set. Right. Right. So yeah, go, I like that distinction. Continue, please.

Right. So like, so that's one way to go. Right. So problem solving matter versus regular. regular. And then there's a whole bunch of questions. In fact, you just did it that feel very natural when you talk about problem solving matter. Like, for example, how efficient is it? How do it originate? How does it adapt? How smart is it? How does it store information? How does it evolve? How does it fail? Right? And how eventually does it go extinct? And some of those questions are shared

with physics. Many of them are not. It's not really meaningful to say how smart is the moon. I mean, some people probably do, but we tend to ignore those people, right? And so, but it's totally natural to say, how much information does an economy store, right? How does a society of insects compute? So in a way, you can define it by the questions that feel natural for it, and all of those are natural to all the scales we study. I like that because what that, when you have that awareness, it means you're not going to force a square peg into a round hole if you're looking at a different system, just because certain

questions and solutions work with one system of machines doesn't mean they're worth all of them and so if you if you keep your interactions native to the phenomenon you're surely likely to get deeper into what's going on for it is that a fair characterization of what you just said no no it's exactly right and i think what happens and this happened to history of physics and is that the domain the what we would call like the ontology the reality um requests from us a certain methodology or set of approaches, we call epistemology. And so physical, the physical world loves principles of symmetry, right, which we encode in various principles in physics, typically conservation laws, right? In complex systems, we like broken symmetries and noise,

irregularities that get fixed, sometimes we call frozen accidents. And that means a different kind of theory. And a lot of SFI is about finding the theory for the messier domain that lives between order and chaos. Hi, I'm Ernie Carducci from Columbus, Ohio. I'm here with my son Ernie because we listen to StarTalk every night and support StarTalk on Patreon. This is StarTalk. We With Neil deGrasse Tyson. So this is chaos theory. So you're looking for patent out of chaos theory.

But if you take a step back, are... But he didn't say it was chaos. Yeah, he didn't. He said it's in between. In between. I did. You're not the guest. I know.

No, what he said was it's in between pure chaos and order. And order. And there's messy systems. that might yield to analysis, as you're saying. Does chaos actually present with certain regularity of patent? Yes, so actually, I mean, just as an aside, I'll get there in a second, a footnote to that.

The answer is yes. Neil's right. I mean, chaos theory is a tiny, tiny, tiny part of complexity. And in fact, weirdly enough, it's a part of complexity that fits very naturally in physics. It came out of the study of things like the so-called three body problem, classical systems, right, that are completely deterministic. There's no noise in chaos. It's deterministic irregularity. So it does present as order that appears superficially

to be random. And we're interested in subjects that have that property, but they add real randomness, like thermal randomness, noise to them. So early in our history, because of a book that was written by Jane Glyke, actually, in 92. Oh, yeah. It's called chaos, right? And it was an important book for SFI, because he talked a lot about our work, but it's actually a tiny part of what we do, and it's the part that's very close to physics. Mm-hmm. Mm-hmm. So, so it could...

So there. That not. Now I know more. I am better educated, so thank you. Okay, so how about other elements? I'm just going to say, but are you looking in the messiness for reason, or if you want to call it order, if you want to call it non-messiness? So that the messiness itself really isn't messy, we just see it that way.

We don't understand it. Yes. Okay, so, right. So interestingly, so where does this word complexity come from? And it gets exactly to your question. In 1948, Warren Weaver, who was a mathematician, he wanted to classify all of regularity or irregularity in the universe. And he said, and we can debate whether this is useful, I find it. quite useful. He said there are simple phenomena. That's classical physics. It doesn't mean it's easy

to understand, but it's to Neil's earlier point. It's simple. There are beautiful laws, elegant mathematical formalisms that explain it. Then there's a world of what he called disorganized complexity. That's the sort of the study of gases, what we would now study with thermodynamics and statistical mechanics, irregular things, but have beautiful descriptions. And then in the middle, there's organized complexity. It's not a gas. It's an organism. It's an ant. It's a brain. It's a city. And in that space, what we realized really at the end of the 19th century is we don't have good theories for it. We can do irregularity beautifully. We can do simplicity beautifully. And then you move to everything that we actually care about in some sense as human beings and as animals, living beings.

we kind of under theorized, beautifully described, beautiful artworks, but what is the kind of mathematics of that zone of organized complexity? And I think you make a really important point, which is that how much order you see is a function of the observer. And if the observer has more computational power, it's going to see more order, right? So complexity is observer dependent in a very profound way, rather like quantum mechanics is observer dependent, in a very different way. But this has to do with our computational capability, if you like. I'm going to tell

you the truth right now, and I'm just going to come out and say it. I actually thought when we started that this was going to be bullshit. But he is making some great points here. That's why he's on. That's why we got the man. That's why he's here.

That's why we got the man. So when you look for adaptive functions, is that the sort of pattern that you're looking for, how it's reacted to, one of a better term, environmental circumstances? Definitely. I mean, I think, you know, one of the challenges, right, is going from, and I think you've had some of my friends on that show who talk about this. I mean, maybe Sean and maybe Sarah. Yeah, we had Sean Carroll. Yes. Right. Right. And so that's a good example. So you go from, you know, what's the difference between a ball rolling down a hill, right? So it's minimizing some

function and an organism adapting and that distinction to this day people debate is there adaptation in physics you know maximizing some function or minimizing some function or is it unique to the living state and so for us agency the agent you know it could be an organism could be a machine actually could be an AI has some characteristics that a ball rolling down a hill do not have that we need a new theory for. And we can talk about what that new theory looks like. Right. But adaptation is absolutely central to complexity because without it, there isn't any. I will tell you this much. An AI rolling down the hill is a hell of a lot more expensive. Than a ball. You have lost a lot of money other than that ball.

All right. So this approach seems potent enough so that it does not need to be constrained to any one discipline. and I'm impressed to see efforts to apply this to society, economic systems, civilization itself. So do you have enough confidence in your modeling and its foundations to then go into something that's way more complex than basic physics because you're now involving human behavior? Yes. In other words, can you solve stupidity? Because we are living in a very high time of stupidity. well that's actually we'll get that my area of interest so I can't solve it but I can diagnose it and so but we can talk about what it means let me just address both I feel this question

I don't know is a real honest answer to this question I think that something interesting happens it's sometimes surprising and then this is very familiar from other sciences right if you try to understand an individual particle floating about it's really hard you know its trajectory is essentially random. But you have enough particles and you average, and order emerges, right? So, for example, a fluid will flow. And so we have equations of fluids that look at things like viscosity and average density, that don't worry about the individual particles and all their peculiarities.

And in just the same way, humans that are fundamentally, I think, unpredictable, I mean, there are things we do that are predictable, alas, but in aggregate, there are predictable regularities in societies, and economists and psychologists and sociologists exploit those facts. So if you aggregate right, then a regularity in a pattern can emerge again. And I think the big question for us is, is a city, is a civilization one of those things that actually shows emergent regularity, which will give us some handle on them that we might be able to control?

This is not a pushback, but I need some clarification here. Because when you talk about a city and you talk about, Let's just say the aggregate of a group, okay? All right. People. People. Thank you. We talk about people, all right?

A neuroscientist will tell you that we are predictable enough that if given enough data on the person, I can tell you exactly what they're going to do. And if given enough... No advertisers. Advertisers know that. Yeah, advertisers. You don't need neuroscientists. Right. And if given enough data on a group, I can tell you exactly.

how to manipulate them for them to become violent, for them to become docile, for them to become agitated. So where is the emergence in that? We haven't talked about emergence yet. Oh, okay. Well, you're right. Okay. Okay. But it's still a great question.

I felt like he was moving towards emergence with that. No, no. But the bridge there is he's brilliantly analogizing the fact that you have gas fluid particles. And individually, you don't, there's no, there's no flow law for an individual particle. You're not worried about the particle, but when they come together, they actually present these properties that make them as a whole act differently. And you basically just said what he said, that an aggregate, we behave in ways that may be predictable and be describable analytically. Yes, but here's the difference. As human beings, that's due to the fact that each one of us has to be in a very particular place for that to happen.

Do you understand what I'm saying? Unlike particles, all the particles is going to do the same thing. All of them all the time. All right, David, question for you. If you go back to the reference point of cities, does it change for an unplanned to a planned city where the predictability is as same or very different? Well, that's interesting. Yeah, I mean, good question.

I mean, I think that for the things that we study, not so much, interestingly, okay, I want to somehow, I want to thread these points. All right, good. Because they're all really good, and they're actually getting at the heart of why this is difficult. So the first point about, you know, yes, humans are not like little, you know, brownian motion particles. We have histories. We have desires. We have beliefs, right? And so we're heterogeneous in a way that particles are not. That is absolutely right. It's what makes me difficult. That's what I'm saying.

Yes, that's just point. Sorry for taking so long to get there. No, but I think that is really important, right? And yet, right, if I half the price of the groceries, you're going to go out and buy more of them because you're fearful that the price will go up again next week. But you see what I mean. There are regular patterns of behavior despite all of that heterogeneity, which is important in our lives. And so the question is at what scale? So now that's going to cities, right?

It turns out, actually, that cities are so constraining of the supply of energy. and the supply of resources and the interactions between people and neighborhoods, that there are emergent regularities that come out of those constraints. So, for example, if you look at the growth of GDP as a city gets larger, it follows a universal law that looks like it's a law of physics. You get a scaling of GDP that goes as essentially the population size to the 1.15 power. So whatever. But the point is that, weirdly enough,

when you impose these strong constraints on society, they start looking more like physical systems. They have emergent regularities. Oh man, that's great. And as you remove those constraints, then of course many of these systems, these kind of theories fail. So let's blow open

the topic now then of emergence. Yeah. Because that's a topic of great fascination, especially in biology, but to the extent that that can apply to other systems, that would be amazing to get some insight into what's going on on the frontier where we don't understand what's going on. And just because you and Neil, and Gary probably too, because he did all the research on this,

can you please define emergence? Because I hear people use the term, and I think a lot of times they're not using it correctly. So can you please just tell us, what is emergence? That's a completely fair observation. I spend most of my life, you know, in horror. So, okay, first of all, I want to say it's a difficult concept, and I think it would be completely accurate to say that we still have huge amounts to work to do

to understand formally what it means. But I'm now going to, having said all that little caveat, I'll tell you what I think it is. So let's just start with physics, because it's easy. So we just talked about it. We talked about gases, like the kinetic theory of gases, put loads of those particles together, right? And you can get solids at the right temperatures and pressures, and you can get fluids.

And it turns out that the mathematical equations you use to describe those two systems are different, right? And the dimension, the simplicity of you like, the number of terms in those equations are different. And so we like fluid dynamics because it's actually quite a elegant way of the way of describing the behavior of loads and loads of particles in a particular temperature impression. That is emergence, the fact that you have two things, a new state of matter with properties that wouldn't really seem to apply at the individual particle level, and it has a new language, a new language of description, and a new language of prediction. Those are

the two hallmarks. Now, it's interesting, you were talking about psychology, advertising, and neuroscience. And that's a beautiful example. So let's imagine, right, that in order to be a really good psychologist or marketer, you had to be a great neuroscientist. Of course, a really good marketer doesn't need to be a good neuroscientist because all of that detail is a little bit like a particle and a gas

relative to a fluid. A good marketer is doing something like fluid dynamics by analogy. They're understanding collective properties that have their own language. And I think a lot of kind of pseudoscience, to be honest, is where a level that has its own perfectly adequate language starts using the more reductionist language

to give it legitimacy. So, emergence. One, new states or phases of matter or organization. Two, new languages and descriptions, typically mathematical, it doesn't have to be, right? And three, the tricky one is, not everything deserves to be called emergent. And so finding that emergent level

is actually part of the challenge. And a lot of, we would argue, a lot of economic theory would be better off being replaced by psychology because the language it's derived or invented doesn't really work. It's not like fluid dynamics. So the failures of emergence are also really interesting. Cool, man. So when you get to the state of emergence,

what is, what is the prediction accuracy for you being able to say, this is most likely going to be happening here? Yeah, that's actually, That's actually the criterion, right, that if you'll allow me, I'm going to use another analogy that helps. Okay. So let's say that I'm proving a mathematical theorem, okay?

The way that works, you write down your axioms, right, your assumptions, you put down your equations, and then you have a bunch of kind of a toolkit for doing deduction. You apply calculus or group theory, whatever you like. And out pops an answer, right? And the correctness of that answer has nothing to do with your psychological state, nothing to do with what you have for breakfast in the morning, nothing to do with the economy, but nothing to do with neurons. It has to do with the formal logic of mathematics.

And that's the most beautiful example of emergence for me, because you wouldn't gain additional insight into the correctness of the proof by knowing what the neurons were doing. They're irrelevant. And the technical term for that is screened off, truly emerging phenomena, screen off microscopic degrees of freedom. That's this fancy language. And we know it's true, right? But we know it's true of mathematics.

It's probably the best example, I think, of a truly emergent language. And because, to your point, it does predict, it does deduce, you do get to the right answer, right? You don't have to go down. So I'm curious about something here. So when we look at flocking birds, that is a macroscopic group behavior where, as far as I understand it, you cannot, there's no known way to analyze a single bird in any way that will tell you that in the company of other birds it will flock. And I've always thought of

emergence as just such a system. And you hinted to that with the gas particle and the gas as a fluid or as a solid or as a gas. Well, yeah. Depending on temperature. The molecule. as a solid liquid or gas. So would you agree that these other kinds of systems, you can't look at a termite and say, one day it will build a termite mound? Is it because we don't know enough about it? Do we need to be more reductive or less reductive

in our analysis of the organism to know what it would do macroscopically in a group? Yeah, that's really interesting. So, okay. So I think sometimes that and sometimes the opposite. So let me, again, give you an example. So you're right. Yeah, he's right.

No, it's really interesting because let's say it's sort of interesting, right? If you said, I know everything about the neuroscience of an ant, right, or a termite or a starling, you know. I know about fluid dynamics, hydrodynamics, I know feathers do. I know how far they can see, you know, all of that. So I could predict if I put a bunch of them together how they would behave. And I think that there are going to be cases where that is true. But that doesn't mean emergence isn't still useful. Because you might say, yes, I can.

I need a deep thought, the computer from, you know, hitcher guy to the galaxy, right? To work it out. And it took the lifetime of the universe to do so. But I could do it as opposed to, you know what? Neil, I've got a pencil and paper here. I'm going to write down my little emergent theory and I'm going to do it in five seconds. And so there is a

site to emergence which is just about efficiency. Yeah, yeah. Efficiency. Okay. So we're going to see an equation on a T-shirt sometime soon. We have many of those. Unfortunately, we have too many of those. Too many equations, right? All right, well, I mean, Chuck mentioned AI earlier on

and we don't want to roll it down the hill because it's too expensive. Absolutely. Can you predict the emergence of consciousness in something like an AI. Wow. Or define the complexity of life itself in this context. You want to go even deeper, fair enough. Yeah, I mean, why not?

Well, yeah. I mean, honestly, and wouldn't that kind of be the same? Because if an AI really does have emergent consciousness and truly, truly emergent intelligence, then it's life. Then it really is us at that point. It's just us at that point. Or is it? No, it's us in a different form.

Well, let's find out. Okay, let's find out. I might annoy you now. Okay, go ahead. So the first thing I'm going to say is that AIs have no intelligence. Right. Okay? And then we'll discuss what that means.

But they have tons of capability. And I tell you the difference. I mean, here's my thought experiment. I was asked these kind of zealots of the technocratic era, which is the following. You have two students. Okay. And let's call them A and B. And you set them the same exam.

It's such a general knowledge. quiz, right? And they come back, we've got all the answers right. And I said, which is the better student? You said, I don't know, they've got the same answer. Now I say, you know, A, did the exam in the library, where every time a question came up, they looked up the answer, and B actually took the exam, I don't know, by the side of the ocean, you know. And you said, well, clearly B is the better student. Now, the problem is, so knowledgeable, we know the difference between fake knowledgeable and real knowledgeable because we can ask

did you do it in a library or not? And the problem with a lot of AI at the moment is it's basically fake intelligence as far as I'm concerned. It's a very quick lookup. It's really just... It's very quick, essentially a really clever look up.

It's a plus, and I'm not saying it's not an amazing technology, I'm just say, but it's a very capable technology. If you ask, and again, now we get to intelligence, kind of my field, what is intelligence?

Intelligence is basically someone or something that makes a hard problem easy. If you went to school and you're sitting down trying to work out a problem, you look over at the person next to you, and they've made a problem look effortless, you'd say, oh my God, that's pretty intelligent. No, no, no. You're the intelligent one looking over the shoulder,

because you got the answer easier than they did. That's true. That's like a strategic intelligence. If I may echo your story with an example I give often, where let's say I'm an architect and I'm going to hire, a summer intern and they're both the same on paper and so therefore they get

to come in for an interview, right? And I want to pick one from the interview. Okay. And this is a contrived example but I think you'll agree, David, that there's a church steeple outside my window and I say, just for Grins, how tall is that church steeple? And the person said, oh, it's 135 feet. I will say, well, how do you know?

I memorized all the church steeple heights, it's a thing I do. The other person says I don't know, I'll be right back. It goes away for 10 minutes, then comes back and says, somewhere between 130 and 140 feet. I said, well, how did you find out? He said, well, I know how tall I am, and I measured my shadow.

Then I measured the shadow of the church steeple, and then I did some simple math to get this answer. Who are you going to hire? I'm actually going to hire the second guy. The second one. Because clearly what he was able to do was problem solved. I didn't say he or she, but that's okay. Oh, I said he.

I'm sorry, because they were able to be saved. We're able to problem solve. And is my example resonate with you here? Yeah, very much so. And I think you can see where in notice, right, that in the ear of the cheering test, the imitation game, you just ask how high is the steeple. And if it gives you the right answer, you say,

look, there you go, it's indistinguishable from another human being, but you then went a bit further and asked for an explanation, right? Tell me how you arrived at that answer. Prove to me you understand. And I think these ideas of it, understanding and explanation

are really important to intelligence and under-discussed in place of what Alan Turing did. And, you know, he's my compatriot. I love him. But the idea of the Turing test did a lot of harm because it allows for this possibility

essentially of cheating. I had not thought about it that way, but you've just convinced me. Because that's the litmus test that has always been applied. And then everyone is left thinking there's intelligence on the other side. But they would just unpack that and lays it bare for what it is. However, in defense of Turing, I think it was just the wrong terminology because basically the idea was you wouldn't know the difference. That was the test. You wouldn't not know the difference.

Correct, correct. So he's not necessarily saying that there's not. the same. He's just saying that one is represented in a way that is indistinguishable from the other. Unless you went further and said, how did you figure this out? Well, now you just screwed him. That's right.

David, you spoke about your research into intelligence in the universe. On Aeon, you published an essay, Problem Solving Matter, September 2024. Eon, Aeon, Eon. Yes, it's a journal. Yeah, either way. Yeah. Thank you. And you suggest that life is less chemistry and physics, more like a computational process

that is born out of our need to be problem solvers. Hmm. You're going to need to be some... Shots fired there. No, no, no, no. You're going to need to do a little bit more on. packing there, because that's got people thinking, now I've even just said that much. You can only say that when there's not a biologist within a mile of him, then he can say that.

I think, David has the capability at the answer this. You'd think, all right, go for it. So in that paper with my co-author Christopher Kempe's, we address this question of problem-solving matter, transcends its materials, which is essentially that question, right? But let me give you, here's the thought experiment, and it's extraterrestrial, you're going to like it, which is, so you imagine some. extraterrestrial being visits the earth and they want to know what a computing device are right or is and they arrive on the earth at the time of the jacard loom which was like a first essentially digital computer and they say our digital computer is something mad out of wood out of silk right and uh right okay and it's all right and then

with a foot pedal right exactly right and uh and it's used to be right and it's used to make beautiful items of clothing. That's what I give you. Wait, okay. So, you know, 50 years pass, they come back again, maybe a bit more, 75 years. And they said, what is the computer? It's what is this thing with these giant vacuum tubes, right, thermionic valves, and they're made out of courts, and they're made out of, you know, mybdum and so on. And another 50 years passes, and they come back and say, what's a computer? Actually, no, that's not a computer. It's not something made of wood and silk. It's not something made out of glass and tungsten or whatever. you want. It's something made out of metal oxides and it's based on transistors. And what you realize

is, you know, computing isn't about the material. It's about the logic. They all implement a logic. Now, they implement, as it happens, binary logic, Boolean logic, right? And we know that if you have enough transistors and you put them together, you can do computations. So this is actually another thing that came from Turing that he got completely right, right, is that it's not the material, It's the logic, and you need the material that can instantiate the logic. So not any material will do. Unfortunately, the history of the study of the origin of life has been obsessed with the material because we're made in material and we're the only example we know.

And so you look at the materials that we're made out of, or all life, and you kind of reach this weirdo conclusion that that's the only way it could be done. That's clever and insightful. What that also does is it distracts. No, it misleads us into projecting what things might be like in the future. So for example, in 2001 a Space Odyssey, that was 1968, and they're imagining the year 2001. That's the whole point in the movie. So in 1968, computers were like big, and in 2001, the computer was even bigger, okay? And it was even more centralized. And no one is thinking that we'd all have computers in our pocket. So they're thinking, that it's the material, now just get more of that to make that happen. So it completely distorts

how you might be thinking about the future unless you take David's sensibility to task here. True, true. In addition, could we then think that the materials are us and then seek to replicate not a silicon version of ourselves, but an actual biological merger of the AI

and what causes us to have true intelligence. Again, I mean, two things there. So one is just a lot of the things and concepts that we've been wrestling with, even, you know, consciousness that you asked about, life, intelligence, in their early phases of development, get mapped onto a machine or mechanism or matter that's familiar to us. And there's a good reason.

I mean, that's nothing wrong with that. It makes perfect sense. You've got to start somewhere. Right. And then as our ideas evolve, they become in some sense more abstract. And eventually we culminate in a kind of logical description. But the material matters. Right.

And so to your point, it's really interesting. I mean, and this is an unknown question. There are people out there, and they call themselves functionalists and Turing was one. And he wrote this beautiful essay where he said, I mean, he didn't say it this way, but he essentially said,

I didn't give a shit whether the brain has the consistency of cold porridge. He said, the matter doesn't matter. And, you know, that was his view. But, you know, we don't know that's true. Matter does matter. The matter matters.

Matter matters. I'm all for matter, mattering. Right, but does any kind of matter matter, right? And I think that's sort of the question. And I think, and it is an open question whether the kind of sensual, sensual existence we experience does depend on the particular kind of matter we're made from.

Does consciousness depend our kind of consciousness depend on our kind of matter? I actually think there's a strong claim to be made the answer is yes. It doesn't mean there couldn't be other kinds of life, right? Other kinds of intelligence. Right. But this one depends on our matter. Well, it's the one that we know for a fact does because it's us and we're experiences. it. So from an experiential frame of reference, we understand it. So my point is, why wouldn't we just look for

exactly how we become conscious and intelligent in our formation? And then if it truly isn't something that just happens because all of these disparate things come together, then we might be able to take that and graft it on to a machine of our making. Well, you could argue that that's exactly what's just happened. Uh-oh. So if you look, well, look, I mean, if you look at the history of AI, 30 years ago, it now gets called gofi, good old-fashioned AI. And it was all about, we're going to build a computer that can play chess using symbolic

logic, checkers. We're going to build expert systems that we're going to inform with human understanding. And then this big shift took place in neural networks. And they said, you know what? We're not going to start top down. We're going to start bottom up. And we're going to start bottom up with a system that resembles a brain. And it's exactly, so they did exactly what you just said.

They'll say, let's just try and rebuild something that looks a bit like a brain with lots of units that are kind of light neurons that are connected kind of light neurons. You have enough of them. They'll do something interesting. So I would argue that this kind of biomimesis approach to intelligence that you're describing is the AI revolution of the current moment. Okay. All right, but I got to get to the bottom of something here. Let's bring back emergence into the conversation.

We have neurons that ostensibly are nicely suited for our survival. Okay. When we're hungry, we look for food. If there's danger, we escape it or fight it. and so the brain is doing its thing. And any creature has similar, any creature that cares about living. Similar functionality in their brain.

Exactly, in their brain. But we want to say that we have consciousness as something beyond what we might ascribe to a plant. So what is going on inside of us, either in complexity or from bottom up or top down, that you can call consciousness. And the reason why I ask that in that way is everyone is making a big deal of consciousness today. And the fact that people still writing books about it is evidence to me that we still don't know what it is.

Because if we knew what it was, the last book would have been written and there'd be no further books on the shelf. So, but everyone's talking about it like we fully understand it. And so can you give me some access to consciousness given your tools that you have built to ask questions? yeah i i i share your skepticism i think a lot of this is just baloney um the consciousness stuff to be honest and i think we really don't understand it and hence more and more terrible books being written on the topic oh i think this is a good time for me to announce my forthcoming book the last book on consciousness there are well you know this there are just many schools of thought here. And in my world, most of the rigorous work, and I'm not saying it's great work,

I'm just saying it's rigorous work, is looking for quantifiable metrics, measurements that correlate with the conscious state. So let me make that clear. So I measure your brain, right? I write down some equations. I calculate some number. And I measure your brain when you're sleeping. I measure your brain when you're awake and solving a problem, and I measure your brain under anesthesia. And it turns out that that number that I calculated, I say, wow, look at that. That number is high when you're waking and solving a problem, and it's near zero when you're under anesthesia or sleeping. And so this is sometimes called the neural correlates approach to consciousness. You don't tell you what it is. It just says that there's some formalism that allows

you to know. You found a correlation. Right. You found a correlation, right? And maybe that's useful, right? If you go under anesthesia and you're going to have your big toe removed, I'd rather that thing was near zero than at its maximum. But that's sort of the best of it. When it comes to actually theories of what it is, honestly, qualitatively, it seems to be something about the tiny little attention window that the human brain has to operate on large sets of data. And just to be explicit about this. Every mathematician knows that every hard problem is solved by their unconscious mind. There is a very famous book written on this by someone called Hadamard, and it's called the mathematician's mind. And he interviewed everyone, Institute Einstein, Puancaray, looked at the

journals of Gauss. And they all say the same thing. They say, you know, it's a really hard problem. The best thing I can do is think about it and then stop thinking about it. You know, I have a nice meal, I go for a run, whatever you do. And then somehow, through some epiphany, The solution presents itself to me. I'm pretty sure Einstein didn't go on runs. I just pretty sure about that. Yeah, yeah. You played the violin.

I don't think he was a fitness guru. You smoked his fight. You know, he did something in place of going for a run. A plate of his violin, even, yeah, sure. You played his violin. You went walking with Kurt Gertl. He did this thing. But the interesting point there is that,

so consciousness is not about solving the hard problem. It's about that little window of attention that is focused on some part of the problem. And most of the current formalisms don't really give us much of an insight to how that might work. So I actually do not like a lot of this stuff, to be honest. So in the, I didn't read the book, but I saw the film I-Robot, which is based on the story by Isaac Asimov.

The Will Smith one? The Isaac Asimov. But yes, the Will Smith story. The Will Smith story. Yes, the Will Smith story. So the robots, all humanoid robots, there are these large vans that have robots that are not that are decommissioned or that but they're still kind of alive but they just have no purpose until they're programmed for their utility to be your partner to be your whatever and in the van the robots grouped with each other they weren't just maximizing their distance and their pattern was not random and someone asked about that, what do you know about this? And they said, we don't know what's causing

this. What we do know is that there's a lot of residual programming that was never fully cleared out when we added new utility to these things. And this is exactly evolution, okay? There's leftover stuff in us from a time that we don't need it anymore. So what the hell is it doing in our head? It's the reason why sometimes I feel the need to eat flies. That explains a lot. It's my reptilian brain. Your gecko brain. Yeah, exactly.

Just going crazy. So I was intrigued that it was the leftover programming that was not refreshed in the continued evolution of humans or in the case of those robots that continued layering on to the functionality of them. And there's legacy software that you don't know what it's doing. So I was intrigued by that. I just want to share with you that observation. Yeah. I mean, again, I mean, a lot to talk about. And I think, you know, actually my colleague here, who you probably all know, who recently passed away, Cormac McCarthy, the novelist, he wrote a beautiful essay on this, that he called the Cacule problem, which is about this moment of insight. And this was the discovery by Cacule of the Benzine Ring. I say, oh, he saw it in a dream. And Cormat was fascinated as a writer, as a novelist, with this question of, where is this coming from? I'm sitting down to write a book.

And somehow my brain is instructing my hand that I couldn't tell you exactly what's going to happen at the end of that sentence. It's this sort of coming out. And so he, over the course of time, introspectively, came to believe that he was getting these instructions from his unconscious mind. Now, I'm not in a mystical sense. Just he wasn't working it out, right? And to your point, if you look at the history of life on earth, most of evolution up until very recently took place without language. And presumably, most organisms being run by a set of automatic programs of the kind that you just described in the van with the robots huddling, like Starlings. And that we superimposed above that this kind of very thin layer of abstraction and self-awareness.

But most of the computation is not being done by that thin layer. And so what is true, right, is that that little bit of that. is that that little thin layer gives us one thing that we're not aware any other animal can do which means that we can communicate our understanding

they can communicate other things but we can communicate our understanding you can give you Newton's laws I can tell you about Darwin's theory of evolution and that superpower of humans that comes from very few neurons I imagine

sits on top of exactly all of that programming that evolution gave us over the course of hundreds of millions years. So I got one last question for you. But I've still got more. But you asked your question.

We're asking more questions. I'll sit and talk to myself then. We're in rap mode. We're in rap. Not the MC. No, I know what mode do you mean, yeah. We're in rap mode here. So I just have to go there because it's, it fascinates us all.

Well, I think about it all the time. I can't speak for others. Can you estimate, based on your toolkit, how intelligent we are relative to how high intelligence can get in the universe. So are we smart enough to figure out how the universe works?

Are we just complete idiots and some higher alien is just going to come out and just look at us like we are earthworms in our capacity to deuce the nature of the world? Yeah, that's really interesting. I mean, one of the areas I work on is on tools and artifacts, you know, the abacus, the sextant, the quadrant, the Rubik's cube. Yeah, all that good stuff behind you.

Here comes the shelf. Got my sexton here, and I've got three Rubik's cubes up on the other shelf. Just show and tell. And I'm sitting on the abacus. I hope not. I mean, there's an important point in that, because human intelligence has always been about ingenious outsourcing to artifacts and tools, including mathematics, right? You could not calculate the orbit of a planet without conic sections or the calculus, right?

And I think that, so I think human intelligence in that respect is unlimited because we'll just continue to build tools that are kind of adjuncts to our capabilities. And what makes AI interesting... I just want you to know that I can compute orbits with this abacus. Nice. No, I love it. No, I'm lying. I'm totally lying right there. Yeah, this is an authentic Chinese abacus. Yeah, I can see.

I can see that. But I think, so just to your point, I think that what's really, and I, okay, this is, I'll just tell you very quickly, I classify tools into two categories. What I call complementary cognitive artefacts. That's like a pencil or an abacus or a sexton. And there's another kind of tool that I call a competitive cognitive artifact. And that's like a GPS machine or a large language model, right? One of those sets, the complementary ones, makes you smarter.

One of those sets makes you dumber. And I think it's the choice of humanity to decide what kind of tool it wants to be dependent on. And so my fear now is we're outsourcing our capabilities to competitive artifacts and not to things like future abacai, which actually would make us smart. Wow. Do I have time for my question I've got? I don't know, I don't think so. Well, I'm asking it anyway. Go ahead ask you.

All right. So, David, you said life is problem solving. So why has the universe created life and what is the problem it's trying to solve? Oh. So I, you know, I can tell you the horribly cynical answer to that question. Go on. Yes, please. this horribly cynical answer to that question

is that life is the most efficient way of returning to thermodynamic equilibrium. Oh man, that's terrible. Because life is the most efficient generator of entropy. Right. And if you think about what we do when we build factories and what we're essentially doing

is we're turning ordered states into disordered states and that the cynical answer your question really is, is life is a kind of suicide by the universe. Yeah, I was about to say, what you really just said is, it's all for nothing. But that's not true either. It's what we make it.

Is there an answer that isn't cynical, just so as we can happen on a... It's such a downbeat. No, I know. I know. I like the cynical answer. I love it. And I think the non- cynical answer came from the sort of idealist philosophers and they said life was the universe's way of knowing itself and that's also true

so that's the non-synical version yeah that's very poetic yes i like that that was uh cosmos 1980 okay that was a major theme life is a way for the universe to know itself yeah yeah all right we got it we kind of have to like end it times up right but we could have gone on three more hours yes clearly Clearly. Well, we're delighted to first meet you and to hear what you're into. And I'm glad as president of the Institute, you get to still do work in your favorite topic. Yeah, but the question is, are you glad? I am glad. I am glad. You can't do this job unless you also do science. Yes, good. That's as it should be and often how it's not. Yes. So, David, thank you for joining us here on StarTalk.

Thank you so much for having me. And these horribly dark tines, so I appreciate the fact that we get to talk about intelligent things in a stupid world. I value that. Thank you for that. There you go. There you go.

All right. Chuck. That's always a pleasure. Pleasure, Neil. Feeling good here. This has been StarTalk's special edition, the complexity version of our special edition. Neil degrass Tyson here, as always, bidding you to keep looking up.

You know what I'm going to be.