The Origins Podcast with Lawrence Krauss - What's New in Science With Sabine and Lawrence | Fusion Dark Matter, String Theory in Biology, and Rapid Evolution

Episode Date: February 13, 2026

I’m back with my friend and colleague Sabine Hossenfelder for another episode of “What’s New in Science”. I think this is one of my favorite dialogues that we have had. Spending time with Sabine was a nice chance to step away from my physics lecture series for a bit. I know many of you have been enjoying the lectures, so don’t worry, they’ll be back soon. In this episode, we covered the kind of science news I like best: ideas you can argue about and results that make you recalibrate. Sabine opened with describing a clever proposal that future fusion reactors might double as axion dark matter factories, producing a flux of very light, weakly interacting particles through neutron-lithium reactions in the shielding. That led to a discussion about what people mean by “axions,” why particle physicists tend to be more particular about the term, and why I’m always more interested in dark matter candidates that were invented to solve an actual problem, not just to fill a cosmological gap. From there we jumped to quantum mechanics at the edge of common sense, with a Vienna experiment showing interference from a cluster of thousands of atoms, and a friendly disagreement about whether “collapse” is a real physical process or just the wrong way to talk about what quantum mechanics is doing.We also talked about AI and math, including the recent swirl of claims about machines proving famous open problems, what was hype, what was rediscovery, and what might genuinely be changing in how mathematicians search the landscape. Then we went from equations to extinction, with a fascinating new approach using space dust and helium isotopes to argue that life may have started rebounding after the Chicxulub impact far faster than people had assumed. Sabine brought a surprising example of string theory mathematics finding a practical use in modeling biological networks, and we ended with biology proper in two very different moods: a sobering study in mice suggesting lung tumors can hijack vagus nerve signaling to suppress local immune responses, and then a lighter result about dogs learning words from overheard human conversation at roughly toddler level. My dog Levi, who many of you have seen on the podcast, was asleep next to me while we talked about it, which felt like the right way to end.As always, thank you for your continued support, and I hope the changing of seasons brings you good time with friends and family.As always, an ad-free video version of this podcast is also available to paid Critical Mass subscribers. Your subscriptions support the non-profit Origins Project Foundation, which produces the podcast. The audio version is available free on the Critical Mass site and on all podcast sites, and the video version will also be available on the Origins Project YouTube. Get full access to Critical Mass at lawrencekrauss.substack.com/subscribe

and welcome to the origins podcast. I'm your host Lawrence Krause, and this is one of my favorite segments where I get to talk to my friend and colleagues Sabina Hussenfelder about science and what's new and what's real and what isn't and get her take as well as mine. And it's always a pleasure. Hi, Sabina, how are you doing? I'm good. How are you? Not bad at all. It's very foggy here, but no snow, so that's good. Do you get snow where you are? Yeah, sometimes we did have a few centimeters, but nothing serious. Nothing serious. Okay, well, in any case, it's almost spring. It's February. Anyway, let's begin with you and something to do with fusion reactors, right?

Yes, it's even better. It's not just fusion reactors. It's also about dark matter. Oh, okay. That's a favorite subject for both of us. So it went through the news around the turn of the year, and I didn't really have time to look at it then, but I've now had a look at the paper. So it's about the possibility that nuclear fusion reactors would actually produce a lot of dark matter particles. And that sounds kind of crazy at first. But if you know how we've measured many of the properties of neutrinos, then it makes a lot of sense because nuclear fission reactors produce a lot of neutrinos. And it's not just that it's a lot of them.

It's a lot of neutrinos from fairly close by, as opposed to coming from the sun. And also, we know pretty much exactly how many of them must be produced if our theories are, because we know the amount of power that is produced in the machine, right? So we have a pretty good estimate for the flux of neutrinos that comes out of the fission generators. And we also know the distance quite precisely. And from all of this, you can derive equations that will tell you what you should expect to see in your detector. And so for the neutrinos, this has been used to measure neutrino oscillations. But it's kind of a similar idea for using nuclear fusion reactors to produce dark matter particles.

And this is about a particular dark matter particle, which is called the axon. So that's a very light particle. So, I mean, it has a very small mass. And that's why it's really difficult to find this kind of particle with particle colliders. So like the Large Hadron Collider. So if we make collisions of protons at very high energy, these particles could also be generated. But because they're so light and they interact so rarely, they don't go into the detector and the only signal you get is a teeny tiny little bit of missing energy

which you can't read out of the data. So what you'd want to do with the nuclear fusion device is just to produce a huge amount of them and then put your detector right next to it. And the higher the flux of the axioms, the bigger your chances of detecting it. So basically that's the idea. and the paper puts numbers to this.

What could you expect? And they've also looked at quite precisely at the production mechanism for these axions, which mostly comes from the, so they look at the most common deuterium fusion, which generates a neutron. The neutron has very high energy, and it's actually somewhat of a headache

because neutrons are nasty to deal with. So this is why you have to cheer these reactors. And what they plan for a lot of those reactors is that they shield them with something that contains lithium in one or the other form, either in some kind of molten salt or actually a kind of metal or whatever. Because if the neutron hits the lithium, it will create tritium, which is one of the fuel. which is one of the fuel components and one that's quite rare on Earth. So this is the chitium breeding. And so the authors of the paper say that a side effect of that very reaction, the neutron hitting the lithium nucleus,

is that these axons will be produced. And so, you know, it's kind of a neat idea. Of course, it's very speculative in the sense that we don't actually have a huge nuclear fusion reactor that we could use for this. And also, like, we don't know that this particle actually exists. And even if we measure it, like, in the best case, you know, they put this detector next to the nuclear fusion reactor, and it measures the axioms.

We still wouldn't know if this is what dark mitres made of. But, you know, I think if we ever have a nuclear fusion reactor, why not give it a try? You know, it's, it makes a lot of sense to me. So I quite like the idea, actually. Well, look, I was intrigued when I saw it. It takes me back. In fact, the process that they talk about is a company called Bremström, which may be a German word, probably.

And it's a kind of reaction where more or less, when a neutron interacts with another nucleus, it sort of throws off things. Normally, this is known from physical processes when light is emitted by particles when they collide with other nuclei and they sort of emit a light in the process. But other light particles can be emitted and it's called Bremselang. It's actually a process that produces axions and this takes me back because actually I think the first real calculations of this process as it happens in the sun was done by me and my colleague Frank Wilcheck 40. years ago. And it's a process, and we used it to drive limits on axions because they would take energy out of the sun. And also when there was a while, again, in Germany, it turned out a false reading about some weird process, probably before your time. And it looked like it may be doing a new type of axiom that we had proposed. And again, you could look at Bremsler-long reactions to try

and rule it out, which we did. So it's a neat idea. I have to say I'm dubious. And I noticed in the article that you quoted, it said axions and axiom-like particles, because there are pretty strong limits that we derived on axioms, you know, so they don't take out all the energy of the sun. And my suspicion, and you can correct me if I'm wrong, because I didn't look at the article, is that when it comes to actual axions, the kind of things that are predicted in the models that solve various particle physics problems, the rates are probably too small, even if the thing existed, probably too small,

to detect. But there are lots of other axiom-like particles that have proposed that have different properties. And my bet is that those are the ones they can constrain than not the real axioms. Is that the case? Is that the parameter space on the two axioms is very constrained already. But people who work in cosmology have been using the word axon very, very loosely. Yeah. So I don't blame them for it. Yeah, no, I don't either. As a particle physicist, I, I, I, get frustrated because it's easy to invent particles that have different properties. But what makes Axion's interesting beside their cute name is that they actually saw,

the original ones and the ones that have that name solve, as you know, a real problem in particle physics. And I'm always, I always think of it as, I call it kosher of dark matter particles, the ones that have been invented to solve a problem in particle physics, and that just happen to be dark matter, I find much more interesting and likely than ones that have been proposed just because you could invent a particle to be dark matter. So I'm more skeptical. In any case, I think it's important. It's a really neat idea, and I agree with you, why not?

And in fact, you're right. After all, the neutrino was discovered not by looking at cosmic neutrinos or neutrinos from the sun. It was discovered by Fred Rhinis and his collaborator Cowan by putting a detector right next to a reactor, because there are lots of neutrinos. And so why not look for this? But I suspect, as I say, the particles that are likely are not the ones that interest me. But, of course, that doesn't mean nature didn't choose them.

Didn't they first try with a nuclear bomb test? I think people talked about it, but I don't think it was ever really done. By the way, just, well, that's interesting because people had talked about using neutrinos as a way to detect nuclear weapons explosions. And actually, again, I worked on this a long, long time ago. It's not too exciting. When we worked it out, in order to detect a nuclear explosion looking at the neutrinos,

you have to be a few kilometers away. And there are probably other ways to detect nuclear explosions if you're a few kilometers away. In fact, you probably don't want to be a few kilometers away. So in any case, it's a few kilometers away. it works, but it's one of these destructive experiments. Let me call it that way. In any case, it's nice to see that nuclear fusion reactors would have a side product, benefit, but I suspect if they actually work, it would be remarkable for other reasons.

But as you know, and we've talked about it in the past, and I've been a skeptic, although I know there's a lot of new work going on, but they've always been 25 years in the future. And it'd be great if they're not. But so far, that seems to be a constant 25 years in the future. But we'll see. Well, I could now talk about nuclear fusion for like half an hour, but I would say let you tell us something about Schrodinger's cat. Yeah, exactly.

Schrodinger's cat, another, our German links are just overwhelming here. And this is Schrodinger's cat, who's a German cat. Anyway, the famous, this is a quantum mechanics thing. Actually, I think maybe because I told you I wanted to talk about it, you didn't introduce it. I was surprised you didn't because I know anything quantum always intrigues you and the public. And this has to do with the fact that the world is quantum mechanical. And that still seems to surprise some people. And so experimentalists continue to try and do experiments to demonstrate that the world is indeed quantum mechanical.

because quantum mechanics is so crazy. And Schrodinger, who was one of the developers of quantum mechanics, gave the famous example of how crazy it was when it comes to the real world, because in a fundamental level, quantum mechanics says that quantum objects that are behaving quantum mechanically could be in a superposition, namely they can be in several different states at the same time, either in different places at the same time or their spins can be pointing up and down at the same time. And the actual particle can be a common.

of many different states. And when we measure it, we measured one state. It seems highly, well, it seems ridiculous if you have classical reasoning to think the particle could be doing many things at the same time. And classically, it shouldn't be doing that, but quantum mechanically it is. And of course, over and over and over again, in spite of the fact that it defies common sense, experiments have shown that that's the way the world behaves at very small scales. Particles are in superpositions and you can, and there's the famous double slit experiment that shows that things like electrons, single electrons can interfere with themselves. They behave like waves and they can be in superpositions of different places at the

same time. And all the experiments show that. But of course, that doesn't happen with you and me. We're not in separate places at the same time. Be nice if I could be there with you at the same time as I'm here with me. So we could talk. That way. but it doesn't happen. And Schrodinger pointed out the craziness of this by this famous hypothetical experiment where you have a cat in a box and there's a radioactive element somewhere.

And of course, radioactive decays are governed by quantum mechanics. And if the particle decays, then a bullet, a gun shoots the cat. And the point is because the radioactive decay is governed by quantum mechanics and superpositions, the cat is, and therefore the cat is a superposition of a live cat and a dead cat until you open the box, you don't know whether it's live or dead. And it seems crazy. And so experimenters have tried to come as

close as they can to macroscopy objects to see if these crazy results from quantum mechanics work. And this recent experiment in Vienna involved 7,000 sodium metal atoms, and that's an object that's eight nanometers wide, which doesn't seem very big, but it's as big as viruses. So, it's almost classical. It's something you could see with a microscope. And what it showed was that even if you could have a superposition of clusters of these 7,000 sodium metal atoms that were separated by 133 nanometers, which again, on a human scale is small, but an atomic skill is large. It's again, macroscopic. They're separated by the amount you could see with a, with a microscope, that those two clusters can interfere with each other. You basically can have a cluster

that is in a superposition of being in two different places at the same time. And they did it by a modern version of the double-slit experiment. This was involved three different gratings, but basically using lasers, they were able to collimate a beam of these clusters into small, into separate little groups like you would with the grading. And then they showed that individual cluster particles that go through this grading can basically spread out like waves and interfere with themselves and be observed on a third grading. It's a standard way of showing that particles, that quantum mechanics works, that an individual

quantum state can be viewed to be in many different places at the same time. and interfere with itself. And that's what happened. And the reason this gets difficult is that in order to see interference, you have to see wave-like effects, but the more massive an object is, the smaller its wavelength, according to the standard arguments of quantum mechanics. And for very small wavelengths, it's very hard to see interference.

Particles behave like particles. And so any noise or anything else can get in the way of disturbing this beautiful, interference pattern that you see as waves. They managed to do it, and it's nice that it still works. The reason that it's claimed in the article I saw it to be interesting, to me it's neat, but I'm not totally, totally expected. But some people who still don't like the idea that quantum mechanics governs everything think that there's some place where suddenly quantum mechanics breaks down and the classical world arises instead of some continuum where classical effects

become more significant and quantum effects become less observable. And I think these people are called collapse theorists or something like that, and I wish they'd collapse myself. But this shows that at least at this scale, quantum mechanics works. And it's interesting from a technological perspective, because the larger the scale you can make quantum mechanics work, the more interesting and strange things you can do. And their next goal, of course, since these are the size of viruses,

is to do something like do the same thing with a biological virus. It wouldn't, from a physics perspective, it would be any different. But it would seem a lot more interesting, and I can guarantee you it would get headlines if you showed that living systems behave quantum mechanically. To the extent that viruses are living, and again, there's a debate about that, because they don't really satisfy all the properties of life. But anyway, once again, quantum mechanics works, and I'll throw it to you to see if you think it's more interesting than I do.

Well, you'll be shocked to hear I'm a collapse physicist. I do think that quantum mechanics, I mean, look at the world, right? As you said, like, we don't observe things in superpositioned. So at some point, they'll have to go away. And decoherence doesn't do it. Decoherence just creates a lot of entanglement. And the thing that's supposed to lead to the emergence of the classical world is not a physical process. Like, if you're tracing out the environment, like this is a math thing.

It doesn't really happen. No one believes this. So I think that's decoherence explanation is logically incoherent. So, but, you know, to say the truth, I hate the most popular collapse theories because they just make things worse. So I'm like, you know, I'd rather stick. with quantum mechanics the way it is. So I have my own collapse theory of quantum mechanics, which is the best of all theories, of course. But leaving aside my personal biases, I think that's a very reasonable thing to do.

Like you want to push the edge, you want to push to more massive, to bigger things just to see if there's something new to find there. And I think this is a way to push the limits of what we know about the fundamental laws of nature in a different way than going to higher energies or looking at further distances. So we go to quantum effects, more mass of things.

So this is why I think this is a very neat experiment. I want to add one thing, though, just so that people don't get confused because you were talking about the interference that they've measured and so on and so forth. And that's a very direct way to prove that an object must have had, quantum effects because it could interfere with itself.

So it must have been in two places at the same time. But there are other ways that you can prove that an object must have had quantum properties. And so there are people who try to work on other ways to prove the quantumness of heavy objects are usually by some sort of entanglement that might be more feasible to do with bigger things. Because as you say, you know, if you're trying to look at something that, depends on the wavelength of these things. If they get too heavy, it just becomes impossible. Okay, well, I agree with you, complete.

If you're an experimentalist, you should always push the edge because you never know what you're going to see. And I think most experimentalists aren't governed by a theorists. They do what they can do. And I encourage that, as long as it doesn't cost too much money, because you never know where some surprises will come in. I will say that part of the problem here is the interpretation of quantum mechanics and the fact that the world is classical. And whenever I hear concerns about interpretations, I go back to a beautiful lecture by Sidney Coleman called quantum mechanics in your face.

And where he argues, and I think so. I think it's true that we, this interpretation of quantum mechanics, the world is quantum mechanical and we should ask how, we should talk about the interpretation of classical mechanics, namely not trying to explain the quantum world in terms of classical world, but explain the classical world in terms of quantum world. And his argument usually has been, and I don't, I can't say in this case, is that once we accept that we are quantum objects, then it's not too surprising that we see what we see classically, that when we measure something, that we're measuring, you know, a specific thing.

And often what's codified as collapse of the wave function is nothing of the sort. It's just one quantum object measuring another. And I tend to think my own bet is that ultimately that's the right way to think about it. And collapse is that trying to force things into a classical picture is not. the right thing. And years from now, when we teach babies quantum mechanics before they learn classical mechanics, maybe they won't, they won't feel so weird about it. But we'll see. I, I, I'm agnostic about it. But I happen to, as Sidney Kubrick used to say in, in, in, in, in that favorite movie, I've stopped worrying and I love quantum mechanics.

well yeah so for me the issue though is not classicality like i'm fine with quantum mechanics the the issue i have with with the collapse is the non-locality which is the same thing that Einstein worried about right because if you if you take it as a physical process it goes faster than the speed of light so so and and of course you know there is no actual information transfer faster than the speed of light. But it kind of, you know, makes me feel very uneasy. And I think it's like the collapse is I think the major reason why we haven't managed to combine quantum physics with gravity, which is something that I've worked on for a long time.

So this is why I'm interested in. Yeah, sure. I mean, there's no doubt that quantum mechanics makes us uneasy. That's one of the reasons I love it. Okay. Now, speaking of uneasy, people are very uneasy about AI lately. And some people are concerned that it's going to take away their jobs, and some people are excited by what can do.

And I think you're going to talk about AI, both hype and real, in terms of math. So go ahead. Yeah. So interesting things are happening, and they're really beginning to happening. I don't know if you saw this, but I think like yesterday, nature news, published an opinion piece that basically said AGI is here. You know, I think it's a little early call. Yeah.

But it's like we're getting there. And I can tell you, like, from my own experience, I've been using any available frontier model for maths and physics things for years. And initially it was just ridiculously bad. but it's like in the past six months or something especially chat GPT Pro has become really, really good at math and physics. And so there are now people who are using GPT Pro

to try and prove mathematical theorems, which is a daring thing to do because this machine still doesn't actually understand logical reasoning. Yeah. And so what happened in October or something is what's been that like Erdush Gate is that there was some people from Open AI who claimed that Chad GPT had proved 10 or 11 or so of the open Erdush problems.

So now some words about these Adersh problems. Paul Erdisch was a Hungarian mathematician in the past century, and he just collected his own open problems that he was interested in. So it's not an official list. There's just a website where a mathematician, I think is Thomas Bloom, if I remember correctly, has collected these open problems, and there are about 1,100 or such, and it's like 60% of them are open. And so it's kind of a nice set of problems that they can try their AI on.

I totally understand where they're coming from. But, you know, in all honesty, it's like these problems are not super relevant. Like it's not like the millennium problems or something. And many of them are probably unsolved just because no one really cares. So no one's really tried to solve them. Another thing that happened was that it turned out pretty quickly that actually those 11 supposed proves were just rediscoverys of proofs that had been published but that were not noted on the website. And so this was a little bit of an embarrassment, I think, for the open AI people, because you think that they would have checked this, right? I mean, it's like all they had to do was to ask chat GPT.

Like, is this a known result? And it probably have said, yes, I looked it up and the references here. But yeah. And so some of them deleted their post and so. So this was in October. And like then in the next month, there have been every once in a while. Someone published a paper. And then it turned out that was using chat GPT.

And then it turned out, okay, it was actually known or it was actually wrong. I've seen at least one paper that was actually withdrawn from the archive that was suspected to be written with Chet GPT, which may or may not have been the case. You know, it's all a little vague at this point. But then I think last month or a couple of weeks ago, Chad GPT Pro seems to actually have proved one of those open problems. And that I didn't actually look at the proof myself as totally out of my area. but I saw like an explanation of it that came from Terence Tao, like the Fields Medal winner, and I assume he knows what he's talking about.

And so it's quite interesting that, so many mathematicians embrace this, but I also think like everyone's a little bit insecure where it will go. So at the time, I think AI works as a sort of discovery tool because it can very quickly scan the literature and find things that might help you make the next step on a problem. And that is worth something and it can actually move things forward. And so this is happening now in mathematics. So mathematicians, I think, are.

you know, slowly warming up to it. And I think that in the next months, we're going to see much more of this. And it won't be long until this also happens in physics. You can basically see it coming. Yeah, well, in fact, coincident at the same time, and I thought you were going to mention it. I don't know if you saw that.

I saw it as a tweet, but, you know, about the Institute of Events study. Did you see this guy who was saying that there was a closed-door meeting at the Institute of Advanced Study, and the physicists were saying 90% of what they do can be done by AI right now, and what are they going to do? And it was throwing up their hands, and I thought it was a good timing to talk about this. So, you know, I can't, I think when I, when I actually listened to it, I thought it was a little bit of hype. A lot of it was talking about 90% of what they do in terms of their daily life, in terms of their bank accounts and their and their emails, they can get done by AI.

It was really quite interesting. When I heard the guy talk, it was basically saying one or two of these, and there are people who do it who have AI who handle their banking and their email and all the rest of the stuff. And that may be 90% of what they do, but I don't think it's what qualifies

is the best part of their science. But it is true that AI is having a bigger and bigger impact. And while there's hype, and you gave the example of these people claiming things that hadn't been done, I suspect it'll be very useful. The key thing will be to recognize what it does, what it can do, and what it can't do,

and to not make claims about it that it can't do. It's very good, as you point out, at searching things and finding things that you can't find that may help push your work forward. It's also good at coding and doing, and so there's a lot of things that can help with. It's a tool. It's not a thinking machine yet.

And I'm, and I think it'd It's going to be a long way off before we have thinking machines. But it, you know, so I thought it was nice that this was a double-edged story. First, the hype and then the reality. And it's nice to know that there's sometimes some reality beneath the hype. And AI is improving. I'm a dinosaur. I don't really use AI at all.

I don't trust it for many things. And so maybe that's just because it's, maybe it's, you can't teach an old dog new tricks. But, but I'm very, although. I'm tempted to try every now and then, but every time I've used AI, I haven't, I've come up with, not for math and physics, but for information. I found out that the information's wrong. So, so I'm very skeptical. But I do want to elaborate one thing about Airdosh, because it's at least amusing. The Airdosh problems are interesting because Airdosh himself was an amazingly interesting guy. He was a math issue, had no home, and he basically moved in and lived with all his

collaborators and had 6,000 collaborators over time and was fascinating. And so, fascinating that people have created something called the Airdosh number, which is, you know, he collaborated with so many people. How many people have you collaborated with you collaborated with people that have collaborated with people that collaborated with Airdosh? And I'm very happy that my Airdash number is only three. So I have a low Airdosh number. But he was a fascinating guy. And he, and that's why, I think that's why the Airdosh problems are interesting, mostly because Mr. Airdash was interesting, maybe more interesting than the problems themselves. In any case, Now we go from this to extinctions, one of my favorite subjects.

Because extinctions on Earth, and we're, of course, we humans are helping extinctions all the time. We're in an era of incredible extinction for many life forms because of what we're doing. But there have been numerous mass extinctions in the history of Earth. And to find out both why they happen is important because some are due to space events. Other ones are likely due to things that happen here on Earth that can happen. And again, like massive super volcanoes, that actually mimic often effects that are happening by human-induced things like global warming. So it's interesting to look at it.

But what's really neat is the fact that in the history of life on Earth, extinctions for life haven't been that bad, namely by getting rid of lots of species, you open evolutionary initiatives for new ones. And one of the ones that was particularly important for us was the death of the dinosaurs. 66 million years ago when a 14-kilometer-wide asteroid hit near Mexico in a place called Chick-Selub, which is now in the ocean off the coast of Mexico, it killed off 75% of species, including ultimately the dinosaurs, likely making room for these little mammals to thrive, who later on would evolve into other mammals who would later on evolve into us.

So that extinction was probably good for us. We might not even be here if it hadn't. if that hadn't happened. But what's interesting, the question is how quickly does life arise after such an extinction? And theoretically, people had argued it would take evolution hundreds of thousands of years to basically have life fully, you know, new species evolve. And what's been discovered empirically is that actually evolution is much more effective and can, and life forms can evolve much more quickly. So the way to see this is you look at the layer of, you can see the layer that happened when the asteroid hit. There's lots of ways to look and

see exactly when that happened. And then you can try and look at the evolutionary life forms. And there's this kind of life form called a planktic forum, which is a grain of sand size object near the surface of the oceans. It lives within a calcium carbonate shell. and it's very sensitive to the conditions of the time. It absorbs nutrients, it's response of its receptive acidification and temperature changes. So seeing how this thrives tells you how quickly life is thriving. And 90% of these planktic forms sort of went away when the dinosaur killing asteroid hit. But the question is, when do you see them start to resurge?

and you can see them start to research. And the question is, how long when you look at layers of Earth, how long did it take between the layer that shows the dinosaur killing and the layer where you begin to see these forms emerge? And in the past, what had been done was to try and estimate how long it took by asking how quickly does sediment get laid down. And an estimate was 30,000 years after the dinosaurs you see these things form, which itself was already a surprise.

But the problem is no one knows what rate at which those layers were laid down. After all, this is right after an incredibly catastrophic event on Earth, and things were very different than now where you can sort of measure how quickly sediment arises when things are kind of stable. And so there was a big loophole in this estimate. And what new researchers have done, I found very interesting, is use what's called space dust. There's dust raining down on us from space all the time.

And that rate at which it's happening has been constant. It's not independent of what's been going on on Earth. And by looking at isotopes of helium, which are carried by space dust, you can measure, you can see its existence in these layers. And by looking at therefore the space dust and the rate of which you measure these helium isotopes, you can ask how much space dust has accumulated between the dinosaur killings, event and the point where the forum are resurgent. And the new estimate comes up as only 2,000 years. 2,000 years after the dinosaurs, you already begin to see new species of Placic Forum. And that's

kind of amazing. It's like saying there was an extinction event in Rome and already, you know, you have new source of life. It means that if it's really true, evolution is incredibly effective at regenerating new species in the absence of. when there's an evolutionary gap. So I think it's a really interesting result. I think it seems solid. The rate at which space dust has been coming down is kind of a cosmic rate.

It doesn't depend on what's going on Earth. And it completely astonishes people, and I don't think anyone quite understands how evolution could be so effective. But it's an interesting result, and maybe heartwarming, because it maybe means after we extinguish ourselves, at least there'll be new species that'll come by relatively quickly.

But it's an interesting result. And I like anything, of course, it flies in the face of previous expectations. So that's my take. Any thoughts on that? Well, it's very interesting. I don't really have a lot to say about it. I just want to pick up on something you said about us causing extinction and so on, because I looked at this very closely, recently,

and it turns out that this has been super overstated. Oh, probably. So as you said, like when the asteroid hit, there were like 75% of all species were gone. Yeah. And I think technically, like, it's a mass extinction, if it's more than 50%, or I might just remember this, but something like that.

But if you look at the, even generously, going by the estimates, the extinction that we have caused is below a percent. So I'm not saying that this is a, you know, a goal to reach. But it's, you know, we're very far from causing a mass extinction. I think it's the rate that matters. We haven't been around for very long. And given how long we've been around, we're very effective at extinguishing these. So the rate is high. The absolute number is not great, but the rate, the derivative is high.

Yeah, let's hope it doesn't stay that way. But, yeah, no, no, it's quite amazing. Like, I generally find it interesting, this intersection between biology and physics or astrophysics in how you can figure out what happened in the past. Like something like carbon dating is like an obvious example of this. space dust and there are other examples about particular isotopes that get stuck in rocks and whatever. Yeah, you know, it's, I agree.

I love when you can use these new techniques. And the last thing we should say about the, since you brought up mass extinction, just remind people that while the dinosaur one is the, is the one that gets all the attention, it certainly was nowhere near the biggest mass extinction on Earth. I think it's the Permian extinction, which were not just 75%, but basically 90%, of life forms died, and it was probably due to a supermassive volcano carbon oxide releasing event in what is now Siberia, and much bigger, much bigger extinction. The dinosaur one gets all the attention because we all love dinosaurs, as you pointed out behind me. You can see a dinosaur

that's a picture of it. I think it's in Basel, so it's near you. Anyway, but that's, anyway, Okay. Well, speaking of uses of physics and biology, it's probably a good segue to a next topic, which you could say is the use of string theory in biology, which when you first mentioned it in your email to me talking about what you want to talk about, I thought, what? And then I started to read it. So why don't you tell me what it's all about? That's exactly what it is. It's a group of actually not biologists, but network theory. I don't know what do you want to call it. So pretty famous people working on network theory, well-known people who've used string theory to better understand biological networks. So, I mean, things like blood vessels or roots of trees, corals, that sort of thing. They all form some kind of network.

They have nodes where things branch. And the question is, how do you describe this? It's like what's the math that gives the right network so that we can properly describe it? And people have come up with some ideas, simple rules for how such networks grow biological networks. And they don't properly work because they tend to default on. very particular branches where typically they're planer, so they all lie in one plane. So, and they only have three links coming out from a node. So, and if you look at real word networks, they just typically don't look this way.

So what did these guys do? They took the maths of string theory and said, let's actually. actually look at what's called a closed string. So closed string is basically a loop. And in string theory, what you do is, so you think of this loop, it describes a particle. Typically it's a graviton, so it's not like an electron or something. So it's a very specific type of particle.

And that particle can decay, say, into two particles. and what happens is that you get, if you imagine how this looks like as a process in time, you get this splitting of these tubes, which kind of looks like trousers or something upside down. So with the legs going up. And this is exactly what they do in string theory. So they describe all these processes of these tubes splitting into other tubes basically, so long as you're dealing with the closed strings. And so over the decades,

they've developed a lot of mathematics to deal with this. And so those guys said, well, why don't we just use the string theory maths? It's already there. Maybe it helps. And to make a long story short, it does help. And so it turns out that thinking about the surface of these things, rather than just the length of the connection is what you need to get these biological networks right,

at least some of them, like the ones that they looked at. Because for one reason or another, it's actually the surface area that matters and not as was previously assumed the length of these connections. So I read the paper. I think it's a really neat paper. Like it's kind of a nice use of the mathematics that was already there. Of course, it's not the kind of string theory that people talk about when they think of theory of everything or something, which is a quantum theory. So they didn't use any quantum properties on their strings, right?

So blood vessels are not very quantum. And also, as I said, it's this particular closed strings. that they use and not the more general things that you use to describe, say, apart in the standard model. But nevertheless, I think it's a nice paper. And so it's been, I was a little sad to see that some people dismissed it as hype, like a string theory hype. It's not how I look at it.

I think it's a nice interdisciplinary application of some genuinely useful mathematics that came out of string theory. Absolutely. I mean, again, I think here I agree with you 100%. I think what's neat about is that the networks that they were looking at were in the brain, which always interests me anyway. But the point is that, you know, I've been called a critic of string theory for 40 years. I was a critic of the string theory hype for 40 years ago. But the point is you might say you can't justify something by its side benefits. And that's one can always have that debate. But you can't. You can. can to have a lot of smart people, and there are no doubt there have been a lot of smart people doing string theory working on something without getting anything useful. And what's clearly been useful is the mathematics. String theory has pushed mathematical mathematics itself, pure mathematics, and, you know, Ed Witten won a Fields Medal because of that. But it's also produced applied techniques that have been useful in other areas of physics, too. Doing the, you know, looking at these Feynman diagrams with closed loops and other things have been allowed people

to apply those techniques in other areas. I mean, string theory has been a, I don't know, I mean, I'll get hate malephism. One could say an abject failure or doing what it really was supposed to do, which was the theory of everything, or prove that it was a theory of everything. Maybe it's still well one day be. But it has been, but it has produced techniques that have been useful in physics and now in biology, the mathematics of string theory is fascinating. The tools of string theory is have come up with are useful, and it does a disservice to pretend that nothing useful has come out of it. There have been remarkably useful mathematical tools that have been used in other areas of physics, and I have colleagues who've used them in other areas to understand other particle.

To some diagrams, it couldn't be summed otherwise, et cetera, et cetera, et cetera. And now it's neat to see them used in biology. So I agree. It's not, I don't think it's hype in any way. There's lots of other hype you can criticize. for string theory, but that's not one. And in fact, the fact is it wasn't string theorists doing it, right? It was it was biologists, so that should make you even more,

more comfortable that it wasn't string theory hype. So good for that. Nice to see that I'm always happy when work that's been done turns out to be useful, even if it's not for its original idea. Okay, and now to some biology proper. Yeah, yeah. Now we're going to, we've made segue into biology again, So this one amazed me because, of course, you know, life is so remarkable.

And it's so tempting to think that there's some intelligent design and some purpose. And these things are acting with intentionality. And of course, when they're not. But there's a new result, which I found fascinating it. And it has to do with tumors. These are cancer tumors, hijacking the nervous system so that they can, to protect themselves. You know, there's, I mean, when it comes to higher life forms,

there's all sorts of examples of parasites that take over other beings and, you know, and affect their nervous system in a way that, you know, causes them to drown so the parasites' eggs can, you know, life is horrible in that sense. I mean, it's just amazing the kind of things that can happen at a biological level. But here, it turns out, cancer tumors are associated with nerve cells

and people kind of didn't quite know. what the connection was. And some people claimed there was a connection, but it was very hard to measure because the nerve cells, the neurons are very big. And we do a biopsy with a tumor, the actual DNA associated with the nerve cell can be far away from the tumor so you don't measure these things. But what these experimentalists have discovered is that certain tumors basically hijack neurons that give signals the brain, so-called vagus nerves, that then basically cause the brain to give a signal that suppresses the immune system in the region of the tumors so the tumor can grow. So it's basically a tumor hijacking a system that would otherwise cause it to not be able to grow.

It almost sounds like intentionality. Of course, it's not. But it is interesting that this vagus nerve generally, so it secrete something called neo-adrenaline, I think, or noradrenaline, which suppresses, it's used to normally suppress inflammation, you know, when you have too much inflammation around a problem. And so this suppresses inflammation, but it therefore also suppresses the kind of antibodies and things that normally would fight some growth.

And what they were able to do to prove that this happens is use a genetic technique that knocks out certain, basically knocks out chemicals that would or genes that would cause, that would basically cause this vagus nerve to produce this

this inflammation suppressing object. And when they knock that out, when they knocked that out from the nerves that were connected to tumors, the tumors, the growth of the tumors was reduced by 50%. So they were basically showed by knocking out that, basically that pathway to the brain

that then otherwise was hijacked and produced something that would stop the immune system for stopping tumor growth. When they knock that out, the tumor growth was suppressed. And I find it remarkable how, you know, when you think about how fighting cancer is so difficult, that biology is incredibly effective over time at basically everything you can think about.

And tumors have, over time, the ones that survive have been effective at hijacking biochemical systems that would stop their growth. And maybe it's not surprising that they can do that because if they couldn't do that, they wouldn't grow. But it's still, it's amazing. It almost sounds like intentionality. And it once again reminds us how complicated biology is. And also, therefore, how complicated it is to fight. things like cancer, which has been around for a long time and figured out ways to, and by saying, figure out, I don't mean intentionality, but life, but, but, but, but, but, but, but, but, but, but, but, but, but, but, but, but, but, but, but, but,

it's not so easy to counter them. I found it a fascinating piece of, um, of, um, of work and a fascinating fact. Anyway, comments. Was this a study in, in, in humans or was it mice or what? Oh, oh, that's good. Thank you. It was in mice. It was in mice and it was long, I think it was lung cancers and mice, which are, I mean, it does, so it doesn't mean all tumors do this, but it shows the kind of capabilities. And I think it's particularly, from what I understand, it's particularly from looking at this that lung cancer tumors are ones that seem to have nerves near them. That's why I think they, they, they, they looked at that. Yeah, I think one of the main messages that I've taken away from the cancer research to the extent that I understand it.

Yeah, yeah. In the past 20, 30 years is that it's a much more complicated disease than we thought. Like, it's incredibly complex and it's also not just one disease. It's many different ones. And one tumor is not like the other ones. And it depends on many other factors that are going on in any individual. body can affect how these tumors grow or don't grow. And it depends on things like hormones, on other problems that you may have or not have on your

nutrition. Depends on what time of the day you take the drugs, that kind of stuff. And so now they're trying these targeted therapies with, you know, specially adapted gene modified, whatever. And I think it's super amazing. And so it's like, I think like at some point, like maybe 50 years ago, we have this idea like there would be one, one drug that would just kill cancer, right? Because it's all the same thing.

But I think that by now it's like, it's like clear. This is not how it's going to go. Like it's just too diverse this. illness. Yeah, and it's complicated. Yeah, exactly. It's diverse. There's probably many different routes that it uses, many different defenses. But it is fascinating

to see how it can actually hijack even the brain in some sense, which amazed me. And it reminds me once again why I'm so happy I'm a physicist because it's so much easier than biology. But, you know, we've done our three apiece, but I have to squeeze in.

in one little extra thing. Do you have a dog? No, so you, not something. My dog Levi is sleeping here next to me, and I love dogs, as people may know. But there's a, there was a result I can't resist just throwing out. Because I watched it. There's video of these dogs.

Certain dogs, not all dogs, basically have the capability of understanding words and language at the level of an 18-month-old child. And it's really amazing. And of course, they're the kind of like, dogs that heard sheep, you know, the really smart ones. They're generally the ones that can do that.

You can watch it. So there's a video of, and there was a study done, but the video is amazing, of someone talking to their neighbor about a toy that they just bought and mentioning the name and talking to the neighbor. And he can see the dog looking back and forth between the two of them, listening. And then later on, that toy is put in a room

with lots of other toys. And the owner will say, go get that toy and the dog will get that toy. It listens to conversation that picks up linguistic cues at the level of an 18-month-old child and I I just, I love dogs but it amazed me. It's really amazing is amazing. You know, we're learning about the cognitive ability of other species and as a and as a vegetarian

I'm always pleased when I to be a vegetarian and and because lots of species have that we're learning that the cognitive abilities of other species are greater than we thought. And anyway, it convinces me that once again, dogs are amazing. And I couldn't resist throwing it in because I think it's a lovely, a love. And I suggest you can probably look it up online, look at the video. It's kind of fun. Anyway, a little plug for dogs.

I didn't look at the video, but I looked at the paper because I see a lot of those pet studies. They're usually not particularly good. Yeah, yeah. They have miserable statistics and everything. But this one was fine. Looked at the paper. And it was a very small sample. Yeah.

And the point is that it's very, very specific to a few kind of species. And even among those species, only a few dogs. But it is remarkable. And again, it also indicates something that's relevant to our cancer discussion. Is that the thing about biology is that, you know, it's a distribution. And so there are outlawing. and that's a very important part of biology. It means that you can't always assume, you know,

that the norm applies to all cases, and that's probably one of the other aspects that's probably going to be difficult when it comes to fighting cancer. You know, there's so many, there's such diversity of, because of the way evolution works with mutations and there's such diversity in a way thing to work that the norm doesn't necessarily, unlike in physics,

where, you know, there's a mechanism. In biology, there may be many mechanisms. And that's why we're lucky to be physicists. And I'm lucky to have you as a colleague and as an explainer of science as well. And this was a wonderful chance to have a chat. I really enjoyed today. Thank you very much, Sabina. It was wonderful.

Well, thank you. Thank you. Hi, it's Lawrence again. As the Origins podcast continues to reach millions of people around the world, I just wanted to say thank you. It's because of your support, whether you listen or watch, that we're able to help enrich the perspective of listeners by providing access to the people and ideas

that are changing our understanding of ourselves and our world and driving the future of our society in the 21st century. If you enjoyed today's conversation, please consider leaving a review on Apple Podcast or Spotify. You can also leave us private feedback on our website if you'd like to see any parts of the podcast improved. Finally, if you'd like to access ad-free and bonus content, become a paid subscriber at Originsproject.org.

This podcast is produced by the Origins Project Foundation as a non-profit effort committed to enhancing public literacy and engagement with the world by connecting science and culture.