Into the Impossible With Brian Keating - There's a New Law of Nature — And It Changes Everything We Know About Life

Episode Date: March 23, 2026

Please join my mailing list here 👉 https://briankeating.com/list  to win a meteorite 💥 Michael Wong recently discussed theory of evolution on TED: https://www.youtube.com/watch?v=DSrf7ErdHWA  — in this conversation, we go deeper on the law of increasing functional information and what it means for life, complexity, and the future of science. 🔔 Subscribe for new episodes each week  🎧 Ad-free episodes on Patreon: patreon.com/drbriankeating  INTO THE IMPOSSIBLE — where Nobel Prize winners, physicists, and bold thinkers explore the biggest questions in science. Is there a second arrow of time? Astrobiologist and planetary scientist Dr. Michael Wong joins Brian Keating to explore the law of increasing functional information — a proposed new law of nature that may explain how complexity evolves across minerals, biology, AI, and the cosmos. Dr. Michael Wong of the Carnegie Institution for Science has been working with a cross-disciplinary group of scientists — the "Missing Law Group" — to propose something bold: a new law of nature. Their law of increasing functional information argues that evolving systems, whether biological or not, tend toward greater complexity and function over time. In this conversation, Brian and Michael unpack what functional information actually means, how it differs from Shannon entropy, and why it may describe something fundamental about the universe we live in. From the formation of minerals in stellar atmospheres to the evolution of life on Earth, Michael walks through how this framework applies across planetary science, astrobiology, and even cancer research. Brian pushes on the hard questions — the Fermi paradox, the boundary conditions of the law, its relationship to the second law of thermodynamics, and whether it truly qualifies as a "law" at all. Sean Carroll blurbed the book and later called it out on his podcast — Brian asks Michael about that tension directly. They also get into panspermia, the contingent role of cosmic collisions in shaping Earth's evolutionary history, the search for biosignatures on Mars and beyond, and what the rise of generative AI looks like through the lens of selection for function. This is a wide-ranging, technically honest conversation about one of the most ambitious proposals in contemporary science. Key Takeaways: 00:00:35 There May Be a Second Arrow of Time — and It Points Toward Complexity 00:01:35 The Law of Increasing Functional Information Explains How the Universe Evolves 00:07:15 Functional Information Measures How Well a System Performs a Specific Function 00:11:30 Three Universal Selection Pressures Drive All Evolving Systems 00:18:00 Mineral Evolution on Earth Is a Measurable Proof of the Law in Action 00:22:00 The Law Is a Tendency, Not a Guarantee — It Doesn't Resolve the Fermi Paradox 00:27:00 Life and Non-Life Contain the Same Building Blocks — the Difference Is in the Distributions 00:32:15 Cosmic Collisions Didn't Derail Evolution — They Opened New Possibility Spaces 00:50:00 Cancer Behaves Like an Evolving System — and That Could Change How We Treat It 01:05:15 AI Is a New Form of Evolving Life — and Without Stronger Selection Pressure, It's DangerousFeatured Guest ➡️  Follow Michael Wong  🌐 Website: https://miquai.myportfolio.com/  📚 The book: Time’s Second Arrow (co-authored with Robert Hazen) : https://www.amazon.com/Times-Second-Arrow-Evolution-Nature-ebook/dp/B0FJ2ZJL53 🏄‍♂️ Twitter: https://x.com/miquai — Brian's Links Join this channel to get access to perks like monthly Office Hours: https://www.youtube.com/channel/UCmXH_moPhfkqCk6S3b9RWuw/join  📚 Get my books: Think Like a Nobel Prize Winner: https://a.co/d/03ezQFu  Focus Like a Nobel Prize Winner: https://a.co/d/hi50U9U  Losing the Nobel Prize: http://amzn.to/2sa5UpA  Dialogue Concerning the Two Chief World Systems (audiobook): https://a.co/d/iZPi9Un  Follow me to ask questions of my guests:  🏄‍♂️ Twitter: https://twitter.com/DrBrianKeating  🔔 Subscribe: https://www.youtube.com/DrBrianKeating?sub_confirmation=1 📝 Mailing list: http://briankeating.com/list   ✍️ Blog: https://briankeating.com/blog  🎙️ Audio-only: https://briankeating.com/podcast  — #physics #astrobiology #complexity #evolutionoflife #briankeating #intotheimpossible Learn more about your ad choices. Visit megaphone.fm/adchoices

Well, first, to be just provocative, there's definitely life on Mars, right? We brought it there. Our rovers have hitchhikers on them. Microbial spores, right? They weren't completely sterilized. Not that the building blocks of life are found all over the universe. Maybe life is just made of the building blocks of the universe. We are literally generating an alien presence on our planet that co-evolves with us and was

in some way spawned by us. Today's guest comes bringing a provocative claim that we've been wrong all along and actually that there's a second arrow of time. Michael Wong has come all the way from Washington, D.C., where he works at the Carnegie. Is it the Carnegie Institute? Carnegie Institution for Science. Home of many great scientists.

What's a definition of time? How do you define time? Oh, my goodness. How do I define time? I don't know if I have a definition for time. I think that time is, you know, some axis upon which we can start to make measurements and derive,

metrics to measure things as they change through time. I'm not a physicist by training. So, you know, maybe I should leave that to the physicist to define. Yeah, yeah. They literally say things like it's what a clock measures. I mean, that's effectively what a physicist will say. Yeah, yeah. So I think, you know, time is essential because what the laws of nature try to describe is

change through time, right? if I apply a force to an object, how does its acceleration change in the future? And there's a certain kind of change through time that we describe as evolution. And so our new conjecture, our new postulated law of nature is a law of evolution, of evolving systems, of changing patterning, order, diversity, complexity in the universe that we see all around us and that we understand has occurred since the Big Bang through the general. of heavier and heavier isotopes to more and more complex molecules and minerals to planets. And then at least one planet in the universe that has life.

And this is where the thread ties to myself and why I would be thinking about arrows in time as a planetary scientist. So, you know, my training is in planetary science, but my heart really is in astrobiology. So that is the search for life elsewhere in the universe. And life is the quintessential, complex evolving system. And so when I think about how should we best go about the project of looking for evidence of this kind of complex evolving system elsewhere in the universe far away, whether it's just on our neighboring planet Mars or on a distant exoplanet,

it would really benefit astrobiology to understand the generation of complexity in general, right? And so I started thinking about, okay, well, how do you think about looking for evidence of life, especially given the increasing anticipation of astrobiologists these days, wondering about whether or not the particular instance of life on Earth is truly representative of life everywhere else in the cosmos. That is to say we have very specific sort of constraints on our biology given by our particular evolutionary trajectory that has been sculpted by the environment that we find ourselves here on Earth. So we use a very specific and narrow set of organic molecules to construct ourselves. You know, we've got five bases in RNA and DNA, which are our genetic polymers. We have

an alphabet of roughly 20 amino acids that create our proteins and a limited set of all possible lipids creating our cell walls. But life on another world could potentially use very different molecules from us. So what should we actually look for when we're looking for life? And in conversations with my friend and colleague Bob Hazen, my co-author on this book, we started thinking along the lines of maybe information as sort of this more agnostic, take a step back metric for looking for evidence of life in the universe. But then we quickly realized that information, is not only part of the biological realm. It exists in all kinds of complex evolving abiotic systems, including minerals, which Bob, you know, is a

mineralologist, and then I studied planetary atmospheres in my Ph.D., and there's information content in the networks of atmospheric chemistry on different worlds. And so maybe what we should seek is understanding information, better, information in the laws of evolving systems. and we started working with a wide range of scientists in a group that we kind of just tongue-in-cheek called the missing law group. If there's a law of complexity, a law of information out there that we've yet to discover, you know, it might be a missing law. And this group gradually expanded to include people of all different kinds of disciplines from organic chemistry and the origins of life to informatics and data science to complexity science and theoretical physics. And then even including some philosophers of science, because when we just,

decided, oh my goodness, I think we might be trying to propose a new law of nature. What is a law of nature? And who decides that? And how do these things develop? Well, history of science and philosophy of science has dealt head on with that kind of topic. And so we've got a couple of philosophers on our papers as well. And this group put together this proposal for a new law of nature, the law of increasing functional information, which we published in a paper in 2023 and forms the basis for this popular science book, this longer exposition that we want to give to the general public. Now, I bet you think I'm going to ask you, how can there be two arrows? But I'm actually going to ask you, why aren't there 20 arrows or a thousand arrows? Why aren't there as many

different arrows as there are things that evolve? You mentioned in the book, what we call the biological arrow of time, that, you know, we get gray hair as we age, and the babies start small, and chickens start in eggs. And then there's the psychological arrow of time, you know, someone I'm listening to one of my lectures, my experience a time dilation effect. That might not be so good. Then there's physical types of time, but time is relative.

Time and simultaneity is not an absolute concept in physics. So why not multiple? Why only, why are you being so stingy? Come on, Michael. Give me more hours of time. You know, I think that our proposal for a new arrow in time does not preclude there being other arrows as well. And so I think the reason why we call it the second era of time,

time is to sort of try to put it in a place that talks to the first arrow, which is the given by the second law of thermodynamics, that increase in entropy of closed systems over time. And we do have that kind of conversation about where this stands in relation to the second law of thermodynamics throughout the book. But I don't think that this means that you can't describe other arrows in time. We're just proposing that there is yet another that has to do with the increase of functional information. Okay, so let's start there. What is functional information?

What's information? What's functional information? What's Colmogora of complexity? What are all these different things? And keep in mind, I think it was Silard, who said that, you know, if you want to explain something in physics, just call it entropy, because no one understands what entropy. So first of all, what is information, functional information, and so forth.

Yeah, so there are many different flavors of information. You've mentioned already quite a few. And so, you know, you can think of information in the Shannon sense or what is often called comagara of complexity, which is the amount of information you need to describe some entity, right? And so functional information differs in a really important way. And so let me just explain what functional information is mathematically. You calculate functional information in units of bits

by taking the negative log to the base two of a fraction. And what is that fraction? So that fraction is the number of configurations, of a system in question that can perform a given function, the function that you're interested in, divided by the total number of possible configurations that that system can have. So maybe the best way of explaining this is by taking, you know, you've got a coffee cup right here, right? So that coffee cup is comprised of probably around 10 to the 25 atoms or so, give or take a few zeros. And so you can imagine scrambling the configurations of the atoms in that coffee cup on this table in all

sorts of different ways, and you can come up with Googles of, you know, different configurations. I don't know exactly how many there are. But only a very small fraction of those configurations successfully hold a hot beverage that you can then pick up with a nice little handle and then pour into your mouth, right? And so that fraction is really, really small. And because there's a negative sign in front of the negative, you know, the log base two, that functional information is really, really, really high for that system of 10 to the 25 atoms in relation. And so, you know, to their function as a coffee cup. Now, you can imagine other functions

for those 10 to the 25 atoms. Maybe you're interested in weighing down some paper on a breezy day. And so there might be a slightly larger fraction of configurations that you could stramble those atoms into that would succeed at the function of being a paperweight. So the functional information of that system as a paperweight is slightly lower than that of the functional information

of that coffee cup as a coffee cup. And then you can think of all sorts of other functions for which the functional information of that system is basically zero, like, you know, being a screwdriver. That coffee cup is not going to cut it. And so, you know, the functional information calculation is almost subjective in a sense because it is dependent on the function that you are interested in calculating for a specific system you're interested in describing.

But who is the you that determines this? If I have an egg, you know, it could either be a chicken or it can be an omelet. So who gets to decide who's determining the functionality and the fitness function? Yeah, that's great. And I love this idea of this chicken or an egg, right? It's like there's so many different ways of describing a system, so many different kinds of complexities that we might be interested in knowing about. And the chicken example is really brilliant. So you can have a chicken egg, a chicken. You can have imagine also a chicken, a live chicken, and then imagine a dead chicken. And I'll give you three kinds of complexity we might be interested in. Genomic complexity, structural complexity, and behavioral complexity. And so that's the thing that the chicken egg has the genomic complexity, right, but it doesn't have the structure or the behavior. The live chicken has all three. It's got the genomic, the structural, and the behavioral, and the dead chicken has the genomic and the structural, but not the behavioral. And so it really is kind of like what, okay, so how then do you decide what is the fundamental sort of function that you should be assessing

the functional information about? And in many cases that have to do with our cells, you know, we're interested in understanding the functional information of, I don't know, an experiment that we've got going in the lab. You know, that's very constrained to what we care about. But in the book, we try to relate the idea of selection for function back to fundamental selective forces in nature. And so there are going to be three of those. The first and most simplest is selection for static persistence. That is the ability for an arrangement of matter to just be exactly. exactly as it is over some time without bending or decaying toward higher entropic states.

And so long-lived stable nuclei or metastable crystals that formed deep within the earth but are now out of equilibrium here on the surface, yet they persist. These are examples of static persistence, selection for that which basically is able to exist. Then there is selection for what we call dynamic persistence. So the existence of these statically persisting entities in the universe creates out of equilibrium structures that can be dissipated to drive the emergence of dynamic systems like convection cells or wildfire or stars or even life, right? These are all dynamically persisting entities in the sense that the exact material structure of these dynamic systems because of the fact that they are open systems, always exchanging matter and energy and information

with their environment, you know, it's always changing. You know, our cells turn over on what, years or something like that. We're always breathing in, breathing out. There's nothing static about us, but yet what is persisting is our activities, our homeostatic activities and autocatalytic activities. And so you can talk about persistence in that dynamic sense. And there's a selection for that as well

at the level of entities like you or me or even, you know, stars and things, convection cells. And then we think that there's this third level of selection for novelty generation. Because if you're able to generate and discover new functions, new ways of being, of exploring your world, of tapping into energy sources, think of the evolution of sight or flight or any kind of locomotion for that matter. You know, there's a selection pressure for being able to generate new configurations that can then persist dynamically better, or at least in a new way. And so those three selection pressures for static persistence,

for dynamic persistence, and for novelty generation, I think, is a way to tie back what function are we actually talking about when we should perform a functional information calculation about a natural system. So how do you determine the scale? Like I always hate it when physicists, myself included, on occasion. So I think, oh, like, entropy is, you know, you can envision a cup of coffee and a cup of cream. And they're separated.

That's very low entropy. It makes them together. It's very high entropy. But that's only because we are rescaling the problem to something that's much more macroscopic. If you were to say, what's the entropy of the coffee cup at the level of quarks or quantum fields? It doesn't change at all. It's basically water and water with, okay, there's a couple more fat model, whatever. The amount of information is basically the same.

And so you would be not able to tell it if you didn't have this massively simplifying operation of filtering by what looks blended together. and you couldn't maybe separate, but it is basically identical at the level of quarks, I'm saying. So how do you determine what scale to start assessing or calculating this logarithm at? Yeah, no, that's a really great question.

And it, of course, speaks to the boundedness of this law in terms of the scale that you're thinking about. So I completely agree that at the level of quarks are the most reductionistic, whatever, the quantum fields and things like that, the amount of information in that sense, you know, of like the amount of information you need to describe the fundamental

particles of that coffee cup is constant through time, right? Every particle has a position and a momentum and that, you know, is what you need to describe that system. But when we get into what we call the macroscopic laws of nature, that's essentially what we're trying to propose here is a new macroscopic law, a law that pertains to our everyday lives, something that you can see in existence all around us as our planet evolves, as we, a species evolve, as technology, AI, continues to evolve at a breakneck pace. These are the types of things at those scales. And what exactly the boundary is?

I don't know. I think there's probably a fuzzy boundary. But for the things that we care about macroscopically, for evolving systems, I think that this law of increasing functional information is meant to try to describe those. Let's apply this to your field of, you know, just crushing domain expertise over anything I could imagine, which is planetary science. Take me through a calculation or the framework or even, you know, qualitatively. How does this apply to, say, the assembly of a planet or, you know, the format? What is being selected for?

What is being evolved? What is surviving? All these Darwinian principles, which are paramount throughout this book. Take us through the formation of Earth. I have a meteorite here. Oh, yeah. yours. Oh my goodness.

Oh, you're so kind. And if you are like Michael and you have a dot edu email address and you live in the United States like Michael, you will win a meteorite guaranteed, Briankeetn.com slash edu. If you don't,

I give them money sometimes to people that don't have EDU accounts in the U.S. So take us through the formation of that. How did that get here? I can tell you, you know, how I acquired it. But I don't want to ruin my secret source. I got a guy.

I got a guy. But tell me, like, why did that get here, but it's not a chunk of uranium or something like that? So you talk a lot about iron. That's an iron meteorite. So this is formation of this thing. Who's selecting? What's selected for?

What's the function? Take us through, and keep in mind the most magnificent audience in the multiverse that has been selected for watching this podcast. So don't be afraid to be technical. Yeah. No, of course. Now, this is lovely. First of all, thank you so much, Brian.

I will cherish this. Iron meteorite, which is basically the core of a proto planet. they got smashed apart and fell to Earth and now it's in my fingers, which is just an incredible story to think about as a planetary scientist. So yeah, I think that the best way of trying to explain one of the key concepts in our book is the evolution of things like this, of minerals through time. And we dedicate quite a number of pages to exploring this and actually trying to quantify it in the book. And so, yeah, just to state the big idea first and then we'll get into the detail. is that the law of increasing functional information

helps us understand evolution and evolutionary theory beyond the narrow constraints of biological evolution. So Darwin did a great job of explaining how species evolve here on Earth, how we get new complex structures and increased diversity and things like that. Okay, so that's great. And we all understand how Darwinian evolution by natural selection works. What we're saying is that you can see evolving systems not just in life, but in things from the abiological realm as well.

So minerals evolved. That is to say that they have grown in diversity and complexity and patterning and distribution over time, starting from minerals that first condensed in the stars' atmospheres, you know, early on to the minerals of planet formation, like the iron, nickel, alloy that I'm holding in my hand right now, to the minerals of the first crust of planet Earth. And then as that crust recycled itself and underwent this process that we call plate tectonics, new metamorphic minerals emerged. As water-rock interactions took place, that birthed a new generation of minerals.

So minerals continue to evolve and to be produced as planetary environments sample new pressure, temperature, and composition spaces. And as planets evolve, they subject their material. to these new conditions, which allow new minerals to be generated, there was a time in the universe before there were any planets whatsoever and before the periodic table of the elements was filled in all the way. And so the possibility space of minerals was very, very small back then and has grown enormously over the past 13.8 billion years.

And then the actualized space of minerals, those that have been actually produced in stellar and planetary environments, has also grown. And so one can do a functional information calculation, for minerals with respect to the function of static persistence, of just hanging around for us to be able to study them. And that functional information calculation looks like information in units of bits

is negative logs, the base two of a fraction. And that fraction in the case of minerals is the number of minerals that we see at a certain time of Earth history or solar system or even the history of the universe, divided by the possibility space of minerals at that time. And we can actually quantify the these numbers. Functional information, one of the trickiest things is actually that it's very

hard to quantify because it's hard to define and know what those possibility spaces are for many complex evolving systems. But thanks to the fact that people have been studying minerals, especially my colleague Bob Hazen, who has compiled a whole system of mineral evolution, that data is right in front of us. So we were able to very quickly put together the functional information calculation of minerals through time, again, from those first minerals born and stellar atmosphere, so the minerals of planet formation to the minerals of the Hadean Earth, all the way up to the minerals generated by metamorphic and igneous processes, and then, of course, life. Life is responsible for roughly a third of Earth's present-day mineralogical diversity through

processes known as biominerization, the things that living things precipitate and influence. And so we are able to see this monotonic trend in the functional information of minerals over deep time. And that was really, really interesting for us to see. Does this have any impact on, you know, the Fermi paradox, for example? I mean, if it were very likely that things tend towards high functional information, then we would expect to see an abundance of life, which we don't see, right? I don't think I'm miss speaking here, but I don't think there's any plausible evidence for life outside of the, of Earth right here, right? Right now, not evident, hard evidence. It doesn't mean there's no possibility. The log space is huge,

right. But what does it have to say about the potential resolution to the Fermi paradox, perhaps? Oh, that's a really great question. So this tendency for evolving systems to grow in functionality and hence in their functional information, I think should not be used to, you know, to try to argue for or against the plethora of life across the cosmos. And the reason is this, it's a tendency law, just like the law of thermodynamics, the second law of thermodynamics, the universe has a tendency toward higher entropy. Here we're saying that evolving systems, those that experiment with combinatorial diversity and then are subjected to selection for function, have a tendency toward higher functional

information. But it doesn't say at what rate or how probable it is that certain transformations or leaps in complexity must be. So the evolution of life on Earth, especially that toward complex life that could explore the cosmos by, as Sarah Walker says, anti-accreting itself into the cosmos. I think that's a beautiful phrase. We don't know what the probability of those transformations might be. And we'd say that the universe has a tendency to produce that if given conditions, again, combinatorial exploration followed by selection for function, but is it going to be likely on any planetary

environment that you will eventually evolve that kind of functionality? We're agnostic to that. Okay. So you just mentioned Sarah Walker. She's been a guest many times on the podcast. podcast. She talks about assembly theory as sort of tracing the contingent history of different objects and their fitness for surviving them. How does this, you know, relate to the assembly index and the functional information? Are they related through a logarithmic transformation or a foray transform or something, Michael? Help me out. I would love, if that were the case for the theory that Sarah and colleagues are working on assembly theory to somehow, you know, mesh with what we're talking about in terms of functional information.

So what I can say, well, first of all, for the audience, maybe we should recap what assembly theory has to say. So if I am characterizing assembly theory correctly, it would go something like this, that you can basically quantify the complexity of objects in the universe. In other words, the amount of selection or the amount of information needed to generate those objects, or as Sarah has been calling it, how deep they are in time, through two numbers. their assembly index and their copy numbers. So imagine an object. Its assembly index is going to be the number of unique transformations required to construct that object. And that could be, you know, if you're thinking of a word,

how many, you know, arrangements of letters you need to create that word. If it's a molecule, it would be how many, you know, different molecular transformations through bond additions you might need to make that molecule. So that's the assembly index, how complex that object is. And then its copy number is just how, it's number density, how much of it you find in a given unit of volume in the universe. And those two give you the assembly, which is the complexity of that object. So the way we're thinking about looking at, you know, objects and the amount of selection having to do with their generation is at a different scale than assembly theory.

So we're looking at collections of objects in systems. And so both assembly theory and our work deals with the question of how to look for life in the universe. And according to assembly theory, it's by looking for these high assembly objects. And for us, it's by looking at the distribution of objects in a system in question. And this is a prediction of the law of increasing functional information, that the selection for function in biology should skew the components of biological materials toward distributions that look very different from the distributions of those components in abiological sources.

So let me go back to this lovely, lovely meteorite that you gave me, Brian. So in primitive bodies like bodies that this meteorite came from, you know, asteroids and the asteroid belt, the leftover planetesimals from planet formation, we have discovered the building blocks of life, you know, nuclear bases and amino acids, you know, all the building blocks of life seem to be just scattered in these most primitive objects in the cosmos. The Osiris Rex mission has shown this for asteroid Benu, for example. But we also have seen them in carbonaceous chondrites that have fallen to Earth. And so that seems really cool. You know, when people discover this, it's like, oh, the building blocks of life found in an asteroid.

But another way of thinking of that is, okay, well, maybe it's not that the building blocks of life are found all over the universe. Maybe life is just made of the building blocks of the universe, right? There's no mistake that we're made up of these particular organic molecules because, you know, the universe generates them. Right. Which begs the question then, how do you differentiate between a suite of organic molecules taken from some black sludge on an asteroid that contains all the building blocks of life? and a cell, which also contains all the building blocks of life, right? The difference between those is in the distributions of those building blocks. They may contain the same building blocks,

but life has been sculpted by selection for function, skewing those building blocks toward high abundances of those that perform biochemical functions, whereas the contents of a meteorite or of an asteroid, well, those organic building blocks have been selected purely for their static persistence of just hanging around in space for 4.5 billion years, waiting for a spacecraft to go and grab them and bring them back to Earth for analysis, right?

So the distribution is different, and we've been able to show through some of our biosignature research that you can train machine learning algorithms on the patterns of these organic molecules and samples that are alive and not alive to then accurately characterize a sample that you've never seen before. So, you know, you go grab something from another world, you break down all of its organic chemistry,

you look at the building blocks and the relative distributions, there could be hundreds of thousands of different organic molecules in that sample, but if you have a machine learning algorithm that has been trained to recognize patterns in that hundred-dimensional space, you can learn something about the provenance, the biogenicity of that sample. And so, you know, we're looking at collections of samples. Assembly theory is dealing, sorry, we're looking at collections of molecules within a sample,

and assembly theory is looking at quantifying the complexity of individual molecules, and I hope that our approaches provide complementary avenues for searching for life in the universe. Hey, everybody. I'm usually the one that asks my guest to judge their books by their covers, but today I'm asking myself to judge my own book by its cover. My newest book, Focus Like a Nobel Prize winner, is chartful of advice, life tips, and focus and productivity tips from nine of the world's greatest minds.

Nobel laureates ranging from economics to peace to physics, of course. So go to Amazon and get the KindleCopy today. So staying with the meteorite theme because I enjoy it so much. The fact that you and I are having this conversation is the result of at least, well, you'll tell me, I mean, I'm sitting in one of the world's great experts on this, but at least four different types of enormous cosmic collisions, the Thia body collision with the Earth that created the moon, then allowed a lot of tidal mixing and allows some gravitational protection bodyguard functionality of the moon.

the great bombardment events that ceded the planet with water, without which we wouldn't have the universal solvents that life seems to depend upon. And then we would be out-competed by enormous lizards, terrible lizards, dinosaurs, if not for the Chicks-Love impact in the Jurassic period, 66 million years ago, right? So at least three different, you know, major bombardments of the earth had it take us to us being the last, you know, and so-called pinnacle, of this functional evolutionary pyramid. So how do you couple into extraplanetary,

you know, kinds of bits of information addition or deletion? There's another theory that suggests that life may have come from earth-bound meteorites and a theory called panspermia, which I'm anxious to know the current status of that. But all these different collisions seem so random. I mean, literally. So how can the function sort of persist

despite all these external events increasing entropy greatly and then eventually decreasing entropy because we become much more functional as the persistent surviving dominant species on Earth. So I realize it's a long question, but how do you take into account the fact that all these random events could have changed our contingent history so much that we wouldn't be having this conversation for millions of different reasons? Yeah, no, this is a great question. We are one particular evolutionary threat. We are the outcome of 13.8 billion years of evolution, but in a very small pocket of the universe that has only

experienced a very small fraction of all the possibilities that could exist across the wider expanse. And so what roles do these cosmic collisions play? Well, I see a couple of roles. So the earliest ones, the ones that set our planet up for the origin and evolution of life, really gave us a vast possibility space to play with. And so this is one prerequisite for any evolving system, biological or abiological, there's really just three factors that come into play. Large numbers of interacting components, that's factor number one. Once you have large numbers of interacting components, two, you need to find ways to generate new configurations of those components.

And then finally, you need to subject those configurations through selection for function. And so some of these large impacts early on in the solar system's history gave us the raw materials to do one and two, right, to basically have a large number of interacting components, you know, geochemically speaking, mineralogically speaking, organic chemistry, you know, and then the ability to experiment with all of these different components and generate new configurations that can then be selected upon by the environment to ratchet up in complexity. And then we've got these mass extinction events, which seems to decrease the complexity of the biosphere by making most species go extinct, right?

And so you correctly state that we are the result of one that killed off the dinosaurs, and then that opened up new niches for mammals to explore, right? So the decrease of functional information during mass extinction events essentially means that one or more of those three prerequisites for evolution has gone away. So the ability to create new configurations or to have them be selected, for functions seems to just disappear during a large impact event, right? The one that would wipe out the dinosaurs. But subsequent to that, all of a sudden, you've opened up new niches, new possibilities

to explore for other kinds of animals. And so I think all of these things are happening at once, and these large impact events are ways to open up new doors to new configuration spaces. And essential in the contingent evolution of life on earth that brought you and me to this very point. But you're right, you know, some other planet with different initial conditions and a different history of bombardment may have, you know, a different kind of conversation happening right now between dinosaurs. I don't know. And that should be probably totally well. They only had a better space program. Right, exactly, yeah. They only had Bruce Willis and

Ben Affleck. So the book is blurred by many, many outstanding intellects, Jack Shostack, winner of the Nobel Prize is a past guest on the podcast, Stuart Kaufman, who's an upcoming guest, and Sean Carroll, who blurted my book also from Norton. And Sean says, time, second arrow is bubbling with ideas all inspired by one of the biggest outstanding questions in science, the origin and evolution of complexity. The insights here help to move us towards a unified understanding of the physical and biological realms. But then later, he said in a podcast maybe perhaps with you or, um,

you and Bob, he said on his podcast, your proposal falls short of what we would normally call a law. So my question is, the second edition, are you going to rip the blur? No, I'm just kidding. What do you think Sean meant by that? Yeah, no worries. Yeah, yeah. So this is one thing that, of course, we will all struggle with and is something that isn't really up to us to decide whether or not it should be called a law or a theory or a proposal at this stage, you know, obviously, yeah, a new law of nature, right? You know what? We've got to sell

copy some. I'll be provocative. That's what the, that's what the, yeah, no, yeah. But it's, it's, so, so the reason why we think that it is appropriate to say that we are proposing a new law of nature is because of the cross-cutting nature of our statement of, of, of this, you know, law of, or this tendency of increasing functional information. And it is cross-cutting in the way, well, let's just go back to the way that natural laws develop over time. You know, a natural law basically is able to describe, explain, quantify, and predict a large number of disparate systems under a unifying framework based on conceptual equivalencies.

So what might be a conceptual equivalency from a law that we all know and love? Well, Newton's Law of Gravity, right, is basically. based on this property of having mass, which is a property that was very esoteric at the time that he proposed it. But this idea of having mass and then therefore being able to be described by Newton's law of gravity applies to so many disparate systems, whether it's an apple falling from a tree or a planet orbiting a star or stars in a galaxy orbiting a supermassive black hole, or even, you know, the tiny microscopic atoms and molecules that we see in planetary atmospheres can be described,

in part, by Newton's Law of Gravity, by virtue of their having mass. And so what we're saying here is that there may be a law of evolution, of complexification, of functional information that crosscuts all different kinds of evolving systems, whether they're biological or a biological. And I think that's what Sean meant in his blurb, that we are hopefully pushing the fields a little bit closer to understanding how complexity arises in general, not just, through Darwinian evolution in biology, but also seeing how more complex systems evolve

in physical and chemical systems. But, you know, it may not be a law, and I think that's really just up to the community to decide over time, and I'm okay with whatever outcome. When I went on Joe Rogan's podcast, I brought him a tiny, tiny sliver of Mars, and perhaps if I come on your podcast someday, I'll do the same.

Okay. No, no promises. Very expensive, Michael, as you can imagine. They really are, yeah. But, you know, it seemed to me reading this book and thinking some thoughts I've had about it before, that we do want to set some kind of constraints from your theory and others on whether or not, you know, this big question of whether or not we're alone in the universe. And I think one of the ways I kind of have tried to tackle that problem is thinking about

in the reverse way that perhaps, you know, if life could come to Earth from elsewhere, then certainly life from the earth has gone elsewhere. And I actually have a, as I said, I'm a rock from Mars. That means there's rocks from Earth on Mars. And presumably we've been exchanging material. You're the world's expert, right? With Mars for literally billions of years, long since we've had biological material on Earth.

And there must have been some time period where the sun was hotter and the Mars was greener and had liquid water flowing on it. Yet we see almost no, let me say that more specific. we see no evidence for life on Mars currently, and we see some claims that sometimes get refuted for past life. So shouldn't the non-existence of any evidence on Mars, you know, at least be a data point in the information landscape of how easy or hard it is for life to grab hold, not how it is to originate, but shouldn't the fact that, you know, we should have McDonald's on Mars right now, shouldn't that set some limits on how fertile information really is?

Wow. Okay, such a great question. And so many sub-questions backed into that, Brian. No, no, no worries. I love it. I love this conversation. I love where it's going. I would say that, well, first, to be just provocative, there's definitely life on Mars, right? We brought it there. You know, we, we've, yeah, well, we haven't sent astronauts to Mars yet that have, nobody's, nobody's pooped on Mars, but it is beyond a doubt that our rovers have hitchhikers on them. Microbial spores, right? They weren't completely sterilized. And so there is life on Mars. We definitely brought it there. And that opens up this possibility that you were talking about

in panspermia, the fact that Earth and Mars being rocky planets right next to each other in orbital space around the sun have been trading material back and forth. You've got Mars here on Earth and Mars surely has Earth over there. And so this idea of, well, maybe Mars was a more genial environment to the origin of life early on in history, in the solar system's history, has been around for several decades. Certain flavors of leading candidates for the origin of life, mainly those that require dry land, situate Mars as a very likely place to have an origin of life because although Mars may have had a hemispheric ocean

early on in its history, it probably wasn't completely covered in water, whereas geophysical and geochemical models of the early earth indicate that Earth was probably completely covered in water, maybe with the exception of a few very tall volcanic sea mounts. And so if you require land, and the reason why people like land is because of wet, dry cycling, another way of creating combinatorial diversity in organic chemistry is to constantly cycle between, you know, wet, and dry conditions, that might, you know, be a way to generate some of the earliest, you know, large-ish building blocks of life like RNA molecules and things like that. Now, there are other origin of life theories that do not rely on dry land.

And so it's quite possible that there are many ways to generate many forms of life from abiotic geochemical scenarios. Nonetheless, I think it's a really interesting idea to think about the trading of material from Earth and Mars. And you talked about Joe Rogan wanting, you know, smoking a piece of Mars. I would love to smoke a piece of Mars, just not in my mouth, but in my instrument in the lab called a pyrolysis G-CMS, Pyrolysis gas chromatograph mass spectrometer. I think Joe has one of those too.

He's got everything, I guess. But we've got this instrument, a copy of which is currently sitting in the belly of the Curiosity rover on Mars. And basically what it does is it smokes material. It heats planetary materials up to roughly 600 degrees Celsius that volatilizes the organic matter in those samples and they flow through a little tube pushed by helium into a thing called a gas chromatograph mass spectrometer which separates all those organic molecules out

so you can count them and identify exactly what is in your sample. And this is one of the ways in which we can get to learn the patterns of organic molecules in life and non-life. And if we had a way to essentially smoke material from Mars, we might be able to use that, again, pattern recognition machine learning software to determine if we found life. Where does the arrow of time fit in, you know, in terms of the time scale itself, if you will? Because we know in physics that the most sacrosanct law of physics is CPT, you know,

and variance, that things under charge, parity, and time reversal look the same. Certainly, you know, I have Newton's balls somewhere over here. If you have pendulum swinging back and forth, nobody can tell me if they, they just walk in the room, which side of the, you know, the displacement it started with, right? And similarly, if you looked out, you know, we don't know if you looked at the solar system upside down, you would see it going backwards according to time, but that means the laws are symmetric with respect to times, discrete symmetry. And yet, something must encode the, you know, what we perceive as the flow of time or the flow of functional information.

When I talked to Lee Cronin on the podcast many years ago, he said, you know, it's really only hold for molecules, you know, like that sort of concept. Is there a similar notion in functional information flow that you have to coarse grain over certain scales before it becomes valid? Yeah, I think this gets back to the question of scale that we were talking about before. And you're right, you know, most of the laws of nature

are, you know, symmetric with respect to time, except for, you know, the second law of thermodynamics, which emerges, I guess, you know, once you have a large collection of fundamental, particles. And if you have this past hypothesis where the universe started out a very low entropy and then is progressively moving towards states of higher and higher entropy toward what we call a heat death, right? And so the idea is quite similar where the universe or, you know, a system would start off with very low functional information. And you're right, again, you know, if you are

not course-graining, if you're just looking at the information content of the position and momentum of the fundamental particles, that won't change over time. But the functional information, once you go to the higher level and you assess this information based on whether or not it is persisting as a macroscopic structure or a macroscopic entity that is constantly ingesting new sources of free energy and raising the entropy of the universe around it, or generating novelty and ratcheting up toward new and more complex possibility spaces, that is when we would make a functional information assessment and try to essentially catch it into this era of time in action.

seeing how it increases empirically over time. I've seen some interviews you've done with Bob, including one with my friend, Hakeem Al-Ashei, who's a past and future guest on the show. And, you know, friendly in his way, tried to, you know, some gotchas, right? Like, what level does it have to be? I mean, this isn't this hubris, chutzpah, as we say,

you know, call it a new law. But I think, you know, kind of, you know, versus a hypothesis, proposal. But, you know, if you look at things like Newton's, law of universal gravitation. We call it a law. Is it correct? Yes. Is it wrong? Yes. It's both at the same time. And I think that's what, you know, ordinary people don't get maybe. So maybe for a layperson, what is a law, what is a guess, what is a hypothesis, what is a fact, what is evident. Take us through the philosophy of science, because you said you are working with philosophers as well. Yeah, yeah,

yeah. And so this is exactly right. So you look at Newton's laws of motion or of gravity, and we see how they are bounded in their applicability. And most laws are, right? So Newton's laws don't apply to the quantum realm or to, you know, the superman, the enormous, right? That's where you either need quantum mechanics or general relativity to take over. And so those three prerequisites, you know, combinatorial diversity, configuration, and selection for function didn't exist back then. Nor will they exist at the heat death, you know, in 10 to the 100 years.

When everything cools down, you no longer have the free energy to try new configurations, well, then the law of increasing functional information will cease to describe any pocket of the universe. But that helps us actually, I think, very beautifully and profoundly contextualize the moment of the universe's history that we exist, that we get to, you know, walk under the sun and have podcasts, right? The law of increasing function information. The highest form of evolution. Describes. What he called two guys with a microphone.

Exactly. It describes what we see around us right now. And so we live in, as Sean Carroll would put it, the interesting part of the universe's history. You know, the part where these thermodynamic drivers allow for new complexity to emerge. And what we are basically trying to say in our proposal is that it's not just the laws of thermodynamics

that are necessary and sufficient to describe everything that we see around us. There might be laws of information that are missing and that we still need to seek. And science progresses this way too, Brian, right? It's like, you know, combinatorial exploration, not just of matter and molecules followed by selection, but combinatorial exploration of ideas,

new proposals, hypotheses, theories, call it what you will, you know, a law. But undergoing this process of generating new possibilities for describing the universe and then allowing those possibilities to be selected, upon for their utility for actually explaining and predicting and quantifying what we see around us is the scientific method. And so whether or not this law of increasing functional information ends up being useful or ends up being one of the many ideas that have been scrapped and

fall into the wayside over time really doesn't matter to me. I think it's my responsibility as a scientist to generate new ideas and then maybe some of them will stick for some amount of time and be applicable to some region of possibility space. Or maybe they will just lead to something else, or maybe they'll just be proven incorrect. And that's fine, because that's the way science works. How would you apply this in your research? So you're looking at planetary dynamics, geology of other planets. Does it make a prediction that could be falsified?

Take us through how you apply this or how one would apply this, or me as a cosmology? Okay, yeah. Okay, a couple of thoughts. Well, we've already discussed how the search for life elsewhere in the universe may hinge upon in some important way, this search for evolution, for selection, right? And so if we come up with many creative ways of trying to identify selection for biochemical functions, whether that's through the patterns of molecules that we see in our PiG CMS instrument, or through the assembly number for complex organic molecules,

or any number of these very creative proposals that are coming about, I think that might be one inroad to identifying life as we do not know it. Another way, you know, and something that's been really interesting to see is the number of folks from very different kinds of disciplines latch onto these ideas and take them in their own directions. So you mentioned Lee Smollin earlier in the podcast, and so there's this idea that universes themselves sort of evolve and there's a selection pressure for universes that can then birth new universes, right? I think this was one of Lee's ideas. And so some cosmologists have been interested. in applying our ideas to that scale, to all the multiverse level. I am not a cosmologist.

There was a brief moment in, I think, sophomore year of my undergrad, when I took a great physics class from a cosmologist, and I said, I want to be a cosmologist, but then I took a few planetary science classes and decided that was more for me. But, you know, so I would love to partner with folks who are engaged in cosmology like yourself to see if that is a viable application of our theory. and another very intriguing application that has come out recently is in the field of oncology, of cancer studies, and describing tumoragenesis through the lens of selection for function.

So this has been shocking to me because I'm not a biologist, I'm not a medical doctor, but to see people engaged in health care describe their systems in a new way through this lens has been mind-blowing. So let me just quickly explain. So tumors, you know, cancers in our bodies are systems that seem to ratchet up in complexity and functionality in response to the biological selection pressures of our bodies become the environment. And we've got many different ways of fighting cancer, but eventually cancers will persevere, they'll persist, and they'll grow in ways that allow them to be virtually undefeatable. And this is very sad for us.

So cancers do this kind of development in a way that is not within the strict context. of Darwinian biological evolution. They're not self-replicating themselves and doing this sort of natural selection in the way that lineages of species do. Nonetheless, they grow in complexity, often to very detrimental ends. And so if you switch this view of evolution away from the Darwinian paradigm and just see it as selection for increasing function and persistence and novelty, then you can start to think of cancers in this evolutionary lens and then design new therapies for trying to combat cancers as you understand their evolution increasing in functionality. Some oncologists have proposed designing evolutionary traps for these cancers,

where you hit them with a certain kind of therapy that forces them into a vulnerable state to some other therapy, which then forces them back into the vulnerable state that they were in initially, catching them in this causal loop where they're now being selected for functions that repair that kind of ailment to themselves, but not no longer ratcheting up in, you know, becoming all spread about our bodies. And so as somebody who is not an oncologist and has never thought about cancer evolution before in my life, in the past few years, just seeing these people from vastly different fields take our ideas and apply them to their systems has just been mind-blowing.

Let's talk about the simplest, and you mentioned Lee Smollin, so we have to go back to that topic for just a bit. The black hole, in some ways, it's the simplest possible object you could imagine. It's only characterized by three different parameters. It's mass, it's spin, and it's charged. It's a commodity like, you know, like salt or something like that, right? It basically is one salt grain is like any other salt grain, pretty much. You've seen a black hole, you've seen them all, except they can have a certain amount

of spin, up to a maximum amount, occur spinning black hole. They do cool things when that happens. they can also have no spin, although that's pretty rare. And they have a certain mass, and they can have charge. The charge is very hard to see, but you can see some of their effects downstream. And all these effects have been served by instruments like the Event Horizon Telescope. We have interviews with Shep Dolman and of course LIGO. We have all three of the Nobel winning scientists involved with it on the podcast.

But I want to ask you, in basic understanding of black holes, according to folks like Stephen Hawking, Roger Penrose, latter is a guest many times, says things that, you know, black holes have this intriguing property beyond the event horizon. Information goes in, and then it comes out very, very slowly

via hawking radiation over immense amounts of time, Beckenstein Hawking relation. So given that black hole, its entropy can only increase and its information seems to be forever walled off by the firewall, what are the implications for your theory?

I mean, does it just not apply inside of black holes or near, and you don't even need to singularity. You could have a naked black hole. You could have an ordinary black hole. So what are the implications for black hole, you know, scientists with regard to functional information? I mean, can there, is there any signature of, or could you settle this debate once and for all?

And show those, you know, crafty physicists like Lee and Leonard Tuscan and all these others claim information could be destroyed, right? So how can this realm of functional information? How can this be used to help our understanding or maybe settle this once and for all? Black holes, do they destroy information or not? I am sorry to disappoint. I will not be able to settle any great debates about black holes here. I think it's important for me to just acknowledge that I don't know.

And I find that sometimes scientists don't do that enough. So I don't know enough about black holes to be able to settle any debates. But I will say that I think. I think you're right on track where, you know, our law of increasing functional information may be outside of its bounds when we're talking about black holes. It's fine. Right. You also have to apply in every regime. Exactly.

Even when I'm speeding officer. The law does not always have to apply. Yeah. Yeah. So black holes might be outside of the regime, mainly because we're talking about, well, first of all, I think the information that Suskind and others were talking about and debating about was very different from functional information. And so when we're talking about, you know, increase or decrease of functional information, we're thinking about these ordinary scale systems of, you know, not relativistic or quantum mechanical. And we're thinking about systems, again, in which there is combinatorial diversity,

configuration, generation, and then selection for function. But, you know, once you cross the event horizon of black hole, I feel like those things cease to be true. And so maybe functional information is not relevant, except, again, going back to this, you know, universe spawning idea of Smolens, where, you know, if a universe is primed or tuned to be able to generate black holes that then spawn new universes, there might be a selection pressure at the level of universes themselves that cause, you know, universes to evolve in a sense that they propagate further. And again, because I'm not a cosmologist, I'd love to ask you, actually, Brian, what do you think about that idea? Should we apply the idea of selection for function, where function is spawning new universes to the level of the multiverse? Or does that idea not make sense at all? So I think that brings up a lot of things. And one in my mind is this question of purpose and, you know, the teleology, which we kind of have been dancing around but haven't really gotten to.

And that typically comes up when you talk about fine tuning, who's tuning, you know, it's fine. fine tuned for what? Past guest Stephen C. Meyer, who's a, you know, the Christian apologist, effectively I've had on young, former young earth creationist, I've had on noble pro, you know, I'll have anybody who's got an interesting idea to consider, right?

But they claim, you know, look, anytime we see information, it has a source, that source can be traced back to a mind, right? That's the claim. If you saw in a tunnel, you know, on Mars, and you saw, you know, the, you know, G.C. AT, you would say, well, it probably came from, you know, it could come from the winds, right?

The catabatic, you know, Martian winds, right? Or it could come from, you know, some intelligence that was there, you know, leaving the four most important words that can ever be found. Please like and subscribe, okay? You know what the deal is, right? You've got to do that, right?

I even say that around the dinner table to my kids. But my, you know, I mean, the main thing is like, what is the, you know, who is the, you know, informant, who is the information, you know, transmit, a recipient. The concepts of black hole, you know, kind of facundity in the universe is more, is a little bit controversial, I think, in the cosmological community. First of all, presupposed the existence of the Montpelerverse, right, these things.

In other realms, it doesn't explain the low entropy starting state of the universe that we seem to observe. Black holes have enormous entropy, right? So if they persist through the big crunch or through whatever the Aeon theory of Sir Roger Penrose is, they have to explain what happens to the entropy. And I don't fundamentally understand what happens to entropy there. There are other models that suggest that there is no multiverse and there is actually no inflationary cosmological scenario, which is the main thrust of my research, looking for B mode polarization on the CMB. We'll do a quick lab tour later on. But the challenge is, you know, how do those cyclical models, what do they do with the entropy?

And so Paul Steinhart, Anaegis, they've come up with theories that invoke this conformal cyclic behavior, get rid of the conformal cyclic problems of Sir Roger Penrose by having, you know, the dilution of entropy to very, very high degree, such that the next bounce can occur without a singularity. So basically you don't have an event horizon, you know, the singularity in the next universe. these are cutting edge things and hopefully my research was such some light onto it but yeah I mean the challenges to the Smolin theory I think are substantial

I mean that if you have to have a multiverse and you have to start with low entropy those two things are very hard to reconcile it doesn't mean that they're wrong but ultimately you guys are not shy about talking about purpose you talk at the second and last chapter

what does it mean for you so you know without you know we have like I said foremost atheist in the world from Sam Harris to Dan C. Dennett, to Dan Lill C. Dennett, and many others have been on the podcast to, you know, people like I already, John C. Lennox and Stephen Meyer. But tell me, when we do consider the ultimate meaning questions and so forth, can one look at this in a way as a religious person and come away with it bolstered in his or her religion? I often say, if you want to know the mind of God, you're supposing that God exists, why not learn science? It's like the assembly language. of the universe, right? So what is your take on it? What should a religious person? Should it be scared? Look at this notion of this evolving, you know, kind of dispassionate. If there is a creator, he or she, it's not moved, right? So how do you react to that? Yeah, that's great. We've had,

you know, a variety of reactions to our proposal, to the papers that we've published, to this book. Some people write us saying, thank you for proving the existence of God and others write to us, thank you for disproving the existence of God. So it's interesting that, you know, you can read into it in very different ways, and everybody brings their past experiences and preconceptions to the book. And I feel like, well, I'm with you, Brian. I feel like, you know, if you do believe in a creator, then knowing science, knowing the laws of nature, will be that window into the mind of God.

If you don't believe in a creator, then you better also understand the laws of nature because that's all you've got. And we write that in the book, and I think that was a very important passage for me to write and for us to have in this book. Because I see so many science communicators out there, some very prominent ones, immediately dismissing the God hypothesis and immediately putting down theological thinking. And I feel like that really walls off a huge segment of humanity. Most individuals on this planet would identify as religious, right? And so if we make that kind of distinction between science and religion, then we are basically saying that science is for a very small subset of humanity. And if you happen to be religious, then it's not for you, but it's not true at all.

Science is for everyone. I don't happen to be religious, but I think it's important that we don't create a false dichotomy between science and religion because the laws of nature matter to you, no matter who you are. and they're important to understand no matter what your belief is in a creator or divinity or spiritual nature in this universe. Yeah, I would say the amount of, well, there's actually quite a good deal of cosmology and planetary science in the first chapter of Genesis. But beyond that, there's not much science. It doesn't really purport to be a science, but I mean, the moon is formed on the fourth day,

the sun is formed on the fourth day. It's obviously they knew it was, you know, the day was determined by the sun. So that must have other purposes. But I guess, you know, in conclusion as we, you know, kind of bring the spaceship into land, as they say, when, you know, we look at the future and the evolution of how we're evolving. And this very interesting period, this guy on Twitter whose cousin, I think I follow, Andrew Chen, his brother or his cousin, Siki Chen, you know, put on something on X Twitter today and said something, you know, like the proof that we live in the singularity is that we're at this transition point. we're worrying about the singularity. Basically, we're coming towards this end. And a lot of that's driven by, you know, fears and hopes and optimism about AI.

We've had on people like Adam Becker is very, you know, very much opposed to the kind of Silicon Valley billionaire overlords that seem to be dictating everything that we are thinking about when we think about things that think, right, the metacognition. What is, I mean, could somebody have predicted the existence of AI? I mean, you take some rocks, you know, that you understand. you'll put some electricity through them, you know, and then all of a sudden you've got access to like superhuman, you know, superhuman intelligence across every domain of knowledge, your PhD, my PhD, is it worthless? Would have been predictable via some element of the framework that you and Bob present? In other words, is this inevitable the artificial intelligence explosion? And, you know, what comes next is sort of the most terrifying question, right?

Einstein said, I don't know what weapons will be used to fight World War III, but World War IV will be fought with sticks and stones. So, more minerals. So tell me, it would it be inevitable that some rocks on a planet, heavy, rocky planet, not far from the sun, the habitable zone, that we would have this artificial intelligence. It seems to be just so literally magical. This is very similar to the question about the Fermi paradox, right? because just like you might ask about the prevalence of the origin of life elsewhere and its evolution and trajectory toward civilization and then space-faring nature, you could ask about the possibilities and probabilities for the second genesis of a kind of life here on Earth,

not one made of organic matter, but one made in silica, right? So some new form of life that we are literally generating, an alien presence on our planet that co-evolves with us and was in some way spawned by. us. And I think that, again, my answer, in relation to the law of increasing functional information would be we're agnostic to the probabilities of these major transitions, of which we are in one right now, a major transition in our information processing capabilities and scale to a planetary level of intelligence. And so where it goes and where it leads, I'm not sure. And whether it's inevitable, I don't know. But I do think that there is a tendency toward greater information processing power.

So if there were origins of life elsewhere in a universe, given enough time, then sure, why not? There could definitely be artificial intelligence there as well. One thing that we definitely want to emphasize in our paradigm of selection for function is what it means that now that we have AI systems. So machine learning systems have been involved in science for many years and decades, and we use them in astrobiology all the time to make sense of the enormous amount of information and data that we are generating about our cosmos as scientists, as humanity. But another end of this, now this bloom of generative AI, well, what does that mean for us?

Well, again, through the lens of selection for function, we have a way of rapidly increasing combinatorial diversity of whatever outputs these generative systems are creating, new strings of text, new blog posts, new pictures and videos and all that stuff. And if we take the framework that we've presented seriously, then the thing that will increase the functional information of these systems will be ever more stringent selection pressures. If we as human beings, as users of AI systems, as regulators of AI systems, as governments, as entities, as corporations that create and craft the uses for these AI systems, if we do not increase the selection for functional outputs, you know, text that actually makes sense, that isn't full of fallacies or... Sycophanty. Yeah, exactly.

Then we're in for a lot of trouble because the combinatorial exploration, this diversity of possibilities has been greatly expanded, which all by itself would decrease the functional information of a system if we do not select for increasing function, for things that work, for things that benefit our lives and our persistence on this planet. Dr. Michael Long, thank you so much for coming down to UC San Diego all the way from the East coast making your trip. Come back to your brother's alma mater. You know, shout out to the Tritons.

Go Tritons. I want to present you with one final gift. You may have been wondering what this thing is just sitting here. Well, in the movie 2001 in the Space Odyssey, there are these monoliths that are kind of maybe they're time capsules, maybe they're talisman, maybe there are some sort of artificial intelligent entity like how, like our little eyeball there. But this is actually just some piece of polymer that happened to self-assemble and, and Put your name on the rim of it.

So on the front is Arthur C. Clark, who came up with the title for the podcast. The back is the monolith. And on the edge is a little thank you to you for coming here. It's not the Nobel Prize. We call it the Keating Prize for Impossible Imagination. I want to thank you. Thank your co-author, Bob, for me. And hopefully he'll come on the show sometime too.

Thanks so much, Brian. This was a real treat.