Latest / From First Principles / Winter Olympics Deep Dive: Ice Physics, Performance Pressure, and Climate Change (EP. 26)

Transcript

Winter Olympics Deep Dive: Ice Physics, Performance Pressure, and Climate Change (EP. 26) · From First Principles

• 0:00
  It's been 200 years and we still don't know why ice is slippery.
• 0:06
  We're getting close, but you know, it's... What?
• 0:09
  Yeah, it's kind of crazy.
• 0:11
  We have to lock in for this next story because it's about the neuroscience of choking under pressure.
• 0:16
  That's right.
• 0:17
  Which in this Winter Olympics we have seen in real time.
• 0:21
  The future of the Winter Olympics is not looking good, okay?
• 0:25
  Climate change is coming, and the Winter Olympics specifically are very much at risk.
• 0:32
  Hello, Internet.
• 0:32
  This is your captain speaking, Lester Nare, joined as always by my co-host and our resident PhD, Krishna Chowdhury.
• 0:40
  For those watching, if you could not already tell, this week we're going to be doing a deep dive on the Winter Olympics today.
• 0:48
  We'll be right back.
• 1:03
  With a story about ice being slippery.
• 1:05
  Yeah.
• 1:06
  In the Winter Olympics, every sport depends on ice being slippery.
• 1:09
  But the science and physics specifically behind it is actually non-trivial.
• 1:13
  Yep.
• 1:13
  We'll follow that up with a neuroscience story around the idea of choking.
• 1:19
  right?
• 1:20
  Obviously, the stakes are high, especially at the Olympics.
• 1:23
  And there is really good neuroscience around why do people choke when the stakes are so high.
• 1:28
  We'll wrap up with our climate change story and a couple of other interesting tidbits as it relates to our deep dive on the Winter Olympics.
• 1:38
  You're going to learn stuff today because we're going to break down the science from its fundamentals because this is from First Principles.
• 2:02
  My friend, we are donned and prepared for our Winter Olympics deep dive.
• 2:09
  Yes.
• 2:09
  My first question for you is...
• 2:12
  What event?
• 2:13
  Because I know what you're good at.
• 2:15
  And so what event are you going to trial for?
• 2:18
  Yeah, I think the event that I compete in is the full send.
• 2:22
  The full send.
• 2:23
  The full send event, which is you're at the top with your skis, and then you just French fry the whole way down.
• 2:29
  No pizza.
• 2:29
  No pizza.
• 2:30
  You just French fry, and then you end up getting dragged out on one of those ski patrol sleds.
• 2:37
  Yeah.
• 2:37
  Who can do that the fastest?
• 2:39
  Yeah.
• 2:40
  With the least amount of actual injury.
• 2:43
  I compete in it all the time.
• 2:44
  It's great.
• 2:45
  Mammoth knows all about it.
• 2:48
  This is so good.
• 2:50
  So we're going to go ahead and get started with our physics story of the day.
• 2:54
  And the physics story is starting out with this idea of ice being slippery.
• 3:00
  Why is ice slippery?
• 3:02
  It turns out it's apparently not such a simple problem.
• 3:06
  And scientists have apparently been working on this for 200 years, but we still don't have a foolproof answer.
• 3:13
  So let's kind of talk because there's so many events in the Winter Olympics that deal with ice.
• 3:18
  Yeah, and they're all dependent on this central physical fact that ice is slippery.
• 3:23
  Yes.
• 3:24
  Right?
• 3:24
  And it's been a 200-year-old question.
• 3:27
  It's pretty incredible, the slipperiness of ice versus the friction on other solids, right?
• 3:32
  This is a great example that we saw in Minnesota.
• 3:36
  Our local Gestapo agents were...
• 3:39
  Named ice slipped on ice.
• 3:43
  This is not a political statement because this is not a political show.
• 3:45
  We are not a political show.
• 3:47
  I just think it's funny that ice slipped on ice.
• 3:52
  I just think that's funny.
• 3:54
  I don't think that's weird to bring up in a science podcast.
• 3:59
  We're talking about the slipperiness of ice.
• 4:01
  Yeah, we're talking about the slippery slope of ice, but I'm talking about the physics.
• 4:05
  I'm not talking about the extrajudicial powers of our federal police.
• 4:11
  No, no, no, no, no.
• 4:12
  I'm talking about the physics.
• 4:13
  So don't you dare come after me.
• 4:16
  in the comments.
• 4:17
  Actually, please do.
• 4:18
  I want to see what you'll come up with.
• 4:20
  But in any case, that's the central question that I want to answer today.
• 4:25
  And in fact, I'm not going to answer it because the research is still ongoing, which is quite incredible.
• 4:30
  It's such a simple question, right?
• 4:32
  Like, why is ice slippery?
• 4:33
  It's relevant to stuff like glaciology, like glaciers moving off continents into the oceans and then raising our sea levels.
• 4:41
  It's relevant for transportation.
• 4:43
  You don't want to slip on ice.
• 4:44
  Mm-hmm.
• 4:45
  um and obviously it's relevant for winter sports um the standard explanations are pretty bad okay okay they seem satisfying from like a high school physics level oh okay there's a layer of water and then the water makes it slippery but when you just go a little bit deeper it doesn't make any sense and it's it's actually quite interesting so
• 5:09
  Let's go to the 200-year-old question, right?
• 5:12
  200 years ago, there was this guy, Michael Faraday.
• 5:15
  He is the greatest of all time in terms of experimental physicists.
• 5:19
  You can come at me in the comments again.
• 5:21
  This guy is the GOAT, okay?
• 5:24
  Um...
• 5:25
  He's also actually the founder of scientific communication.
• 5:28
  It wasn't Carl Sagan.
• 5:30
  People think that Carl Sagan was this first guy.
• 5:32
  No, Michael Faraday used to have Christmas lectures at the Royal Institution where he would demonstrate electricity and magnetism for the common public.
• 5:41
  And there are these very famous...
• 5:43
  you know, outreach events that happened in the 1800s in London.
• 5:47
  So, you know, this guy's an OG and I love everything about him.
• 5:51
  And one of the many things that he did, he did, you know, electricity and magnetism is what he's really famous for, but he actually did experiments all over the place.
• 5:59
  And he's the first to observe something called regulation, which is this idea that there's a permanent liquid-like layer of water around every ice block.
• 6:10
  Yeah.
• 6:11
  Okay.
• 6:11
  And people sort of took that.
• 6:14
  observation and said oh that's why ice is slippery because there's a little bit of water on top right i just want to make a quick note that we are not talking about the system that puts a premier league team in england is that called regulation it's called relegation oh relegation yeah i've heard that i've heard that from my cousins are really into arsenal and they're like oh they're gonna get so for example used in a sentence tottenham hotspur
• 6:44
  is potentially in the relegation zone.
• 6:46
  Oh, are they really?
• 6:47
  Yes.
• 6:48
  However.
• 6:48
  Did they have a terrible season?
• 6:50
  So they're going to go down?
• 6:51
  They're potentially going to go down.
• 6:52
  How do you feel about that?
• 6:54
  You're a Chelsea fan.
• 6:55
  I am ecstatic about it.
• 6:56
  Right.
• 6:56
  Yeah.
• 6:57
  Okay.
• 6:57
  That's what I thought.
• 6:58
  I'm ecstatic about it.
• 6:59
  It's happened again.
• 7:00
  It's happened again.
• 7:01
  Tottenham Hotspur.
• 7:03
  It's happened again.
• 7:04
  Okay.
• 7:06
  But we're talking about.
• 7:09
  Regulation.
• 7:09
  Regulation.
• 7:10
  Which, just to be clear here.
• 7:12
  Yeah, that's the liquid layer of water on top of ice.
• 7:16
  And this liquid layer can be nanometers thick.
• 7:19
  It can be only like a few molecules to tens or hundreds of molecules.
• 7:22
  Okay.
• 7:22
  Right?
• 7:23
  And for the longest time, Michael Faraday did some experiments that showed that this is true.
• 7:28
  And for the longest time, it was like, oh, like those water molecules are what's causing the slipperiness.
• 7:33
  Right.
• 7:34
  When we started getting into some deep thermodynamics, the Thompson brothers, the most famous of whom is Lord Kelvin, he started looking at the statistical mechanics of what happens at this interface between ice and liquid water.
• 7:47
  Liquid water is very interesting because just water in general is very interesting.
• 7:52
  It's one of these unique compounds.
• 7:53
  Because of the geometry of the molecule, it increases in volume.
• 8:02
  When you freeze it.
• 8:03
  This is very strange.
• 8:04
  This is why ice floats.
• 8:06
  Ice is less dense than liquid water.
• 8:12
  The solid phase of water is less dense than the liquid phase, which is why ice floats.
• 8:17
  This has everything to do with the geometry of the water molecule.
• 8:19
  The fact that it's 104 degrees between these two bonds of oxygen, hydrogen, oxygen, hydrogen.
• 8:25
  Um...
• 8:26
  And for the longest time, that's been kind of the dominant explanation is because the volume increases.
• 8:35
  Suppose now I were to press down on ice, right?
• 8:38
  If I stand on ice with skates or something like that.
• 8:41
  I'm pressing down.
• 8:42
  I'm increasing the pressure.
• 8:44
  And what physics wants to do at all times is resist change.
• 8:48
  And so when I press and I compress the volume, it's going to undo the freezing part and become liquid.
• 8:56
  Does that make sense?
• 8:57
  Like when water freezes, it expands.
• 8:59
  I'm pushing in, so I'm going to make it unfreeze.
• 9:02
  I'm going to make the water sort of come back as liquid instead of being in the solid form.
• 9:08
  And that's the...
• 9:10
  I mean, I've heard this explanation even from Feynman.
• 9:13
  There's a really famous interview with Feynman where a guy asks him, you know, why, how do magnets work?
• 9:21
  And he goes on this random tangent stream of consciousness about what do you mean?
• 9:27
  Why does it work?
• 9:28
  And the interview is like, I don't know.
• 9:29
  It's a simple question.
• 9:30
  Why do magnets work?
• 9:31
  And he's like, well, you know, suppose there were an alien around.
• 9:34
  And I said, grandma slipped on the ice.
• 9:36
  Yeah.
• 9:37
  How far deep would I have to go into the explanation to tell him what I mean?
• 9:42
  And then he talks about, oh, well, grandma slipped on the ice because ice is one of these unique compounds that is slippery when you stand on it because of the pressure and all this stuff.
• 9:49
  And he even cited this thing, right?
• 9:52
  But let's just like try to hone in on this explanation.
• 9:58
  The explanation is...
• 10:01
  I squish the ice.
• 10:03
  The ice becomes water.
• 10:04
  Okay?
• 10:05
  That works for negative temperatures like negative 2 degrees Celsius, negative 3 degrees Celsius.
• 10:12
  Okay.
• 10:12
  When you're getting to something like negative 20 degrees Celsius to negative 30 degrees Celsius, I can still skate on that ice.
• 10:20
  But this explanation won't work because I'm so cold, no matter how much I press, I'm so cold.
• 10:26
  The physics at that scale, the statistical mechanics of this bulk body, it's not going to let the water come out.
• 10:34
  The idea being when temperatures are colder, simply applying pressure from your ice skate onto the ice is no longer going to change that surface layer from its solid state to its liquid state.
• 10:48
  And so...
• 10:49
  You actually got it right there, the surface layer.
• 10:52
  There's some different physics happening at the surface layer.
• 10:54
  Interesting.
• 10:55
  And we can't count on the big equations that we use in statistical mechanics.
• 11:00
  For example, the main one is the Clausius-Clapeyron equation, and that's the one that gives us this phase diagram.
• 11:06
  We can't actually use that in this scenario.
• 11:08
  Mm-hmm.
• 11:10
  What we need to do is think about what is actually happening at the surface level.
• 11:14
  It can't be pressure because of what I just told you, right?
• 11:17
  It's not enough to explain the slipperiness of ice at really, really cold temperatures, at negative 20, negative 30 degrees Celsius.
• 11:26
  So in 1939, Bowden and Hughes came up with an explanation.
• 11:30
  They said that actually there's heat generated by friction, and that heat is what causes melting.
• 11:37
  Okay, so now we've got a good way to get a layer of water in between whatever I'm standing on.
• 11:44
  Let's say skates.
• 11:45
  Let's use skates for this whole example.
• 11:47
  I'm standing on skates.
• 11:49
  Between my skates and the ice, there's going to be a tiny liquid of water because layer of water because I'm going to move with the skates.
• 11:55
  That's going to cause friction, and the friction is going to cause heating, and that heating is going to cause some melting, and I'm going to get a layer of water.
• 12:03
  What it makes me think about is when, you know, you were in kindergarten or first grade, you're running around and you're in the classroom and the classroom had a rug.
• 12:11
  And somehow you ended up sliding on your knees like Harry Maguire after scoring a header and you get rug burn and it would feel hot.
• 12:18
  Yes.
• 12:19
  Because your knees were moving on the surface.
• 12:21
  That was friction and it was heat and the heat burned my flesh.
• 12:26
  Yeah.
• 12:26
  But similarly, the.
• 12:28
  But also, if you were to measure the temperature of the carpet.
• 12:30
  Yep.
• 12:31
  It would be a tiny bit warmer.
• 12:32
  It would be warmer.
• 12:33
  Yeah.
• 12:33
  And so the point is, you know, this contact between the bottom of the skate and the top of the ice layer, we understand intuitively as just regular people.
• 12:44
  Yeah.
• 12:44
  That when we rub our hands together or we slide our knees on rugs, there's heat there.
• 12:48
  There's heat there.
• 12:48
  And that's the entry point potentially for what they're trying to say here.
• 12:52
  Yep.
• 12:52
  Yep.
• 12:53
  Exactly.
• 12:53
  Bowden and Hughes.
• 12:54
  Yes.
• 12:54
  That was Bowden and Hughes in 1939 out of the Royal Society.
• 12:58
  Okay.
• 12:58
  That's all fine.
• 12:59
  Okay.
• 13:00
  Except the lubricant in this case is still liquid water.
• 13:04
  Yes?
• 13:05
  Yeah.
• 13:05
  In between the two contact surfaces, right?
• 13:07
  There's ice.
• 13:08
  There's my skates that are metal.
• 13:10
  And the lubricant in between is liquid water.
• 13:12
  And that's what you're saying is what's causing slipperiness, right?
• 13:15
  Liquid water is a terrible lubricant, okay?
• 13:20
  Because it has very low viscosity, right?
• 13:24
  point yeah the reason why life works the reason why water can go up the phylum and xylem of plants is because it's very low viscosity so the leaf when the leaf pulls on some water it'll go all the way up to the roots and that force will get relegated right that makes sense so it's it's actually a feature of water that it's very low viscosity yes
• 13:44
  But that's not going to work in this case because whenever we want to lubricate metal on metal, for example, we use oil, right?
• 13:52
  Like car oil and things like that.
• 13:54
  Those things have very high viscosity.
• 13:56
  The point of high viscosity is if I have a contact...
• 13:59
  between two pieces i don't want the liquid in between to give way right i want it i want there to be a film in between such that there's no metal on metal contact or in this case solid on solid contact you need high viscosity such that the the the liquid actually resists that tension the pressure the pressure yeah yeah
• 14:22
  Liquid water is not going to do that.
• 14:24
  Got it.
• 14:24
  Liquid water is just going to get the hell out of the way.
• 14:27
  And so the thickness of that film, if you were to actually the hydrodynamic thickness of that film is going to be nanometers or less compared to something like a highly viscous substance like oil.
• 14:40
  That's why we use oil in our engines and not water.
• 14:43
  If we used water, the engine parts would be rubbing up against each other.
• 14:46
  It would not be a good time.
• 14:47
  Mm hmm.
• 14:48
  Right.
• 14:49
  So that's the central discrepancy.
• 14:51
  Now we're at 1939.
• 14:52
  Yes.
• 14:53
  We figured out why there's liquid water.
• 14:55
  Yes.
• 14:56
  But we haven't quite figured out how the liquid water becomes this lubricant.
• 15:00
  It really shouldn't.
• 15:01
  Right.
• 15:01
  Because because what we're saying is just by the skate blade in our analogy contacting the ice.
• 15:07
  Mm hmm.
• 15:08
  On its own, based on what we know, it should not create that nanometer scale layer of water.
• 15:14
  Yeah, yeah.
• 15:15
  Like, how does that happen?
• 15:17
  Yeah, like, if it's a nanometer scale... No, if we do the physics, the scale should be nanometers.
• 15:24
  It should be nanometers.
• 15:25
  But it needs to be a lot thicker.
• 15:28
  Nanometers is like tens of atoms One nanometer is ten hydrogen atoms That's not enough To keep me from contacting the ice Right?
• 15:37
  You know what I mean?
• 15:38
  So that's the idea How do we get a layer of water
• 15:44
  And we get that layer of water to be more viscous than bulk liquid water.
• 15:48
  Understood.
• 15:48
  Okay?
• 15:49
  And that's where there was this beautiful paper in 2019 in the Physical Review X by Canale and other authors, Nanoreology of Intrafacial Water During Ice Gliding.
• 16:00
  This is a paper that tried to understand why the water was slimy between my skate and my ice.
• 16:09
  Okay?
• 16:10
  Is slimy a technical term?
• 16:12
  Yes, I believe slimy is viscosity is anything above whatever oil is.
• 16:17
  I don't actually know what the numbers or the units for viscosity are.
• 16:21
  That's something I should look up later.
• 16:22
  But the point, it is meant to be a technical description.
• 16:25
  Yes.
• 16:26
  It's meant to be.
• 16:27
  It's much more slimy than normal water.
• 16:28
  Normal water is not slimy.
• 16:30
  Yes.
• 16:30
  Right?
• 16:30
  I can just, like, get rid of it.
• 16:32
  It'll evaporate away.
• 16:33
  Slimy stuff doesn't evaporate.
• 16:35
  It's viscous.
• 16:35
  It's sticky.
• 16:36
  Things like that.
• 16:37
  Okay.
• 16:37
  And we need that.
• 16:38
  Okay.
• 16:39
  So the key challenge is that this film between the ice and where the water is sitting between my blade is an ephemeral film.
• 16:48
  It only exists during this dynamic act of sliding.
• 16:51
  Let's get into some of the chemistry behind this.
• 16:53
  what ice is, and how that layer works.
• 16:57
  So you've got a solid layer.
• 16:58
  The solid layer is very high molecular order.
• 17:02
  You've got nice water molecules that are stuck in a crystal form, regularly spaced, characteristic of solids, right?
• 17:10
  So all of these water molecules are sort of, you know, interacting with each other, but they make this hexagonal lattice.
• 17:17
  The layer in between my ice and my skate is going to be the pseudo liquid layer where there's going to be a lot of molecular disorder, right?
• 17:27
  Because if you go down, the order is really high.
• 17:30
  And if you go up, it's fully liquid.
• 17:32
  There's going to be some transition where the water molecules are behaving in a very unique way, right?
• 17:37
  Mm-hmm.
• 17:38
  That's what this paper was trying to solve.
• 17:40
  What is the chemistry and the physical properties of that interfacial layer of water molecules?
• 17:46
  Where there is a high disorder.
• 17:48
  Yes, where there's high disorder.
• 17:50
  And can that actually tell us something about why the viscosity is so high?
• 17:55
  Right?
• 17:55
  And...
• 17:56
  What they used is something called an atomic force microscope, but with a tuning fork.
• 18:00
  So an atomic force microscope is this incredible technology.
• 18:03
  This is one of the technologies that I thought should win the Nobel Prize this year that I called.
• 18:08
  Didn't get it, but I think it's coming.
• 18:10
  The AFM is an amazing tool.
• 18:13
  It's effectively a cantilever.
• 18:15
  So you've got a little pointy thingy that's pointed downwards, right?
• 18:19
  And if I've got a sample with molecular level irregularities, like I'm talking tens of nanometers to less...
• 18:26
  When I bring this thing up, it's going to move up and down, right?
• 18:29
  Because the literal atoms on my surface are going to resist the cantilever.
• 18:34
  And if I have a laser that's shining on the cantilever, and then it's on a detector, as the cantilever moves, the laser on that detector is going to move, right?
• 18:43
  And then I'm going to be able to sense these tiny atoms.
• 18:47
  surface level atomic irregularities and that's what the atomic force microscope is doing it's it's been revolutionizing surface chemistry studies like all over the place from biology to chemistry everything right
• 19:02
  I think if I remember correctly, the way we analogized it was it's like when you have a record player.
• 19:09
  Yes, exactly.
• 19:10
  And it's reading a record, but just with a laser for precision on the readings.
• 19:15
  And the records are at atomic scale.
• 19:17
  Atomic scale versus whatever records are currently.
• 19:20
  But just as a visual analogy for folks, that's kind of the reference.
• 19:23
  Yeah, that's exactly right.
• 19:25
  And with the atomic force microscope...
• 19:27
  The problem is you're only going to read this sort of X, Y... Or sorry, the Z displacement, right?
• 19:32
  The up and down displacement.
• 19:34
  What we want to know with ice, because we're sliding, is we got to get this, right?
• 19:39
  So they made a tuning fork, Autonomic Force Microscope.
• 19:44
  A tuning fork is if you're a musician, you know these like two prongs of metal that will vibrate at a specific frequency.
• 19:51
  So if you want an A...
• 19:54
  the size of these prongs tells you that it's only going to vibrate at an A.
• 19:57
  And so when you bounce it, it's going to be like, and it's only going to be at this really clean tone.
• 20:04
  They took a tuning fork microscope and they attached it.
• 20:07
  Sorry, they took a tuning fork and they attached it to an atomic force microscope.
• 20:11
  They attached that to a bead that would interface with the ice.
• 20:15
  So the bead is kind of like my skates.
• 20:17
  Yes.
• 20:17
  Yes.
• 20:18
  I have a tuning fork on top.
• 20:20
  Yes.
• 20:20
  Okay?
• 20:21
  That tuning fork is going to vibrate sideways.
• 20:23
  Yes.
• 20:24
  And that sideways vibration is really tapped in to whatever resonant frequency that tuning fork is at.
• 20:30
  Right?
• 20:31
  Yes.
• 20:31
  If I've tuned that tuning fork to an A, it's only really going to vibrate at an A.
• 20:35
  Right.
• 20:35
  Right.
• 20:36
  Right.
• 20:36
  Yes.
• 20:37
  And the up and down can be the atomic force part.
• 20:39
  I see.
• 20:40
  But the sideways part can be my tuning fork vibrating at a very specific frequency.
• 20:44
  Yes.
• 20:45
  Okay.
• 20:45
  So now I can get both the up and down motion.
• 20:49
  Yes.
• 20:50
  And the side to side motion.
• 20:52
  And I get two resonance modes in this case.
• 20:55
  I've got a tangential mode, which is my stroke.
• 20:58
  And that's going parallel to the lubrication layer.
• 21:03
  Yes.
• 21:03
  So I can characterize the little thin film of water that's created.
• 21:08
  How is it...
• 21:10
  responding this way and how is it responding this way independently so so what we've done now uh with this uh afm plus tuning fork is we've added an additional variable layer that we can measure before we were measuring in one dimension yeah we were just doing okay what's the height of this thing right but now it's like what's the mechanical response this way and so now we have two different dimensions of measurement
• 21:39
  that can be appropriately and robustly measured independently of each other, which means we can better characterize what's happening because we have more data and information.
• 21:49
  Exactly.
• 21:50
  Yeah.
• 21:50
  We have data in all three dimensions, and we also have data in time.
• 21:54
  That's going to be huge, right?
• 21:56
  Because with the tuning fork, I can make it resonate at a certain frequency, and I can tell what the responses of my...
• 22:04
  material, right?
• 22:05
  So there's going to be two things, and this comes from, you know, if there's any electrical engineers out there, they must have heard of this concept of impedance, which is the idea, it's kind of like a resistance, but for AC circuits, for the alternating circuits, because when you have direct current...
• 22:21
  The current goes through, the resistance sort of just like stops the current from happening.
• 22:25
  And then the output voltage is going to be a little bit lower because you've lost a bunch of energy inside that resistor.
• 22:32
  When you're doing alternating current and you put a resistor in there,
• 22:36
  the output is going to be another alternating current, but there's going to be a phase shift.
• 22:42
  There's going to be a time delay because the resistor is not only reducing the amount of energy, it's also causing a time delay.
• 22:49
  And that's actually what they see with this water.
• 22:52
  This water has mechanical impedance.
• 22:55
  So there's an elastic property, which makes it sort of dissipate energy, but there's also this viscous dissipative component that is creating dissipation.
• 23:06
  poke the tuning fork and the tuning fork is going in this way, my response is going to have a delay.
• 23:11
  And that delay tells us a lot about the viscosity of that layer.
• 23:14
  And it turns out that layer is highly viscous.
• 23:17
  It's not the viscosity of normal water.
• 23:21
  And so this is the first indicator of the slimy water that we talked about earlier.
• 23:26
  On Amit's face...
• 23:27
  Water is not very viscous, which allows it to, in the tree example, float up because it's not introducing new friction.
• 23:39
  Yeah, it's very slippery.
• 23:39
  It's very slippery, which means it can move around very easily.
• 23:43
  In order for skating, ice skating, to work in the way we think about it, at a minimum, the top layer of the big body of ice...
• 23:52
  Needs to be extremely viscous.
• 23:53
  Exactly.
• 23:55
  Meaning not slippery.
• 23:57
  Meaning being able to catch onto, basically grab onto the blade as it's moving on top of the surface.
• 24:03
  And this is the first indication where we can measure and see that, oh, actually that top layer is highly viscous.
• 24:11
  And if you were to now calculate, how thick would the film be?
• 24:15
  Yes.
• 24:16
  It's now 100 to 500 nanometers.
• 24:19
  Instead of just a couple.
• 24:19
  Yeah, it's like two orders of magnitude above what we thought it should be before.
• 24:23
  And so now that layer is enough.
• 24:26
  Right.
• 24:26
  To lubricate this contact between my blade and the ice.
• 24:31
  And the ice.
• 24:32
  Right.
• 24:32
  And this was in 2019.
• 24:34
  It was, I thought it was a really cool experiment.
• 24:37
  Just the way that they constructed an AFM, which is pretty established now.
• 24:45
  displacement as well i just want to briefly note this is a very interesting point about how when tools are built afm crisper etc etc tools are let's say platforms however you want to categorize it they're not uh in a finite state of completion no right you can add your aftermarket attachments yeah on top of it for different use cases and this is a perfect example of taking afm and having a little additional attachment on there that allows you to see stuff a little bit differently that's very cool
• 25:14
  Yeah, yeah, yeah.
• 25:15
  It's very cool.
• 25:15
  And the other thing that they could tell from this new technology is like not only why is it viscous, they could also tell that the reason why that thing was viscous is because there's tiny little ice crystals in that layer of water that's causing the viscosity to go up.
• 25:30
  Interesting.
• 25:31
  Right?
• 25:32
  And that's what makes sense.
• 25:32
  So now when it comes to like, you know, why do we put wax on our...
• 25:39
  on our coating like when we do skis and stuff like that i mean the at the olympics everyone's waxing their skis right and the reason why you want to wax your skis is because they're hydrophobic they they don't like water on it right and that's actually enhancing this viscous property of that layer which is which is kind of cool it's like kind of it's making that
• 26:03
  more scientifically justified before it was just like an observation it's like if i wax like it just works yes and people had hand wavy ideas right but now we're really getting into the physics of it it's very very cool okay um it hasn't stopped there because that was done at you know very very low temperatures
• 26:21
  Now, how do we bridge the gap?
• 26:23
  Right?
• 26:23
  Because this is clearly happening at all temperature scales.
• 26:26
  And that kind of theory only really works for this very specific temperature.
• 26:30
  Ah, that's interesting.
• 26:31
  So it's not done.
• 26:33
  We're not done.
• 26:33
  It's not universally true.
• 26:35
  It goes back to where there's a difference between what happens at negative 3 degrees versus negative 30.
• 26:40
  And so the AFM tuning fork solution and how the slimy water layer was being created is true at very low temperatures.
• 26:51
  Yeah, not like the negative two, negative three that we see in winter sports.
• 26:54
  Right.
• 26:55
  But it's a clue at least as to why that part works.
• 27:01
  It's like the game Clue where we found the candelier in the kitchen and it has some fingerprints on it, but we've not yet done the fingerprint analysis.
• 27:10
  Yeah, yeah, exactly.
• 27:11
  And then in 2024, there was another paper, Hong and others in nature.
• 27:16
  What they did was atomic scale visualization using cryogenic AFM.
• 27:21
  So now they upped their AFM game and they were doing atomic scale resolution all the way down to like angstrom level.
• 27:32
  Wow.
• 27:32
  Okay.
• 27:33
  And what they found was there's this thing called pre-melting.
• 27:37
  Which initiates a kind of defective boundary because the ice is not going to be a clean hexagonal lattice, right?
• 27:43
  There's going to be defects where like there's this domain and then it's like kind of interacting with this other domain of hexagons.
• 27:48
  But it's the hexagons don't like completely line up.
• 27:52
  And what that does is where those things don't line up, you get this atomic defect that compounds to create that layer of like weird water.
• 28:05
  Okay, so that was in 2024.
• 28:08
  2025, there was another paper with Attila and Musser.
• 28:13
  They proposed something called cold self-lubrication at very low temperature.
• 28:17
  So now they're going to negative 40 degrees.
• 28:19
  And they're trying to say that there's some mechanical shear stress.
• 28:22
  So when you have a lattice, the shear is like how much you sort of warp it this way, like contort it.
• 28:28
  And when you have that shear stress, that creates the disordered amorphous stress.
• 28:33
  solid like layer that then becomes this viscous lubricant between our stuff so as you can see like it's it's still a very ongoing
• 28:43
  field of research 2025 was that last paper 2024 was the paper right after 2019 was this afm paper that which was kind of like the really initial like an initial the initial clue that has let us down a like a path yeah to a solution it's just six years ago yeah yeah and and people are still investigating it i mean it's got an incredible importance for our society right if we just talk about climate modeling and climate change which we're going to get to later on in the in the
• 29:12
  episode, if water is a hundred times higher viscosity than,
• 29:19
  right in between interfaces of ice and other solids, then that completely changes the current models about how glaciers slide off continents.
• 29:30
  Interesting.
• 29:31
  Right?
• 29:31
  Yeah.
• 29:32
  Yeah.
• 29:32
  Because before we had this number and we're like trying to model, okay, the glacier is like going, this is the friction, but now we've got a completely different number.
• 29:39
  And the number is two orders of magnitude different, but a hundred times different.
• 29:43
  So now the, the sliding rate is going to change.
• 29:45
  It's going to be get a lot higher.
• 29:46
  Right.
• 29:47
  Just to be clear, your note about the difference is we initially talked about how the assumption was it's, you know, two to three nanometers or a couple of nanometers.
• 29:55
  Yeah.
• 29:56
  Is the slimy water layer that enables the ability to skate, quote unquote.
• 29:59
  Yeah.
• 30:00
  We've now with AFM tuning fork and follow on research identified that that's in the hundreds of nanometers.
• 30:07
  Yeah.
• 30:08
  What we've clearly done, the layer depends on how much pressure you're putting, right?
• 30:12
  Like, if there's a human body weight on a bit of Tiny Skate, I mean, there's a lot of pressure on that tiny little, like, you know...
• 30:20
  millimeter thick blade but with a glacier glaciers like like kilometers thick worth of right worth of stuff so there's going to be a lot more pressure so the the layer is going to scale that way what we've done is confirmed that viscosity is a lot higher yes and that viscosity number is what we're going to put into our models right because now because the viscosity is a lot higher it's a variable when we look at large-scale glacier glacier glaciology mm-hmm
• 30:45
  It's a variable in a larger equation that has meaningful impact on modeling.
• 30:49
  Yes, yes.
• 30:50
  And one of the things that you can think about is like these shear thinning events that happen all of a sudden that lead to a catastrophic flow and acceleration.
• 30:59
  There's this glacier called the Thwaitis Glacier.
• 31:01
  I think that's how you pronounce it in Antarctica.
• 31:04
  This single glacier contributes to 4% of the global sea level rise.
• 31:09
  Yeah.
• 31:10
  Okay, it's massive.
• 31:12
  It's the size of like states in America.
• 31:14
  And it is moving extremely fast.
• 31:17
  Something like two kilometers per year.
• 31:20
  That's really fast for a glacier.
• 31:21
  The surface of the glacier is moving on the underlying landmass.
• 31:24
  Yeah, and you know...
• 31:27
  In this century, it's projected to raise the sea levels by several centimeters.
• 31:32
  If the whole thing went, which I think people are projecting it's going to go in the next three centuries, that's two feet of ocean level rise from a single glacier.
• 31:42
  It's crazy.
• 31:42
  It's crazy, right?
• 31:44
  And now with these new models, it's like maybe this is even faster.
• 31:47
  Right, right, right, right.
• 31:48
  It's crazy.
• 31:48
  Yeah, right.
• 31:49
  This is actually an interesting note on how what seems to be a totally unrelated issue with the question, the entry question of why is ice slippery?
• 31:58
  Yeah, yeah.
• 31:59
  Has all of the second, third, and fourth order consequences on things like how do we model ice?
• 32:05
  uh sea level rise based off of glacial ice uh uh evaporation etc yeah yeah which which makes sense when we talk through it yeah but it's not you don't immediately make yeah you don't yeah you don't i mean you you could be like oh it's such a why do we care well in science there's always other consequences right that deal with these fundamentals right fundamentally why are water molecules so weird has so many consequences right
• 32:30
  So I thought that was pretty exciting because it is such a simple question.
• 32:35
  It's been 200 years that we've been asking it.
• 32:38
  All of our best scientists have worked on it.
• 32:40
  And we're still – the closer you look, the more interesting things are.
• 32:47
  This is this is so and I think what's so fascinating about this is the Winter Olympics have been going on now for I think it's 25 years is what they were saying on the broadcast.
• 32:56
  It's like relatively new as compared to the Olympics that's been around for a very long time.
• 33:01
  I think it's it's been around longer.
• 33:02
  No.
• 33:03
  25 years.
• 33:04
  So we can do a look.
• 33:06
  The way that NBC is... No, 1924 was the first one in Chamonix.
• 33:10
  Chamonix, sorry.
• 33:12
  Chamonix.
• 33:13
  So they must be talking about the way the Premier League talks about the Premier League era versus the eras that are legit or not legit, whatever, what have you.
• 33:23
  It is still fascinating to me that we don't understand the fundamentals of this thing, that we have a massive global event going on that's been around for 100 years.
• 33:31
  Yeah, an industry.
• 33:32
  An industry, billion-dollar industry.
• 33:34
  And we actually don't know why it works.
• 33:37
  Just straight up, why is I slippery?
• 33:39
  For some of the key events.
• 33:41
  Great physics story to start off with.
• 33:45
  we're going to move now into the rundown where we're going to, again, with the rundown, we can't cover every story every week.
• 33:54
  There's so much breaking and frontier research science happening.
• 33:58
  We do our best to cover the stories that we're most interested in and can cover.
• 34:03
  But the rundown gives us an opportunity to share what else is happening in the world of science at a slightly higher level today.
• 34:10
  Before we jump in, I do want to talk about a couple of housekeeping notes.
• 34:16
  For those who are listening to the pod, whether that's on Spotify as audio only, on Apple Podcasts as audio only, we are a video podcast.
• 34:27
  This is funny.
• 34:27
  People say, well, if you have a video, aren't you a show?
• 34:29
  We're not syndicated on an actual network or streaming platform.
• 34:32
  So in my view, definitionally, we are not a show.
• 34:36
  But we are a video podcast.
• 34:38
  And so many of the things we talk about on the show, we accompany with a lot of visuals.
• 34:43
  I think in every episode, we almost have 40 to 60 overlays that we display.
• 34:48
  Yeah.
• 34:48
  So if you're listening to the show audio only and you're having a hard time visualizing the concepts we're talking about, you can always check out the video version directly on Spotify and or on YouTube for the full explanations where we have the overlays.
• 35:04
  It really is helpful to understand the concepts with that visual reference point.
• 35:10
  So if you haven't already transitioned to the video version correctly,
• 35:13
  please do.
• 35:14
  And on top of that, if you are watching on any of those platforms, whether it's Spotify, Apple Podcasts, or YouTube, we need to rise through the ranks of the billionaire algorithms.
• 35:25
  And the best way we can do so is with likes, with shares, with comments, and for Spotify and Apple Podcasts.
• 35:32
  Specifically, if you give us that five star, it helps us get
• 35:36
  our science pod out to more people.
• 35:39
  We love talking about science.
• 35:41
  We'd love to reach a bigger audience.
• 35:44
  And last note on housekeeping here is for folks who are looking to support the show, one of the things that's really key for us is that the show is available as freely and as on many platforms as possible.
• 35:57
  No subscriber bonus episodes, no exclusive access subscriber content whatsoever.
• 36:02
  But if you want to donate to support the show, we now have a new donation portal set up at the website ffppod.com backslash donate.
• 36:13
  You can join one of our three supporter tiers, listener, supporter, or producer, as well as sending over a custom one-time donation of an amount of your choosing.
• 36:25
  It is through this support that we are able to run this podcast with just the two of us.
• 36:30
  Yeah.
• 36:31
  No big, big, bad, evil platform or network.
• 36:35
  It's literally just the two of us.
• 36:36
  It's just the two.
• 36:37
  Just the two of us.
• 36:38
  Yeah.
• 36:39
  We can make it.
• 36:40
  I don't know how it goes.
• 36:41
  We can make it if we try.
• 36:43
  Oh, you can tell who's the singer of the two of us.
• 36:46
  No, but seriously, the engagement's been super fantastic.
• 36:50
  We're excited to grow and expand the show, and your support will do a lot to enable us to do so, including some of our on-site episodes, one of which we have already shot.
• 37:02
  And we are in the edit process for which we are super excited about.
• 37:05
  Now, with all of that housekeeping out of the way, let's go to our first story of the rundown.
• 37:12
  And this one is about AI doing physics.
• 37:15
  Yes.
• 37:15
  So Sam Altman, unclear if he's a human or not.
• 37:20
  or a non-human, non-human intelligence himself.
• 37:24
  But Sam Altman and OpenAR are claiming that ChatGBT is now doing research-grade high-energy physics research.
• 37:32
  What is your take on this recent OpenAI story, which I think was in tandem with a couple of other institutions, maybe namely MIT, if I'm remembering correctly?
• 37:44
  Yeah.
• 37:45
  Yeah, it's a pretty interesting story, and it's a pretty big claim.
• 37:48
  I first heard about it on their Twitter.
• 37:52
  The idea is, in particle physics, we are worried about calculating something called the scattering cross-section of stuff, okay?
• 38:02
  Imagine you're at the LHC or any collider.
• 38:05
  Large Hadron Collider.
• 38:06
  That's the Large Hadron Collider in CERN.
• 38:08
  Right.
• 38:09
  You've got jets of particles that are interacting with each other, and then...
• 38:14
  They have a certain energy.
• 38:15
  And then you've got a bunch of detectors around.
• 38:18
  And what the particles are going to do is, because they're at such high energy, there's going to be interactions of different fields at this quantum level.
• 38:26
  There's going to be a Higgs boson coming.
• 38:28
  There's going to be electrons, protons, like all sorts of random stuff happening.
• 38:32
  And then they're going to scatter off, right?
• 38:35
  And then different particles are going to scatter off.
• 38:37
  And we're going to collect all that data in our detectors from that collision.
• 38:41
  Right?
• 38:42
  What we want to do is calculate what the scattering amplitude and the probability is going to be in each of these directions.
• 38:50
  Because the point is after the collision, these things are scattering in all kinds of different directions.
• 38:55
  All kinds of different directions depending on whatever physics is going on at the point.
• 38:59
  Right.
• 39:00
  And so fundamentally in particle physics, all these guys are doing is calculating these scattering cross sections.
• 39:05
  OK, because that tells us intimately how the fields interact with each other.
• 39:10
  Right.
• 39:10
  How does the electron interact with quarks?
• 39:12
  How do quarks interact with each other?
• 39:14
  And in this case, they're wondering how do gluons interact with each other?
• 39:18
  Gluons are the mediators of the strong nuclear force.
• 39:21
  We went over this in a previous episode, but just to briefly reiterate, the strong nuclear force is what keeps the nucleus together.
• 39:29
  okay nucleus is full of protons which are all positively charged according to electromagnetism they don't want to sit right next to each other they want to blow apart but the strong nuclear force between the quarks inside of these protons is what's gluing the nucleus together inside of an atom and that glue comes from gluons which are the mediating force between all of these quarks that's how the quarks kind of talk to each other
• 39:55
  Now, gluons can interact with themselves.
• 39:58
  A gluon here can interact with another gluon here.
• 40:01
  And the focus of this particular paper is about how those gluons interact.
• 40:07
  There's a certain type of diagram.
• 40:08
  We use Feynman diagrams, which is this sort of tool.
• 40:12
  It's a mathematical tool, really, but also a visual tool that Feynman came up with that helps us understand how these fields interact with each other, with, like, virtual particles, a gluon interacting with another gluon, and so on and so forth.
• 40:24
  And usually when we calculate these scattering cross-sections, we want to figure out all the different ways that a gluon can interact with other gluons and so on and so forth.
• 40:32
  There's a particular way that they interact, which we thought would never happen.
• 40:38
  Because if you were to calculate through the whole integrals of all of the Feynman diagrams, this tree-level diagram, which had no loops...
• 40:45
  The answer came out to be zero, meaning the amplitude was zero, which means the probability that this particular process happens is zero.
• 40:53
  That's what it was thought.
• 40:54
  So the math was saying that there's an outcome that should not happen.
• 40:58
  That should never happen.
• 40:58
  And we feel very confident about the math.
• 41:01
  Yeah.
• 41:01
  So the math was like, this is solid.
• 41:03
  This has been proved true in other use cases.
• 41:06
  And so we should safely assume that in this use case where n equals zero.
• 41:10
  That nothing will happen.
• 41:12
  Yeah, yeah, exactly.
• 41:14
  Right.
• 41:14
  The amplitude should be zero.
• 41:15
  The probability should be zero.
• 41:16
  Then about a year ago, the authors of the study from IAS Princeton, Cambridge, Harvard, they decided that actually it might not be totally the case.
• 41:26
  Okay?
• 41:26
  They went all the way up to N equals 6, which is 6 gluons interacting with one another, and they tried to write out by hand the formula for how probable this outcome would be.
• 41:38
  It's a ridiculously bad formula in terms of just like for a human being to write out all the combinatoric possibilities of this gluon going here and going there.
• 41:48
  What's the probability of this?
• 41:49
  And keeping track of all those integrals.
• 41:51
  They do it by hand all the way up to N equals 6.
• 41:53
  Okay?
• 41:53
  Jesus.
• 41:53
  But what they're getting an inkling of is maybe this is not zero.
• 41:57
  This answer is not all going to cancel out.
• 41:59
  Okay?
• 42:00
  So that's when they employed ChatGPT.
• 42:03
  ChatGPT took those expressions, simplified it, and then conjectured a simple formula...
• 42:09
  That was a general case for all N, not just N equals all the way up to six, but N equals seven.
• 42:14
  You just plug in a number.
• 42:16
  You get the thing out.
• 42:17
  And then very, this was kind of interesting.
• 42:20
  Then an internal open AI model spent 12 hours reasoning behind this thing, came up with the formula on its own, and came up with the proof for that formula.
• 42:29
  That was then corroborated by the authors.
• 42:32
  The proof was verified, and it was correct.
• 42:34
  So let me make sure I'm getting this right.
• 42:36
  So...
• 42:37
  They initially fed this thing to Chachi B.T.
• 42:40
  And it created a simple formula.
• 42:42
  A simplified way of doing that.
• 42:44
  A simple calculator to do the math for n equals any number.
• 42:48
  It figured out some sort of patterns within like n equals 1, 2, 3, 4, 5, 6.
• 42:54
  It maybe figured out a pattern that was going on.
• 42:56
  And it was like, actually, in physics we do this all the time.
• 42:59
  It's called an ansatz.
• 43:01
  It's our best guess for what do you think the general formula should look like.
• 43:05
  So it created its own version of this simplified expression.
• 43:09
  This is a simplified formula.
• 43:11
  And then they fed it to an internal OpenAI model that had a bunch of scaffolding that's specific to scientific research stuff.
• 43:18
  It let it run for 12 hours.
• 43:20
  And this is like an important note.
• 43:22
  Like, you know, the state of the art.
• 43:23
  There was a different model that let it run for 12 hours.
• 43:26
  Correct.
• 43:27
  I do want to be.
• 43:27
  This is not like 5.2.
• 43:29
  Right.
• 43:29
  They had an internal model that ran for 12 hours.
• 43:32
  And so just conceptually for context here, because I think this is interesting, you can take these base models and then what they call, they put scaffolding around it, which is this technical term to mean these additional weights and processes that are hyper-specific to a particular task or use case.
• 43:48
  So this is not just, you could go into chat GPT today and you could then get this result.
• 43:53
  Yeah.
• 43:54
  But the underlying model intelligence is being amplified with this sort of, sort of bespoke use case.
• 44:00
  Anyway, yeah,
• 44:01
  They let it run for 12 hours.
• 44:03
  Yeah.
• 44:03
  And I think this is important because a year ago, you could not have long running agent based autonomous processes.
• 44:14
  And just actually recently in the same time period as this is coming out, the now the maximum time last year, the maximum time was about three hours, a couple hours.
• 44:22
  Uh huh.
• 44:23
  Just about a week ago, the max running time, this was for a Claude Opus 4.6, was two weeks straight.
• 44:30
  And it built a C compiler of 100,000 lines of code from scratch that worked with no human intervention.
• 44:36
  So I just, I want to.
• 44:38
  Yeah, yeah, yeah.
• 44:39
  I heard about this.
• 44:39
  That was crazy.
• 44:40
  It's really, these, being able to run these things for longer enables stuff like this.
• 44:46
  So they ran it for 12 hours.
• 44:48
  And then it came up with a proof.
• 44:50
  Mm-hmm.
• 44:51
  Which then meant the human scientists could then go through that proof.
• 44:55
  And verify.
• 44:55
  And validate that it's true.
• 44:57
  Yep.
• 44:58
  Yeah.
• 44:58
  And it was true.
• 44:59
  And it was true.
• 45:00
  So it is quite fascinating, right?
• 45:02
  And I think it is a big deal.
• 45:05
  Yes.
• 45:06
  Okay?
• 45:06
  From where I'm standing, I think it is a big deal, but there are caveats, okay?
• 45:10
  There are nuances, because I want to understand what is actually happening.
• 45:13
  I do hate these black boxes.
• 45:16
  But, you know, we're in the age of AI black boxes, right?
• 45:19
  Is it actually understanding stuff?
• 45:21
  Or my hypothesis is, well, let's go through.
• 45:24
  Is it actually understanding stuff?
• 45:25
  Is 5.2 understanding something?
• 45:28
  Friend of the show, he's going to be in the acknowledgement section.
• 45:32
  Alan Southworth, he sent me a screenshot of something that he asked, ChatGPT.
• 45:37
  And that's in the next photo.
• 45:39
  I want you to show that.
• 45:40
  Okay, so he asked...
• 45:42
  ChatGPT 5.2.
• 45:42
  Okay.
• 45:44
  I want to wash my car and the car washes 50 meters from my house.
• 45:48
  Do you think I should walk there or do you think I should drive there?
• 45:50
  Mm-hmm.
• 45:52
  You should drive there because it's your car.
• 45:54
  Yeah, right.
• 45:55
  Okay.
• 45:55
  ChatGPD answers, at 50 meters, you are officially in the put-on-sandals-and-stroll territory.
• 46:03
  Right?
• 46:03
  Because in all of its LLM knowledge, it's like, oh, 50 meters, you can walk.
• 46:10
  Because there's probably so many blog posts out there about when is it okay to walk and why to not use gasoline and all this other stuff.
• 46:22
  You know, ChatGPT always asks a follow-up question because OpenAI wants you to keep engaging.
• 46:27
  The follow-up question is, so are you going to get a full detail?
• 46:30
  So it knows that you're doing the car wash the whole time.
• 46:36
  Yes.
• 46:37
  But in the middle, it's like you should walk.
• 46:39
  And then at the end, it asks, are you going to do a full detail?
• 46:43
  Or like, what are you getting about your car?
• 46:44
  Yes.
• 46:45
  Yes.
• 46:45
  Do you understand?
• 46:46
  And the point being, it means it doesn't actually have an understanding of the question and the variables inside the question.
• 46:54
  Because for any, you could ask a five-year-old that question.
• 46:58
  And they'd be like, you should drive the car because you need the car to wash it.
• 47:01
  Right.
• 47:01
  They've associated some type of meaning behind the task that you're trying to do and accomplish, which is get my car washed and the mode of transportation you would need to get there.
• 47:12
  You would have to take your mode of transportation because the task is related to that mode of transportation.
• 47:18
  That's ChatGPT 5.2 that's not able to actually make that connection.
• 47:22
  I actually would be curious.
• 47:23
  Is it?
• 47:23
  Does he have?
• 47:24
  I'd be curious if he had thinking on or off.
• 47:27
  Interesting.
• 47:27
  Okay.
• 47:27
  Because you can't have reasoning on or off with 5.2.
• 47:30
  Oh, okay.
• 47:32
  I'm not saying that it would have made a difference.
• 47:33
  Right.
• 47:34
  But no, that's a valid question.
• 47:35
  I think it is because that's the argument that those on the inside are making.
• 47:40
  It's like, oh, when you just use the base models without reasoning as this additional layer to basically, in theory, check for this kind of stuff.
• 47:48
  Yeah.
• 47:48
  Then, yes, you will get these quote unquote hallucinations.
• 47:51
  But the argument is, oh, reasoning with 5.2 and Opus 4.6.
• 47:56
  covers most of these use cases i'm not saying that's true that is the argument thing that is the argument so alan it would be good to know if you use thinking on or off yeah because if thinking is on it really does i think expose a huge with even within the reasoning pathway it is really not actually understand understanding yeah and from there i wanted to ask like okay like how
• 48:20
  If it's not understanding, how is it able to do this physics?
• 48:23
  Right.
• 48:23
  Right?
• 48:24
  Right.
• 48:25
  I have a hypothesis.
• 48:26
  I don't know if it's correct, and perhaps someone can tell me I'm wrong.
• 48:29
  My hypothesis is that Chachipiti, in all of its wisdom and all of its research—
• 48:37
  in pre-training, found a piece of mathematics that was very similar to this physics problem.
• 48:44
  It's reducing a bunch of integrals that are sequential and all of these processes, and it's trying to find some generalized formula.
• 48:51
  Perhaps in all of its reading of the world's mathematical literature, it saw and made a connection to someone else who had done something similar to find a general use case.
• 49:03
  pulled that bit from its latent space and then injected it here.
• 49:08
  I'm wondering if that's what happened.
• 49:10
  And that would make sense given the structure that we are currently told is what these large language models are framed as.
• 49:16
  Because your point, what you're saying is there is a pattern...
• 49:20
  That the model found in mathematics applied to another use case that for whatever reason we as humans have not yet made the connection to.
• 49:29
  And the model just made the connection.
• 49:30
  Maybe the physicists who are working on this are just not aware of like some fringe mathematical aspect.
• 49:36
  And so perhaps it made that pattern recognition, right?
• 49:38
  Right.
• 49:40
  Regardless, I think this does show the utility of these large language models to do frontier theoretical research now.
• 49:50
  Right.
• 49:51
  Right.
• 49:51
  One of the things, I know we've spent a long time on this story, but I think this is such an important.
• 49:55
  I think it was cool.
• 49:56
  This is really cool.
• 49:57
  And I think it's very important because this is like a zero or one phase phase.
• 50:03
  state change kind of issue.
• 50:05
  Either these models are not able to discover new science or,
• 50:11
  Or they are.
• 50:11
  Oh, yeah.
• 50:12
  And the difference between a world in which they are not and in which they are is an order of magnitude in terms of the implications.
• 50:21
  Yeah, yeah, yeah.
• 50:22
  So it's something that we should cover.
• 50:24
  And, you know, I'm not quite convinced that they can do really new science, right?
• 50:29
  But they can fill in gaps.
• 50:31
  Yes.
• 50:31
  And I think that's what they're doing here.
• 50:32
  Yes.
• 50:32
  Right?
• 50:33
  Yes.
• 50:33
  And even that is itself a very good thing.
• 50:36
  I want to remind people that ChatGPT, which is the first consumer version of the modern era of LLMs, came out three years ago.
• 50:45
  Yeah.
• 50:46
  And so I think keeping the time scale of progress in mind is very important.
• 50:51
  Yeah.
• 50:52
  You don't even graduate college in three years.
• 50:55
  Yeah.
• 50:55
  And we've gone from it can do absolutely jack.
• 50:58
  Yeah.
• 50:58
  Yeah.
• 50:59
  to getting to the fringes of some of these fundamental science concepts, I think it's very unwise to underestimate this.
• 51:08
  I get that there's a whole variety of social, economic, political implications that are all very important to talk about.
• 51:16
  That is 100% true.
• 51:18
  But saying this is a stochastic parrot that can't do anything of real value I think is a gross assumption.
• 51:26
  underestimation of what is currently happening yeah yeah and we and we need to like as a society start grappling with the true potential of this technology right and how we're going to get around it because if it doesn't get there great yeah but if it does get there we don't want to be getting there with our pants down yeah exactly
• 51:42
  Great first story.
• 51:43
  Again, potential watershed moment as this moves into scientific discovery.
• 51:48
  We've talked about several previous episodes where AI is already having other tangential impacts within breakthrough and frontier science research.
• 51:58
  A long story number one will be quicker for the stories.
• 52:01
  Story number two that we have here.
• 52:04
  This is out of the Universita D'Aglia Studi di Napoli, Federico II in Napoli, Italy, and the universities in Berlin in a paper in Scientific Reports.
• 52:14
  This is about feline behavior.
• 52:17
  A new study has shown that a cat's purr reveals much more about its
• 52:22
  individual identity than a meow.
• 52:25
  Yes.
• 52:26
  Cats meow.
• 52:27
  Yeah.
• 52:27
  So a meow is different from a purr.
• 52:30
  Okay.
• 52:30
  Is the point.
• 52:31
  Okay.
• 52:32
  You know, cats were domesticated probably like 10,000 years ago in the Middle East.
• 52:36
  And,
• 52:38
  What they found using 27 different cats and the vocalizations of all these different cats is the cats do not meow to each other.
• 52:45
  The meow of the cat has specifically evolved to interact with their humans.
• 52:53
  Oh, wow.
• 52:54
  Okay?
• 52:55
  That's the point of the story.
• 52:56
  The purr is actually something that is very unique to a cat.
• 53:00
  So one cat's purr is going to be just a...
• 53:04
  foundationally different sound signature than another cat's purr.
• 53:08
  But the meows have a very high variance.
• 53:10
  And the meows are basically used to interact with human beings.
• 53:14
  Cats only meow at each other when they're, like, in a fight, effectively.
• 53:18
  Otherwise, they're, like, communicating, like, hey, this is me using a purr.
• 53:23
  And what they did was they also...
• 53:26
  compared the vocalization of these 27 domesticated cats with the cats in the wild.
• 53:33
  Okay?
• 53:34
  Again, that's what they found.
• 53:35
  The meows are more variable, but the purrs are actually much more related to these wild cats.
• 53:40
  So the purr has kind of...
• 53:42
  been around since domestication and before domestication around 10 000 years ago in the middle east but the meow has actively evolved to communicate with human beings like i need more food i want to go outside i hate you i don't know what cats are like but i think that's what's going on in their head
• 53:58
  So we've identified that a cat's purr is like their fingerprint, the idea that every individual's fingerprint is different.
• 54:05
  But the meow is the way for them to interface with the meat suits that are human beings.
• 54:13
  Yes.
• 54:13
  Yes.
• 54:14
  that they get to boss around as they like to.
• 54:18
  Another follow-up to our Lion Roar story that we did last week in the rundown.
• 54:23
  A lot of great animal science stories.
• 54:25
  We're now going to talk about immunology and the debate of nature versus nurture as scientists discover how life experiences rewrite the immune system.
• 54:37
  It's like we have a molecular diary of our past.
• 54:41
  Yeah, this is an incredible story out of the Salk Institute in La Jolla, California, right outside of UC San Diego.
• 54:49
  It answers this question of variability, which is something that is very familiar to us having gone through the COVID-19 pandemic.
• 54:56
  Some people got completely wrecked by COVID-19 when they got it.
• 55:00
  Other people were fine.
• 55:02
  you know, even with the vaccine on both ends, right?
• 55:04
  Even if you control for those who are vaccinated and not, the variability between responses from one individual to the other is extremely different.
• 55:13
  Okay.
• 55:13
  And this,
• 55:15
  particular paper is trying to answer that question.
• 55:17
  Why are, why is stuff so different?
• 55:19
  They focus on something called the epigenome.
• 55:20
  The epigenome is how the DNA gets expressed.
• 55:23
  Everybody has the same DNA effectively, but like how each DNA strand, how the genes in that DNA get expressed is part of the epigenome.
• 55:33
  And that's a very dynamic thing that can change as we age and has, as we have life experiences.
• 55:40
  Um, um,
• 55:41
  The genetic patterns can be changed through two ways.
• 55:46
  Either it's a genetic thing, right?
• 55:48
  Like there's part of our epigenetics where the cells pick which genes to express and which genes to not based on the genetics themselves.
• 55:56
  And then there's another one where our life experiences actually change genetic expression, right?
• 56:01
  And those competing factors are very much...
• 56:05
  in play when it comes to immune cells.
• 56:09
  Because immune cells have to learn throughout the experience of an individual, right?
• 56:14
  I get some disease, my immune cell has to now remember what that disease looked like mechanistically so that we don't get that disease in the future, right?
• 56:22
  So immune cells are this really great lab to try and figure out what are these competing effects like, okay?
• 56:27
  What they focused on was something called differentially methylated regions.
• 56:32
  Basically, you've got your DNA strand,
• 56:35
  Your cells put tags on the DNA with these things called methyl groups, okay?
• 56:41
  And that sort of gives them a distinction between, hey, this is a methylated piece of DNA.
• 56:45
  This is not a methylated piece of DNA.
• 56:47
  So structurally, from a molecular level, they can tell two bits of DNA apart, okay?
• 56:53
  What they looked for is how these differentially methylated regions...
• 56:58
  were tied to things like genetics and life experiences.
• 57:02
  And what they found was there's actually two different types of markers that
• 57:07
  in our genetics, okay?
• 57:09
  The part that's more genetic is actually particularly in long-lived immune cells, like the T cells and the B cells.
• 57:17
  These are the ones that have memory of our past immune attacks and things like that.
• 57:21
  And all of these methylated regions are found in the stable regions of our genome.
• 57:26
  On the other hand, if you have things like, you know, the...
• 57:31
  Killer T cells.
• 57:32
  These are the ones that are rapid response.
• 57:35
  Those are the ones that have this methylated region in more flexible regulatory regions.
• 57:41
  Okay?
• 57:42
  So you've got these two different effects.
• 57:44
  And so what this kind of cell-specific database can tell you is new insight into how immune responses form.
• 57:51
  And also later on, we can do, you know...
• 57:54
  future treatments that are tailored to each person's unique biology, right?
• 57:59
  Because their particular immune system is going to respond in a different way to somebody else based on their past treatment.
• 58:05
  So if we can read the epigenome, we can then forecast how is someone going to respond to a certain treatment?
• 58:11
  How is their immune system going to respond?
• 58:13
  That's the idea.
• 58:14
  This is very fascinating.
• 58:15
  The concept here being there are, you know,
• 58:20
  Our genome has both these like genetically inherited from long, the long history of evolution that then formulate into these long lived T cells and B cells, which kind of makes sense.
• 58:31
  It's like, oh, these are things that for millennia, for generations have been helpful.
• 58:37
  So we can kind of memorialize it in our long term memory.
• 58:40
  And then there's this sort of like epigenetic layer, which is more about the lived experience of the individual or its most recent predecessors impact the killer T cells and these sort of aspects that might need to be,
• 58:55
  more reactive to real-time changes in the environment.
• 59:00
  And so the point... And we're able to now, in the molecules, see the difference.
• 59:06
  How they're tagged between coming from this long history versus a short history.
• 59:11
  And again...
• 59:13
  I mean, this goes to the importance of like genetic diversity and clinical trials as an example, which is a space that my mom works in because you will have different responses based not only on, you know,
• 59:27
  someone's African versus African American.
• 59:30
  But even that, even if they're both black, the epigenetic layer of the experiences are going to be fundamentally different between those who are part of the African, African slave trade versus Africans who are not just as a use case example.
• 59:43
  Yeah.
• 59:44
  And so these kinds of details are important because it,
• 59:48
  especially if someone is black, like the medical system and us being able to get treatment is a very complex issue.
• 59:53
  Yeah.
• 59:54
  Um, because just like photos and film was initially created for only white people.
• 59:59
  Right.
• 1:00:00
  I didn't know that there's fundamental things about the way that film works
• 1:00:03
  Like actual film works because of light.
• 1:00:07
  If you think about how light bounces off of white people's skin versus dark people's skin.
• 1:00:11
  So until we got later into digital, it was never really made for black people.
• 1:00:16
  Medicine kind of has a similar issue where a lot of the genetic profiles have been or the way in which we've operated is for black people.
• 1:00:23
  a particular group set.
• 1:00:24
  And I think as we continue to do this type of research, it will allow us to expand the efficacy of medicine across a larger swath of genetic profiles.
• 1:00:32
  So fascinating story.
• 1:00:34
  Again, there on talking about immunology, which has been an issue.
• 1:00:38
  We've talked a subject.
• 1:00:39
  We've talked about quite a bit on the pod.
• 1:00:41
  We're going to move on to origin of life for our last rundown story.
• 1:00:46
  Yeah.
• 1:00:47
  second to last sorry second to last i want to do i want to do five this time okay okay okay uh so for our second to last which is going to be this um scientists finding genes that existed before life on earth what does that even what does that mean yeah that doesn't that doesn't mean anything to me right right genes before life
• 1:01:11
  What are you talking about is exactly what I thought when I was reading this.
• 1:01:15
  Because it's like life is genes.
• 1:01:18
  Right.
• 1:01:18
  So how does it before?
• 1:01:19
  Yeah, yeah, yeah.
• 1:01:20
  So here's what's going on.
• 1:01:22
  There might be some genes that are in nearly every single organism today that were duplicated before life shared a common ancestor.
• 1:01:32
  Okay.
• 1:01:32
  Okay?
• 1:01:32
  Because let's think about all of life on Earth, right?
• 1:01:34
  It came from a single common ancestor.
• 1:01:36
  That's what we think.
• 1:01:37
  Yes.
• 1:01:37
  Because we all have DNA and there's all these different markers, right?
• 1:01:41
  Yes.
• 1:01:41
  Now, that single common ancestor...
• 1:01:45
  There must have been stuff before the common ancestor that were building blocks to make that last common ancestor, right?
• 1:01:52
  And these guys are saying they have a way now to find genes that were there before that last common ancestor was even alive.
• 1:01:58
  Fascinating.
• 1:01:59
  Like the building blocks of that last common ancestor.
• 1:02:02
  It's pretty insane.
• 1:02:03
  So when we think about the last common ancestor, right, these are cells that already had membranes.
• 1:02:09
  They've got an inside and an outside, and they've got DNA to store genetic information, right, to store some information for self-replication.
• 1:02:17
  Now, these are essential traits, and they've already been established.
• 1:02:21
  So what happened before that, and how can we tell?
• 1:02:24
  Well, they look for something called paralogs, okay?
• 1:02:27
  Paralogs are groups of related genes that appear multiple times within a single genome, right?
• 1:02:34
  A good example is for us hemoglobin.
• 1:02:37
  We have eight copies of hemoglobin in our DNA.
• 1:02:39
  Okay?
• 1:02:40
  And that hemoglobin came about about 800 million years ago.
• 1:02:43
  That's the thing.
• 1:02:44
  That's the protein that carries oxygen in our blood.
• 1:02:46
  We've got eight different copies.
• 1:02:48
  All of us have eight different copies.
• 1:02:50
  A lot of mammals have eight different copies.
• 1:02:53
  The fact that all of us have that means that hemoglobin is a really old, old protein.
• 1:02:59
  Okay?
• 1:03:00
  So, yeah.
• 1:03:01
  The idea is if there's repeated copying and extra versions of this gene, then the copying must have happened way early because all of us have that.
• 1:03:10
  Right.
• 1:03:11
  So now let's look for universal paralogs, which are genes that are repeated, like they have this repeated tendency and these copies across everyone.
• 1:03:23
  Bacteria, us, bacteria.
• 1:03:25
  Fungi, literally everyone.
• 1:03:27
  What they're looking for is gene families that appear at least two times in at least two copies in the genomes across life.
• 1:03:35
  And so to be clear, you're saying hemoglobin, there's eight copies.
• 1:03:38
  Eight copies, but it only happened 800 million years ago.
• 1:03:41
  Right.
• 1:03:41
  Like bacteria don't have hemoglobin.
• 1:03:43
  Right.
• 1:03:44
  And so we're looking— Plants also.
• 1:03:45
  Plants also.
• 1:03:46
  We're looking for things that have multiple copies, like hemoglobin, but hemoglobin we only see in humans and mammals.
• 1:03:53
  And so we're saying, what else has these n number of multiple copies, but in other...
• 1:03:57
  Everywhere.
• 1:03:58
  But literally everywhere.
• 1:04:00
  That's the other thing.
• 1:04:01
  It's got to be literally everywhere, because then you can make the argument, it happened way back, even before the universal ancestor.
• 1:04:09
  Right?
• 1:04:09
  Yes.
• 1:04:10
  And they found a few.
• 1:04:11
  Okay.
• 1:04:11
  And all of these genes have to do...
• 1:04:15
  with two things either building proteins or managing the membrane how to get in and out of the membrane and that kind of makes sense we've talked about this you've brought this up several times about the membrane and it's like how did that happen how did that happen it turns out these cell membranes and managing transport across the cell membrane is something that might have happened even before our last common ancestor mm-hmm
• 1:04:40
  Just in like random molecular processes and like proteins at a molecular level, not even a cell trying to self-replicate, but like molecules trying to get across.
• 1:04:51
  And if a molecule got across, it had an advantage to like do more of itself.
• 1:04:56
  Right.
• 1:04:57
  And I think that's really fascinating.
• 1:04:59
  What they could also do is reconstruct the protein and produce this ancient ancestral protein that could then go across molecules.
• 1:05:09
  It's molecular Jurassic Park.
• 1:05:11
  Yes.
• 1:05:11
  Yes.
• 1:05:12
  And what's crazy is that molecule that's billions of years old now, it can traverse the same membranes that we have.
• 1:05:19
  So that old use case, it can still do it.
• 1:05:28
  It's like we've found, we've like reburned a CD copy of Now That's What I Call Music Volume 4.
• 1:05:34
  And we're on Now That's What I Call Music Volume 200.
• 1:05:37
  Yeah.
• 1:05:38
  But it's like we found the old.
• 1:05:41
  Oh, yeah.
• 1:05:41
  And it got the same intro.
• 1:05:43
  Yeah, yeah, yeah.
• 1:05:44
  And it plays in the same player.
• 1:05:46
  That's kind of a rough analogy.
• 1:05:48
  But that's actually.
• 1:05:49
  I thought that was really crazy.
• 1:05:51
  This is very cool.
• 1:05:52
  Because, again, what this means is we can continue to log.
• 1:05:54
  Right.
• 1:05:55
  And continue to look for these paralogues across everything that would allow us to identify these ancestors before the currently accepted universal common ancestor.
• 1:06:06
  Yeah.
• 1:06:06
  In terms of traversing the path of evolution from like single cell to multicellular organisms.
• 1:06:13
  Yeah.
• 1:06:13
  Yeah.
• 1:06:14
  Unbelievable.
• 1:06:15
  This is out of Oberlin College, MIT.
• 1:06:17
  I misdated MIT earlier.
• 1:06:19
  This was the MIT story.
• 1:06:20
  And our Midwest companions, UW-Madison.
• 1:06:24
  Of course.
• 1:06:24
  In Cell Genomics.
• 1:06:26
  We're going to move to our last rundown story, which is about astrophysics and exoplanets, which astronomers have discovered a solar system that should not be possible.
• 1:06:40
  Four planets orbiting a red dwarf 116 light years away from Earth.
• 1:06:46
  from earth yeah so what is happening here and why should this solar system not be possible right so we used to think that every solar system should kind of look like ours for very good reasons the way it looks like ours is the inner planets are rocky the outer planets are gas giants right that makes sense because when
• 1:07:04
  solar systems form the star is going to be very volatile so it's going to the radiation from that proto star is going to basically shoo away all of these lighter elements like helium and hydrogen and ice and things like that the rocky elements like iron and other nickel that's going to stick around the star and create rocky planets and then the outside is going to be gas giants that's how our solar system is the physics makes sense yes
• 1:07:29
  They have found an inside-out solar system.
• 1:07:32
  This solar system goes rocky, gas giant, gas giant, rocky.
• 1:07:36
  Oh.
• 1:07:37
  And that really doesn't make sense.
• 1:07:38
  Yeah, that doesn't make any sense.
• 1:07:39
  Okay?
• 1:07:39
  Just from the physics argument that I gave you.
• 1:07:41
  Yes.
• 1:07:42
  Right?
• 1:07:42
  And they're trying to figure out what the hell is going on.
• 1:07:45
  Right.
• 1:07:45
  Okay?
• 1:07:46
  This rocky planet that's on the outside is 1.7 times the mass of the Earth.
• 1:07:52
  It's what astronomers would call a super-Earth.
• 1:07:54
  They tried to figure out, like...
• 1:07:56
  Ways that it could have formed.
• 1:07:57
  Like, suppose maybe it was a gas giant, but then the atmosphere got ripped apart, right?
• 1:08:02
  And so now we're seeing it as a rocky planet.
• 1:08:04
  They did a bunch of simulations, didn't work.
• 1:08:07
  So what they think is happening is that this outer rocky planet actually just formed later in the history.
• 1:08:14
  Because suppose the inner planets formed early.
• 1:08:16
  So you got the rocky planet on the inside.
• 1:08:18
  You got two gas giants.
• 1:08:20
  Those gas giants have hoarded...
• 1:08:22
  All of the gas.
• 1:08:24
  And now when this outside one formed, there was no gas around to really like nucleate on the outside.
• 1:08:31
  Right.
• 1:08:31
  And which is why it's now rocky.
• 1:08:33
  So the idea, the hypothesis here is that there was the difference is because of a time delay, a formation time.
• 1:08:40
  Yes, a formation.
• 1:08:41
  They started forming at different times.
• 1:08:43
  And by the time that fourth planet started forming after the rocky gas giant, gas giant.
• 1:08:49
  all of the materials to make it a gas giant were not available.
• 1:08:52
  And so it became a rocky, exactly.
• 1:08:54
  Rocky planet.
• 1:08:55
  Yeah.
• 1:08:56
  This is a good one out of university of Warwick.
• 1:08:57
  Yeah.
• 1:08:58
  University of Michigan, Ann Arbor, plus several others in science.
• 1:09:02
  Yeah.
• 1:09:03
  We went long on the rundown today, but we love this stuff.
• 1:09:06
  So sometimes we get, sometimes we get, yeah, sometimes we get a little carried away.
• 1:09:10
  Uh, we'll try to keep it tighter, but we know you all like the details.
• 1:09:14
  And so we will always go with the details, but we are now moving back to our theme of the week, which is the winter Olympics.
• 1:09:24
  We talked about why is ice slippery first?
• 1:09:26
  Um,
• 1:09:27
  We are now moving on to our second main story, which is going to be a neuroscience story.
• 1:09:34
  And this story is about choking under pressure.
• 1:09:38
  Not literal pressure like in our first story.
• 1:09:40
  No.
• 1:09:41
  And not literal choking.
• 1:09:42
  And not literal choking.
• 1:09:43
  But the figure of speech, what happens in sports all the time.
• 1:09:48
  There's a sequence.
• 1:09:49
  You've done it a thousand times.
• 1:09:51
  You've got a down pat.
• 1:09:52
  You train for four years.
• 1:09:56
  And then when it comes down to the moment, you just can't lock in.
• 1:09:59
  Yeah.
• 1:10:00
  And there's actually really good neuroscience basis for these.
• 1:10:04
  Yes.
• 1:10:05
  And I wanted to talk about this because of the recent events that have happened in Cortina and Milan.
• 1:10:14
  This Winter Olympics, there was a man named Ilya Malinan.
• 1:10:18
  He was purported as the quad god.
• 1:10:21
  This is a...
• 1:10:23
  Scientific American article talking about the physics of how he can pull off these impossible jumps.
• 1:10:28
  He's a two-time reigning world champion.
• 1:10:31
  Pre-event odds, before he went for the free skate portion, the individual free skate portion, negative 10,000, which is implying a probability of 99% at gold.
• 1:10:42
  That tells me two things.
• 1:10:43
  One, this guy is incredibly good, and everyone is thinking he's going to get gold.
• 1:10:47
  And two, the bookkeepers have really terrible models.
• 1:10:51
  Okay?
• 1:10:53
  How do you not include the variability of the Olympics?
• 1:10:59
  The Olympics is like no other event in sports, especially for these types of sports.
• 1:11:06
  There's the world championships.
• 1:11:07
  Okay, he's a two-time world champion.
• 1:11:09
  The Olympics is different.
• 1:11:10
  I don't watch the world championships.
• 1:11:12
  Okay?
• 1:11:13
  People who are into ice skating will watch the world championships.
• 1:11:16
  Yep.
• 1:11:17
  But everybody is watching the Olympics, right?
• 1:11:20
  I think I just saw this chart the other day where the Winter Olympics is like the number five most watched sporting event.
• 1:11:26
  Only behind the Cricket World Cup, the Summer Olympics, and the World Cup.
• 1:11:32
  And one other event that's escaping me.
• 1:11:34
  But it is, and it's in the billion, like, it's a massive, it's a massive thing.
• 1:11:37
  It's like 10x the Super Bowl.
• 1:11:39
  Like, it's very, very watched.
• 1:11:41
  Yes, exactly.
• 1:11:43
  He was the quad god.
• 1:11:45
  He was going to get gold, 99% chance.
• 1:11:48
  He gets on the ice, and he completely chokes, okay?
• 1:11:53
  He was going to make this quad axle, which was going to be the first time anyone's ever done it in competition.
• 1:11:58
  Everyone was looking forward to it.
• 1:12:00
  This was crazy.
• 1:12:00
  I was watching it live, dude.
• 1:12:02
  He goes for his first jump.
• 1:12:04
  which is the quad flip does it very well.
• 1:12:08
  The commentators say it's an effortless start.
• 1:12:10
  The next one is going to be this quad axle.
• 1:12:12
  And the commentators say, if he's going to go for the quad axle, it's going to be this next one.
• 1:12:16
  He think, I think he goes for it in the middle of the jump.
• 1:12:20
  He abandons it.
• 1:12:21
  He only does one turn.
• 1:12:22
  Mm-hmm.
• 1:12:23
  And from there, it just unravels.
• 1:12:26
  Like mentally, he just like completely collapsed.
• 1:12:29
  Yep.
• 1:12:29
  Yep.
• 1:12:29
  Okay.
• 1:12:30
  Yep.
• 1:12:30
  And I just thought that was so fascinating because this is a guy who's done that routine probably thousands of times.
• 1:12:37
  He's prepared for it.
• 1:12:38
  Right.
• 1:12:38
  And the Olympic ice is just different.
• 1:12:41
  Right.
• 1:12:42
  It's still water.
• 1:12:43
  It's still H2O.
• 1:12:44
  But there's something that is different about that moment.
• 1:12:47
  Yes.
• 1:12:48
  that just completely, the house of cards just completely collapsed.
• 1:12:52
  And it's not a physics issue.
• 1:12:55
  No.
• 1:12:56
  Like we talked about earlier.
• 1:12:57
  No.
• 1:12:57
  It is a, it's a neuroscience issue.
• 1:13:00
  It's a neuroscience issue.
• 1:13:01
  And afterwards he said, it's the Olympics.
• 1:13:03
  And I think people only realize the pressure and the nerves that actually happen when you're on the ice.
• 1:13:08
  Right.
• 1:13:08
  And it was something that just overwhelmed me.
• 1:13:10
  And I felt like I had no control.
• 1:13:12
  This is a quote from Ilya.
• 1:13:14
  And I.
• 1:13:16
  From there, I wanted to talk about sort of what that means.
• 1:13:19
  What is going on in an athlete's brain when something like that happens?
• 1:13:24
  And before we jump in, I just want to be clear here, especially as someone who is a former athlete at a high level, this is not meant to be disparaging to Ilya.
• 1:13:34
  No.
• 1:13:34
  Sports is extremely difficult.
• 1:13:36
  Yeah.
• 1:13:37
  And the stakes are very high.
• 1:13:39
  There's a lot of money involved.
• 1:13:42
  It is just the topic that is...
• 1:13:45
  It's there for us to discuss.
• 1:13:47
  And I think it's interesting.
• 1:13:49
  There's an interesting science angle here.
• 1:13:51
  And so we're trying to understand the neuroscience of this.
• 1:13:55
  This is not about, oh, my God, I can't believe he failed or didn't fail.
• 1:13:59
  No, yeah, no.
• 1:14:00
  We're going to be talking smack, but it's going to be later, and it's not going to be about Ilya.
• 1:14:04
  Yeah.
• 1:14:04
  don't you worry this is an olympics podcast but you know ilia is very young and he's gonna be fine so he's gonna be back in the next olympics and you know one of the things that we're gonna uncover about the neuroscience of choking is the more experience you have the better these pathways get to prevent you from doing that that's why lebron is still like incredible right at like just shooting yes even under pressure right yes
• 1:14:28
  So let's get into some of the theoretical frameworks before we dive into the neuroscience.
• 1:14:33
  Okay.
• 1:14:33
  There's several theories behind why this happens.
• 1:14:36
  One is called the distraction theory.
• 1:14:38
  What it says is basically working memory, which is like your memory of like events that are happening now and what you need to do in a while.
• 1:14:44
  That has a finite capacity and it's flooded with...
• 1:14:47
  task irrelevant stressors like there's fear about failing there's the crowd there's the cameras on you um and so you get this sort of loss of goal-directed control of attention okay that's the distraction theory then there's um the explicit monitoring theory which is actually the opposite which says that okay athletes actually consciously control some automated practice skill and if you overthink
• 1:15:16
  That's when it hurts you.
• 1:15:17
  OK, if you if you start consciously thinking about what you're trying to do, that's actually what's bad.
• 1:15:23
  And then there's another one, which is this over arousal or over over motivation theory.
• 1:15:29
  And that's the idea that there's actually an optimal amount of arousal when it comes to doing difficult and easy tasks.
• 1:15:35
  And if you want to do difficult tasks, you should be aroused less.
• 1:15:41
  than if you're doing easy tasks, which is kind of incredible to think about.
• 1:15:45
  When you're doing an easy task, you want to be more in it.
• 1:15:48
  But when you're doing a really hard task, it's kind of better to...
• 1:15:53
  To be a little bit out of it.
• 1:15:55
  Yeah, yeah, yeah.
• 1:15:56
  It's kind of probably what people describe as that flow state.
• 1:15:59
  Exactly.
• 1:16:00
  It's the flow state versus the intense attentional control that you're putting on something, right?
• 1:16:06
  Right.
• 1:16:06
  And so in the context of these theories, I wanted to talk about the brain regions.
• 1:16:11
  The main brain regions we're going to talk about is the prefrontal cortex, the motor cortex, and the amygdala.
• 1:16:23
  Right up here.
• 1:16:24
  This is something that is very big in humans.
• 1:16:26
  It's also there in primates.
• 1:16:28
  It's not so much there in lower order mammals.
• 1:16:30
  Okay.
• 1:16:31
  This is where you have executive control, working memory.
• 1:16:35
  You delegate micro mechanics to the other subcortical structures.
• 1:16:38
  So this is kind of like our control center in some sense.
• 1:16:42
  I mean, I remember there was this guy.
• 1:16:44
  He's still there.
• 1:16:45
  Professor Bujaki at NYU.
• 1:16:47
  He hates when people say this.
• 1:16:49
  He used to call it the prefrontal cortex, the trash can.
• 1:16:52
  of the brain because whenever someone i was like oh this this is probably the prefrontal cortex but the reason is because it's such a highly evolved part of our brain that it seems that a lot of these higher functions are happening here right if that makes sense yep so there was an experiment that was done um about the prefrontal cortex and how the prefrontal cortex functionally disconnects
• 1:17:17
  with the other court with the other courtesies during high stress it was a pretty in pretty incredible paper um called out of control diminished prefrontal activity coincides with impaired motor performance here's what it was doing okay they use fmri data and they have this bimanual visual motor tasks where um people like
• 1:17:38
  human beings are playing snake.
• 1:17:40
  Okay.
• 1:17:41
  And every once in a while, whenever they get the snake game, right, they get like 10 bucks.
• 1:17:45
  Every once in a while, they'll get a jackpot of like 50 bucks.
• 1:17:49
  And they want to, they want to see how the, how the participants do.
• 1:17:55
  They actually performed less than,
• 1:17:58
  On the jackpot tasks.
• 1:18:00
  It's the same game.
• 1:18:01
  Right.
• 1:18:01
  But when there's this pressure of, oh, I could just win like 50 bucks instead of 10, the performance dropped.
• 1:18:07
  And then when they look at the prefrontal cortex and the motor cortex connectivity during these jackpot trials, it was inversely related to choking susceptibility.
• 1:18:17
  Meaning the less it was connected, the more you were going to choke.
• 1:18:21
  Mm-hmm.
• 1:18:22
  So there's literally a connectivity issue between your prefrontal cortex and your motor cortex.
• 1:18:27
  The motor cortex is the thing that's actually doing the gameplay.
• 1:18:30
  And when that connection goes down, you're likely to choke more.
• 1:18:35
  Now, the question is, why is this connection going down?
• 1:18:38
  During these high-stress situations?
• 1:18:41
  Scenarios.
• 1:18:43
  So the initial intuitive thought process is, okay, the high stress scenario is what is directly correlated to the decreased connection between the prefrontal cortex and the motor cortex, theoretically.
• 1:18:52
  Theoretically, right?
• 1:18:53
  But in neuroscience, we always want something more mechanistic.
• 1:18:56
  Right.
• 1:18:57
  Right.
• 1:18:57
  Right.
• 1:18:57
  Right.
• 1:18:58
  At least I do.
• 1:18:59
  And so now let's go to this next paper.
• 1:19:02
  This was from Smulder and other authors in 2024.
• 1:19:06
  This was in Neuron, A Neural Basis of Choking Under Pressure.
• 1:19:10
  And what a title for a paper, can I just say?
• 1:19:14
  um here's what they did they use you they um they used 96 channel utah arrays in reas macaques so macaques have a prefrontal cortex kind of like we do um this was in the motor cortex and in another brain region called pmd and they had the similar um behavioral paradigm of choking under pressure basically there were these jackpot
• 1:19:38
  scenarios where the monkey knew that if it succeeded it was going to get a really high reward and again the same thing happens you get this increase in performance but when you get an incredibly high reward anticipation your performance decreases
• 1:19:54
  So what we're looking at here is three charts of monkey E, P, and R with six sessions, nine sessions, and 12 sessions.
• 1:20:01
  And on the Y axis is success rate.
• 1:20:05
  And on the X axis is sort of jackpot size, going from small going to large.
• 1:20:10
  And you can basically see this like top of a trapezoid where we have at small to medium...
• 1:20:16
  There's an increase.
• 1:20:17
  There's a huge increase.
• 1:20:18
  And then medium to large, it kind of maintains that higher level success rate.
• 1:20:22
  And you see a very drastic decrease in performance when you move from large to jackpot across all these session lengths and all of these different subjects.
• 1:20:31
  Exactly.
• 1:20:31
  And so regardless of number of repeats, et cetera, et cetera, there's a clear decrease in performance efficacy when the jackpot gets sufficiently high.
• 1:20:41
  Yeah, and the jackpot is way higher than the large is the other big thing.
• 1:20:44
  It's like small, medium, large are like successively higher, but jackpot is like an actual jackpot.
• 1:20:50
  And that's kind of what the Olympics are in some sense, right?
• 1:20:53
  Like the stress is so high and that gold medal that's like sitting there could be yours.
• 1:21:00
  It's incredible, right?
• 1:21:02
  And so how do we manage that?
• 1:21:03
  And so, okay.
• 1:21:05
  We saw the behavioral paradigm, which is the same as we have in humans.
• 1:21:09
  The trick is now we want to find a neural paradigm, something in the neurons themselves that correlates to that behavioral paradigm, because then we found a neural signature for what is happening.
• 1:21:20
  Yes.
• 1:21:20
  Right?
• 1:21:21
  They use something called dimensionality reduction.
• 1:21:23
  This is something that we've been through in neuroscience.
• 1:21:26
  The idea is your brain lives in a very high dimensional subspace, right?
• 1:21:30
  If you're...
• 1:21:31
  If you're listening to 10,000 neurons, that's 10,000 different dimensions in your mathematical subspace of where your brain is.
• 1:21:42
  Your brain doesn't actually live in that subspace.
• 1:21:44
  That's just how we're describing it.
• 1:21:46
  There might be lower dimensions.
• 1:21:48
  that we can move along that tell us something about, oh, how much does the brain care about reward in this scenario?
• 1:21:55
  How much does the brain care about the target, getting to the target and getting that reward in this scenario, right?
• 1:22:01
  We can decompose the brain activity into independent metrics and see what happens.
• 1:22:07
  So what they found was in the reward axis...
• 1:22:10
  there was an axis that basically correlated with reward.
• 1:22:14
  So when it was jackpot, it was really high.
• 1:22:16
  There were a bunch of neurons that were firing when it was a jackpot.
• 1:22:19
  And that was like those neurons care about.
• 1:22:21
  It's like kind of telling the rest of the brain, hey, there's going to be a jackpot.
• 1:22:24
  Like we got to lock in.
• 1:22:25
  Okay.
• 1:22:25
  Then there's another one, which is the target preparation axis.
• 1:22:29
  Okay.
• 1:22:29
  This is the dimension within the target plane that is critical to the reach, which is part of the task.
• 1:22:38
  Like how far am I going to reach?
• 1:22:40
  for the macaques to get this reward, right?
• 1:22:43
  And that target axis has the same behavior as what we saw in behavior, meaning for low reward, I'm not that high on the target axis.
• 1:22:53
  But as I go to jackpot reward, again, I come back, okay?
• 1:22:57
  So these neurons, these specific neurons that care about the target axis, the geometry of the brain has changed to where this target axis is actually the thing that is causing that choking under pressure.
• 1:23:10
  The point being there is now a neurological signature that maps onto the behavioral experimental data around this idea of there being sort of this diminishing...
• 1:23:25
  as the jackpot gets sufficiently high in task completion.
• 1:23:29
  Exactly, exactly.
• 1:23:30
  And now we can actually think about a mechanism, right?
• 1:23:33
  And what they came up with is something called the expansion, then collapse mechanism.
• 1:23:37
  Here's the idea, okay?
• 1:23:39
  For small reward...
• 1:23:41
  I'm influenced by reward a lot.
• 1:23:44
  Okay.
• 1:23:44
  But as I move in this trajectory, there is some optimal space that I want my brain to be in.
• 1:23:50
  Okay.
• 1:23:51
  There's like an optimal location in all of my neurons, in all of my neural space that I want to be in.
• 1:23:56
  Okay.
• 1:23:57
  But when I get this massive motivational signal,
• 1:24:00
  from the jackpot trials what that's going to drew do is put my system into overdrive and it's going to push my system away from this optimal setting yep right yes so where now my neural states abruptly like collapse on top of each other yep and i i can't actually discern between like what to do when i have the task in front of me
• 1:24:26
  Because all of my neurons are in overdrive.
• 1:24:28
  I've got stress neurons coming in.
• 1:24:30
  I've got the future planning neurons coming in.
• 1:24:33
  What am I going to do when I get... What's going to happen in the future?
• 1:24:37
  All of the past mistakes that I've done, that stuff's coming in.
• 1:24:40
  The memory's coming in.
• 1:24:41
  All of the stress is trying to get into this...
• 1:24:44
  Flight or fight response.
• 1:24:45
  And it's putting your brain into overdrive.
• 1:24:47
  And then you get a physical failure because now you can no longer actually do the task as well as you once could.
• 1:24:54
  Just I want to bring up this chart again, which is this neural bias.
• 1:24:58
  Why do animals choke?
• 1:24:59
  And we're looking at an X, Y, and Z axis.
• 1:25:02
  Neuron one is, let's say, X, neuron two is Y, neuron N is Z.
• 1:25:08
  And we sort of see the influence of the reward, small, large, and jackpot, like, you know, in a 2D image meant to represent its 3D positioning on this X, Y, and Z axis.
• 1:25:18
  And the point is sort of saying, I'm saying, the large, the little circle with the dots around it.
• 1:25:25
  That's the optimal.
• 1:25:26
  That's your target region where you want to be at.
• 1:25:28
  That's where you want to be because that's where all the balance of all of the inputs and everything is just right for you to execute.
• 1:25:36
  Yeah.
• 1:25:36
  And what we're sort of saying here is the introduction of a jackpot reward moves you beyond that optimal positioning.
• 1:25:45
  Because there's so much other stimulus coming in.
• 1:25:47
  Coming in and going on, which sort of it makes sense because if you just think about athletes, right?
• 1:25:51
  Some of the athletes in every sport who we mention as being the GOAT, when you look at them playing, almost if you talk about people like Messi, Wayne Gretzky, Usain Bolt, etc.,
• 1:26:02
  visually when you're looking at them with the human eye test, it almost appears as if they are totally unperturbed, disturbed, unfazed by anything going on around them.
• 1:26:14
  And it's like they're floating.
• 1:26:15
  People always describe Michael Phelps.
• 1:26:17
  They're like, they are operating at a plane that is different than anybody else.
• 1:26:22
  Yeah, because they've somehow disconnected that input drive that's coming in.
• 1:26:26
  And I think that comes a lot with experience.
• 1:26:29
  Yes.
• 1:26:29
  Right.
• 1:26:29
  Which is fair.
• 1:26:30
  Which is fair.
• 1:26:31
  Like when it's your first time at the Olympics, like Ilya's,
• 1:26:35
  Mate, like, I don't know what that's like.
• 1:26:37
  That's got to be insane.
• 1:26:38
  You've been working.
• 1:26:39
  You've been dreaming about this moment for the past, like, God knows how many years.
• 1:26:43
  He's probably started, you know, ice skating very, very young.
• 1:26:47
  Like, since he could walk, probably, or something.
• 1:26:50
  You know?
• 1:26:50
  And then you think about the gold medal opportunity.
• 1:26:53
  Everyone is there.
• 1:26:54
  The NBC is calling you the quad god.
• 1:26:56
  Right?
• 1:26:57
  Like, there's so much hype.
• 1:26:58
  99% on Polymarket.
• 1:26:59
  Yes.
• 1:26:59
  Yeah.
• 1:27:02
  Like that's got to be so much pressure.
• 1:27:03
  I would love to see the same group do research on Lamine Yamal because this kid is 19 now and he's dominating grown men.
• 1:27:14
  Yeah.
• 1:27:14
  Yeah.
• 1:27:15
  In every level of competition in soccer, football, globally.
• 1:27:20
  And it doesn't make any sense.
• 1:27:22
  Yeah.
• 1:27:22
  And there has to be something upstairs that is blocking everything.
• 1:27:27
  going to the jackpot yeah space in the in the gradient descent how is he with like free kicks and penalties he's unbelievable because that's that's sort of where i would expect choking to happen right because that's an individual kind of thing this kid with team sports i can i don't you know fair i wonder if we can take these fair no no that's fair because it's a little bit different yeah it's a little bit tiger woods
• 1:27:48
  Because also Ilya in the team event, although that was a free skate that he did individual and he crushed it.
• 1:27:54
  So like over 200.
• 1:27:55
  No, that's a good point.
• 1:27:56
  But then when it came to his individual, which was only a few days later, he followed the same procedure.
• 1:28:01
  And like, that's a very good point.
• 1:28:03
  You know, I only bring it up because he's young.
• 1:28:05
  Yeah.
• 1:28:06
  And like as a counter...
• 1:28:08
  a counter thesis to the experience point, but you, you correct team sports and individual sports, very different context.
• 1:28:15
  Yeah, yeah, yeah, exactly.
• 1:28:17
  And, um, I just want to end with, you know, Ilya.
• 1:28:21
  Now he's, he, I mean, he's like beside himself and one, one of the,
• 1:28:27
  articles was just about, you know, it's kind of the point that winning Olympic gold isn't supposed to be easy.
• 1:28:34
  And I don't know who thought it was a 99 percent because with sports at the highest level, anything can happen.
• 1:28:41
  You know, and one of the things he added was I just felt like all the traumatic moments of my life really just started flooding in my head.
• 1:28:49
  And there was just too many negative thoughts that just flooded in there.
• 1:28:52
  And I just did not handle it.
• 1:28:53
  This is a quote from him afterwards, which is I mean, it's incredible that such a young man has the vocabulary and mental fortitude after such a traumatic event to even speak to the press.
• 1:29:06
  I think I would like give the middle finger and just like leave.
• 1:29:09
  But, you know, I think he's going to be back.
• 1:29:11
  I have seen him do the...
• 1:29:13
  I've seen videos of him practicing that quad axle thing, and it's incredible.
• 1:29:19
  And it's also when he's doing it in the practice, it looks effortless.
• 1:29:22
  It looks effortless.
• 1:29:23
  Which is crazy.
• 1:29:24
  Yeah.
• 1:29:25
  Because it's so ridiculous.
• 1:29:26
  Yeah, it's so ridiculous how many turns.
• 1:29:27
  He's like four and a half turns landing backwards.
• 1:29:30
  It's crazy.
• 1:29:31
  On that thin millimeter blade with the 100 nanometers of slimy water.
• 1:29:35
  Yeah, maintaining balance.
• 1:29:36
  Like, it's crazy.
• 1:29:37
  Yeah.
• 1:29:38
  So he'll probably be back.
• 1:29:39
  But I just took that opportunity as a quick deep dive into neuroscience because I always love that stuff.
• 1:29:45
  No, it's fascinating.
• 1:29:46
  Again, I think, you know, sports, as we move into the era of...
• 1:29:50
  infinite social media and infinite content and TV and movies are sort of losing their luster of all of us watching them for the first time at the same time with binging and all this stuff.
• 1:30:01
  Sports as a medium and the importance of sports as being a connective tissue that's real time.
• 1:30:08
  It's the only appointment that...
• 1:30:11
  media left.
• 1:30:12
  Yeah.
• 1:30:13
  You got to, you got to be there when it happens both in person or digital.
• 1:30:17
  Yep.
• 1:30:18
  And it is, you know, we see these kinds of situations.
• 1:30:22
  This, this was not meant to focus on it.
• 1:30:24
  It happens in all sports all of the time.
• 1:30:26
  Why is it that Zion Williamson in the NBA was one of the most lauded NBA draft recruits that everyone was super excited for him to be the next big thing in the NBA.
• 1:30:36
  Yeah.
• 1:30:37
  Totally flopped.
• 1:30:38
  But now Cooper flag, 19 year old comes in same kind of pathway.
• 1:30:42
  People like he's absolutely dominant.
• 1:30:45
  And it's, it's very interesting to see this play out across different sports and across different contexts.
• 1:30:51
  Yep.
• 1:30:52
  We are going to wrap up the day with our sort of final main section, which is a little bit of a potpourri.
• 1:30:58
  Yeah.
• 1:30:58
  It's a hodgepodge.
• 1:30:59
  It's a little miscellaneous.
• 1:31:00
  There will be a climate change.
• 1:31:02
  science aspect to this but we're going to cover a lot of different areas i think you did have some other random thoughts about the olympics given our olympics uh focus here so why don't we start with some of this potpourri that you got for us
• 1:31:18
  Yeah, so the first thing I want to talk about is climate change.
• 1:31:21
  It is real, and it's going to hurt Winter Olympics a lot more than Summer Olympics, as you can imagine, right?
• 1:31:26
  With the earth warming, the first Olympic Winter Games were held in Shamanics, France in 1924.
• 1:31:33
  All 16 events took place outdoors.
• 1:31:36
  Look at these fine gentlemen in suits.
• 1:31:38
  Like, they're ready to, you know, I think this is the skate race, the 1,000-meter skating race.
• 1:31:45
  The athletes relied on natural snow for all the ski runs and the freezing temperatures on the ice rings.
• 1:31:51
  Now we get to 2022 in Beijing.
• 1:31:54
  All artificial snow.
• 1:31:56
  Yes.
• 1:31:57
  Not a single snowflake there was...
• 1:32:00
  made by the actual weather.
• 1:32:03
  It was all artificial snow in 2022.
• 1:32:06
  And in a recent study, scientists have looked at the past 19 venues on the Winter Olympics to see how each one might hold up, given the climate scenarios that we can forecast into the future.
• 1:32:18
  They found that by mid-century, four of the former host cities, including Shamanix, Sochi, Grenoble, and somewhere in Germany, Garmisch,
• 1:32:28
  parten question that's pretty that's pretty good yeah germany i can do that's pretty good those would no longer be able to um host reliably given the climate that we're coming in even under the best case scenario that's actually okay and if we do the current fossil fuel rate of burning then places like squaw valley and vancouver are also not going to be able to do it right this is getting pretty bad yeah and
• 1:32:54
  Especially when we have something like the Paralympics, which happened, you know, a month or two after the real Olympics.
• 1:33:00
  Snowmaking requirement is going to keep on increasing because we're going to have to keep making snow in March and...
• 1:33:08
  The amount of snow that we'll have to make is going to increase all the way from zero to now like 80% if you want to do it in Eastern Europe and things like that.
• 1:33:16
  Right?
• 1:33:16
  I mean, we're already seeing this.
• 1:33:17
  I talked about this, but at Big Bear.
• 1:33:20
  Yeah.
• 1:33:22
  The only reason you can ski on three or four of the runs right now is because of artificial snow.
• 1:33:27
  Period.
• 1:33:27
  Yeah.
• 1:33:28
  And, you know, we think that, OK, maybe this can be fixed by snowmaking, right?
• 1:33:31
  Maybe we just artificially make all the snow that we ever need.
• 1:33:36
  That's not even going to work either, because even snowmaking works at an ideal temperature, right?
• 1:33:40
  You need to be lower than the dew point, which is when the snow that you're putting out is not going to condense into water droplets.
• 1:33:47
  Right.
• 1:33:48
  Right.
• 1:33:48
  Right.
• 1:33:48
  And so it's you've got to be below negative two degrees Celsius, 28 degrees Fahrenheit.
• 1:33:53
  I mean, depending on the pressure, but usual and the humidity.
• 1:33:55
  But usually that's about the temperature.
• 1:33:58
  Now, that's really bad because we've covered a story on this podcast before about how the world mountains are warming up faster than the planes.
• 1:34:05
  Yes.
• 1:34:06
  Because hot air rises and so on and so forth.
• 1:34:08
  So we have that we have that little tidbit that the mountains are actually warming up.
• 1:34:13
  Yes.
• 1:34:13
  Yes.
• 1:34:14
  So all of these factors are combining to make it really, really quite problematic to do Winter Olympics in the same places that we used to be able to do.
• 1:34:23
  Right.
• 1:34:24
  People are suggesting that they can do something called snow farming, which is a new thing.
• 1:34:28
  I don't know if you've heard about this.
• 1:34:29
  I didn't even know this was possible.
• 1:34:30
  Okay.
• 1:34:31
  You get a bunch of snow and you pile it up with massive piles of sawdust, tarps and wood chips, and that retains 60 to 75 percent of the snow.
• 1:34:43
  Interesting.
• 1:34:44
  I need to look into how the physics of that works because, like, over the summer, you're telling me it doesn't melt?
• 1:34:49
  Yeah, I'm not.
• 1:34:50
  That one... That's crazy, but it's, like, working.
• 1:34:52
  It, like, works and people are doing it.
• 1:34:54
  If you guys want us to cover the snow farming story, let us know in the comments.
• 1:34:57
  Yeah.
• 1:34:58
  Because I am also...
• 1:35:00
  That doesn't make any sense.
• 1:35:00
  That doesn't make any sense.
• 1:35:01
  Like you put sawdust.
• 1:35:02
  Shouldn't it be warmer under the sawdust?
• 1:35:04
  I don't know.
• 1:35:05
  In any case, I'm just saying that like, you know, people are going to have to come up with interesting strategies now to keep the Winter Olympics going.
• 1:35:12
  I will say I do.
• 1:35:13
  I do find sometimes a little bit frustrating where the answer to the problem is always, oh, we'll just find a way to innovate our way out of the problem.
• 1:35:20
  No, without actually giving a specific answer.
• 1:35:22
  Like you can't just be hand wavy about it.
• 1:35:24
  No.
• 1:35:24
  Like, like how, how?
• 1:35:26
  Yeah.
• 1:35:26
  Like, okay.
• 1:35:28
  Yeah.
• 1:35:28
  We'll just, we'll just like, but no, like what, like why does that actually work?
• 1:35:31
  Yeah.
• 1:35:32
  Like fundamentally the physics wise, you can't just make snow, right?
• 1:35:37
  That's still depends on weather.
• 1:35:38
  Right.
• 1:35:39
  And so anyway, anyway,
• 1:35:40
  All right.
• 1:35:41
  The next thing I want to talk about was curling.
• 1:35:43
  Yes.
• 1:35:43
  Curling is an amazing sport.
• 1:35:44
  The most electrifying sport in the Olympics.
• 1:35:47
  Yes.
• 1:35:48
  And all of the curling stones come from a single island.
• 1:35:51
  Wait, really?
• 1:35:51
  Which is crazy.
• 1:35:52
  Oh, okay.
• 1:35:53
  There's two places.
• 1:35:54
  There's a little island in Scotland called Alissa Craig.
• 1:35:56
  I think that's this one.
• 1:35:57
  Yeah.
• 1:35:57
  And then there's the Treffer Granite Quarry in Wales.
• 1:36:00
  Okay.
• 1:36:00
  All of the curling stones...
• 1:36:02
  come from this island.
• 1:36:03
  Each stone is 40 pounds.
• 1:36:05
  And the reason why is because somehow these places have very specific types of granite.
• 1:36:11
  In a curling stone, there's two surfaces.
• 1:36:13
  There's the part that comes into contact with the ice, which is the running band.
• 1:36:17
  And then there's a part that comes into contact with other stones and knocks them over.
• 1:36:22
  And that's the striking band.
• 1:36:23
  And you need two different types of physics for each of them.
• 1:36:27
  The running band, which is the thing that's on the ice, needs to be smooth.
• 1:36:31
  Because you don't want grainy texture there because it's going to be moving along the ice and you want it to be as slippery as possible.
• 1:36:39
  Low friction.
• 1:36:40
  Low friction, exactly.
• 1:36:42
  The striking band, that striking surface, you actually want a more granular structure in that granite because as it hits one another, if it's a really nice crystal, there's going to be cracks.
• 1:36:59
  Mm-hmm.
• 1:37:00
  So what you want is the two different things, the two different parts of the stone to have different sort of chemistry.
• 1:37:09
  Yes, because they have two different functional areas where they're relevant.
• 1:37:13
  Yeah.
• 1:37:14
  One for making sure it gets down the ice quickly and sufficiently.
• 1:37:17
  Yeah.
• 1:37:17
  And the other to make sure when it contacts other stones...
• 1:37:20
  It does so in the most.
• 1:37:22
  Yeah, and it's like a really nice elastic collision that we read about in physics textbooks.
• 1:37:26
  Yes.
• 1:37:26
  Momentum is conserved.
• 1:37:28
  There's no, like, nothing lost to heat and things like that, right?
• 1:37:31
  Okay, so somehow there's only two places where there's these good stones.
• 1:37:36
  I am convinced this is a racket.
• 1:37:39
  It's a racket.
• 1:37:41
  Like...
• 1:37:41
  Rack it.
• 1:37:42
  There's no way, dude.
• 1:37:43
  But look, okay, so people tried to make an effort in Canada in the 1950s.
• 1:37:49
  They tried to use the very black igneous rock called an orthocyte to make these.
• 1:37:55
  And then those things just started chipping way too quickly.
• 1:37:58
  Like the stones would get damaged.
• 1:38:00
  And speaking of the Canadians, they are now cheating in the Winter Olympics.
• 1:38:05
  I don't know if you guys saw this.
• 1:38:06
  Typical.
• 1:38:06
  Typical.
• 1:38:08
  Dude, this guy, so you're only allowed to grab the handle and then let it go before that line.
• 1:38:16
  This guy was grabbing the handle, letting it go, and then pushing it with his finger.
• 1:38:19
  And then when the Swedes called him out, he literally started yelling at them and told them to F off.
• 1:38:25
  And then the Swedish NBC, these guys are the OGs.
• 1:38:30
  They placed their own camera because the Olympic cameras are placed like here and behind.
• 1:38:37
  So the guys could have gotten away with that.
• 1:38:39
  So what are you saying?
• 1:38:40
  The initial comment of the initial complaint was prior to the video.
• 1:38:43
  Yeah, because this is a known thing that this Canadian dude does.
• 1:38:46
  OK. And so the Swedish guys for the Canada versus Sweden match.
• 1:38:51
  Yes.
• 1:38:51
  The Swedish NBC, their television.
• 1:38:54
  Yes.
• 1:38:54
  Put a specific camera.
• 1:38:56
  Yes.
• 1:38:56
  at that spot and caught this guy red red-handed he did it again against the swiss and i think now he's fainted facing penalties after all that bluster the judges were just like not doing anything yeah yeah yeah because they were like well i don't know if he did it or not well now it's like we got a specific camera that's showing that he did it i guess it must be the case that no one was dumb enough to do it which is why there's no like goal line technology for this current
• 1:39:18
  But it seems obviously true to me that like if that's the rule.
• 1:39:22
  Yeah.
• 1:39:23
  And there's an advantage you can get from breaking that rule.
• 1:39:25
  Because you give an initial momentum.
• 1:39:28
  Yes.
• 1:39:28
  And then a tiny, tiny nudge will like.
• 1:39:31
  Yeah.
• 1:39:31
  Because the, you know, the target is meters away.
• 1:39:35
  Yes.
• 1:39:35
  So even a tiny bit of nudge in that angle is going to cause a huge deflection later on.
• 1:39:39
  Right.
• 1:39:40
  It's not nothing.
• 1:39:42
  Very, very, very sad.
• 1:39:43
  Look, we're trying to continue to establish and repair our North American relationships with the Canadians.
• 1:39:51
  There are several things to be upset at us about, but we are now upset at you guys about this cheating.
• 1:39:56
  That's BS.
• 1:39:56
  Let's not, remember, as Bad Bunny said, it's America and they enlisted everybody.
• 1:40:01
  Yeah, yeah, yeah.
• 1:40:01
  And so we got to be on the same page here.
• 1:40:03
  No cheating in the Olympics.
• 1:40:04
  Yeah, yeah, yeah.
• 1:40:05
  I want to beat Sweden as much as anyone.
• 1:40:08
  We're on the same team on this one, okay?
• 1:40:10
  Okay, and the next thing about cheating, I don't know if you saw this.
• 1:40:13
  No, I did not.
• 1:40:14
  This is a family-friendly podcast, and so when I say these words, I am using the biological terminology.
• 1:40:21
  This is penis gate.
• 1:40:22
  Okay.
• 1:40:23
  Okay, ski jumping penis gate.
• 1:40:25
  Allegedly, some competitive ski jumpers may have artificially enlarged their crotch area by injecting...
• 1:40:33
  their genitals with engorging chemicals or stuffing their underwear to make a bigger bulge i'm not kidding what is the what is the athletic advantage of this okay so that's exactly like because that's a lot like what injecting stuff down there what mate what are you doing okay so here's the idea right um ski jumpers ski jumpers they get laser
• 1:41:00
  Whatchamacallit?
• 1:41:01
  Hair removal?
• 1:41:02
  No, no, no.
• 1:41:03
  It's like their body gets laser scanned.
• 1:41:06
  Okay.
• 1:41:06
  And then the suit that they put on has to be exactly fitting to their body.
• 1:41:12
  Oh, it's like Frozone's, where is my super suit?
• 1:41:15
  Yes.
• 1:41:16
  It's exactly that.
• 1:41:17
  Okay?
• 1:41:17
  And the reason for that is...
• 1:41:19
  Their sport has everything to do with aerodynamics.
• 1:41:23
  Okay?
• 1:41:23
  So here's how the physics of that works, right?
• 1:41:25
  You go down this giant ski hill, and I think we've got a visual for it.
• 1:41:29
  Yes.
• 1:41:29
  This is 4-13.
• 1:41:30
  Yes.
• 1:41:30
  We go down the giant ski hill, right?
• 1:41:33
  During that downhill, you want to reduce drag as much as possible to get as much velocity.
• 1:41:37
  Yes.
• 1:41:37
  Now when you lift off, you want to increase the amount of drag because that is going to increase the amount of lift.
• 1:41:44
  No way, dude.
• 1:41:46
  Right?
• 1:41:47
  And what these guys are saying is that if I have a slightly bigger, like, crotch area, that is going to increase my lift.
• 1:41:57
  And so it's going to make me go farther.
• 1:41:58
  And that's why they're injecting stuff down there, dude.
• 1:42:00
  This is so insane.
• 1:42:02
  Bro, what is going on?
• 1:42:03
  The surface area of the bulge increases your lift.
• 1:42:07
  Yeah, yeah, yeah.
• 1:42:07
  It's going to get you a farther distance.
• 1:42:09
  Yeah.
• 1:42:09
  This is a ski jumping thing where they go up and they're just like swanning.
• 1:42:13
  They do like a V so that they can.
• 1:42:14
  Yeah.
• 1:42:15
  And they have to hold this position.
• 1:42:19
  But like just, I guess, a few inches down there.
• 1:42:23
  Look, a few inches has never been more impactful.
• 1:42:27
  Right.
• 1:42:27
  Right.
• 1:42:28
  It's a gold medal versus not meddling.
• 1:42:30
  Don't try this at home.
• 1:42:31
  No, not at all.
• 1:42:32
  This is for the Olympics.
• 1:42:33
  Or even at the Olympics.
• 1:42:34
  What are you guys doing?
• 1:42:35
  This is not for your at-home Olympics.
• 1:42:37
  Yeah.
• 1:42:38
  There's no medals in that sport.
• 1:42:40
  It's a losing game.
• 1:42:41
  Yeah, dude.
• 1:42:42
  Anyways, that was my last story.
• 1:42:43
  That was sort of my stream of consciousness.
• 1:42:45
  Of all the things I was researching, this is the V. And you see, like, if you make the crotch area just a tiny bit bigger, maybe you get, like, two more meters over the other guy.
• 1:42:55
  I don't know what the judicial system is for sports.
• 1:42:58
  I think it's CAS, the central something for sports.
• 1:43:02
  Yeah.
• 1:43:02
  I think you're going to have to look at this issue.
• 1:43:04
  Yeah.
• 1:43:05
  This is ridiculous, guys.
• 1:43:06
  That's...
• 1:43:07
  I mean, okay, look, I think steroids is more reasonable than whatever this is.
• 1:43:12
  Yeah, let's not do that.
• 1:43:14
  Yeah.
• 1:43:16
  That's I'm going to keep it PG and I'm going to leave it there.
• 1:43:19
  Yeah, yeah, yeah, exactly.
• 1:43:20
  This is a family friendly show.
• 1:43:22
  Yeah, but this is a real thing.
• 1:43:23
  That's a real thing.
• 1:43:24
  It's happening.
• 1:43:25
  It's happening.
• 1:43:25
  I think you guys should know about it because because I was alarmed.
• 1:43:29
  That is that is quite.
• 1:43:31
  Yeah, that is quite curious.
• 1:43:34
  And this also speaks to, again, the what we talked about earlier, the stakes as it relates to these high level events like the Olympics.
• 1:43:42
  Yeah.
• 1:43:42
  where the stakes are that high that people will literally physically do anything to their bodies to win.
• 1:43:47
  It's never worth it.
• 1:43:49
  No, it's never worth it.
• 1:43:50
  We want to keep human sports, human, no robotic Olympics when they start, don't support them.
• 1:43:58
  So,
• 1:43:59
  We have gotten to the conclusion of our fantastic Winter Olympics episode.
• 1:44:06
  I am starting to get some beads of sweat on my nose here.
• 1:44:10
  Because we are in LA.
• 1:44:11
  We are in LA.
• 1:44:12
  And it's not that cold.
• 1:44:13
  Although it is raining, it is not snowing.
• 1:44:15
  And so these ski jackets are quite warm.
• 1:44:18
  Yeah, they're quite warm.
• 1:44:19
  But we did a lot of fun stories today.
• 1:44:22
  Again, science is all around us.
• 1:44:24
  We talked about the physics of slippery ice.
• 1:44:27
  We talked about the neuroscience of choking.
• 1:44:29
  We looked at how climate science, climate change, anthropogenic climate change is going to impact the ability to have future Winter Olympics.
• 1:44:39
  So if you have kids right now that you're trying to get on the U.S. Olympic team for the 2042 speed skate tournament,
• 1:44:46
  or I guess any of these outdoor related events, it's going to be relevant to you living vicariously through that future Olympic child.
• 1:44:54
  And our rundown was super fun today.
• 1:44:56
  We had a ton of good stories.
• 1:44:58
  The AI doing physics thing, I think is a, no, that's an interesting one.
• 1:45:02
  I think,
• 1:45:03
  Let's not sleep on this issue.
• 1:45:06
  I understand.
• 1:45:06
  Again, I have similar feelings about where the money is coming from for all these billionaire frontier models.
• 1:45:14
  Why it's bad to replace all human labor with LLMs and software, AI and robots.
• 1:45:21
  And also we have no plan for distribution.
• 1:45:24
  And yes, there are some cases where hype is happening.
• 1:45:27
  Yeah.
• 1:45:28
  But this is happening, and it is moving very, very, very quickly.
• 1:45:34
  And if the last time you tried any of these things was the free version of ChatGPT, even six weeks ago, even two weeks ago, where the frontier is is exponentially better than it was two to six weeks ago.
• 1:45:50
  It is unbelievable, and I just really encourage people to remove their prior biases about AI because...
• 1:45:57
  Not because we're saying it's amazing and brilliant and all this stuff, but because we're saying it's growing at a rate that no other phase shift in technology in our lived lifetime has changed.
• 1:46:09
  And the implications of that are so enormous that,
• 1:46:13
  It's like hard to really articulate.
• 1:46:15
  And it is not just software engineers and everyone else is going to be fine.
• 1:46:20
  This is something I'm just going to keep beating the drum on because I think it's very important.
• 1:46:23
  But we had other great stories today.
• 1:46:25
  We talked another animal behavior study story on cats' meows versus purrs.
• 1:46:30
  We did immunology talking about us discovering how we rewrite the immune system and the epigenetic versus long-form genetics that informs that.
• 1:46:40
  The origin of life beyond our common ancestor.
• 1:46:42
  That's a really fantastic one.
• 1:46:44
  And then the astrophysics and exoplanet stories about us discovering a solar system that should not be possible.
• 1:46:52
  I mean, where else, everybody, can you get the absolute best breakdown of breaking and frontier science across physics, chemistry, biology, all the stuff in between?
• 1:47:05
  It's only here at FFP, the Fundamentally Flawed Podcast.
• 1:47:09
  Yeah.
• 1:47:09
  Yeah.
• 1:47:10
  which is a funny name that we got out of one of the little thingies we use, but it is from first principles.
• 1:47:16
  We are so grateful for our audience.
• 1:47:18
  Last two things.
• 1:47:19
  Comment.
• 1:47:20
  Yes.
• 1:47:21
  What should it be?
• 1:47:22
  Thoughts.
• 1:47:23
  Now, what are your thoughts on the controversies that we encountered?
• 1:47:29
  Whether it's the Canada or the ski jump crotch.
• 1:47:33
  Or the gate, the X gate.
• 1:47:36
  This is a free form one.
• 1:47:38
  Some of you guys have had some really good comments, so we really appreciate you.
• 1:47:41
  Again...
• 1:47:42
  If you love the show, if you've gotten to this point in the show, if this is not your first episode, you're one of the key members of our community.
• 1:47:48
  We appreciate you listening to the pod.
• 1:47:51
  We have now opened up our donations flow.
• 1:47:54
  This is going to be hugely helpful for us to be able to continue to produce and expand the team and ability for us to do more shows with more content.
• 1:48:04
  all the time with the pod.
• 1:48:05
  So any support we can get from you all is really appreciated.
• 1:48:08
  And I will also mention, we are building out our community portal, which is going to have some exciting features.
• 1:48:15
  We have a signup form on the homepage at FFPPod.com.
• 1:48:19
  So sign up for email updates.
• 1:48:20
  If you are interested in donating, go to FFPPod.com backslash donate.
• 1:48:26
  You can find us on all of the socials, X, Instagram, Facebook, LinkedIn, Facebook,
• 1:48:32
  Blue Sky, anyone's I forgot to mention, the main full-form video podcast is on Spotify, Apple Podcasts, YouTube.
• 1:48:41
  We're also on Overcast, iHeartRadio.
• 1:48:44
  Effectively, anywhere you can listen or watch stuff, you will find us.
• 1:48:48
  If you watch our clips, our clips will not be in context of the almost two-hour episode that we did today.
• 1:48:55
  And so, yes, you're going to be a little bit.
• 1:48:57
  You might have a clever comment that is true in the context of the clip, but it is untrue in the context of the full episode.
• 1:49:05
  So we appreciate the feedback.
• 1:49:06
  But so you know, we always email all of the authors of the papers we cover to get their feedback and provide corrections from them.
• 1:49:14
  And if we don't get corrections from them...
• 1:49:18
  then the clip gets to live on in perpetuity with no corrections.
• 1:49:21
  I am your host, Lester Nare, joined as always by my co-host and our resident PhD, Krishna Chowdhury, with another wonderful deep dive here on the Winter Olympics.
• 1:49:36
  We will see you all next week.
• 1:49:41
  Because this is from First Principles.