[Music]

Okay, here we are in the Origins podcast

for one of my favorite times every month

when I get to speak to one of my

favorite people, Sabina Hosenfelder, and

I get to I get to hear her unique take

on science and we have a discussion of

of science topics in the news, uh,

what's new in science. And it's it's

something I've come to really enjoy and

I'm really happy to say as far as I can

tell other people actually enjoy it too.

So Sabina, how are you doing there in in

Europe? Is it hot?

Today is fine. Uh it's actually raining

outside. So um yeah. Well, good to see

you and good to see you in a studio this

time.

Yeah, that's right. I try Well, I try,

you know, following you, I try and up

the game a little bit because you're

always so professional. So it's nice to

be in a studio. Thank you. Well, we're

going to talk about a whole bunch of

interesting cosmic and terrestrial

topics today. And speaking of

terrestrial, you you're going to talk

about a topic that's supposedly going to

change the future of the world. So, why

don't you talk about that?

Yeah. Um, I want to talk about hydrogen.

Um, so this is a topic I have followed

for a long time because I have a

personal grudge uh with the German

government that's invested a lot into

the so-called hydrogen economy that

isn't going anywhere. So, uh, I have a

lot of misgivings about this stuff

because I've always said like it doesn't

make any sense. Like to begin with, um,

you know, the idea that we would produce

hydrogen from renewable energy and then

use that to, um, you know, create

electricity, um, is just it's terribly

energy inefficient. So, it doesn't

really make any sense.

Yeah. Um and and so in in practical

terms, what's actually happened is that

uh we're paying taxes for what's been

called the hydrogen ready power plants.

What what what are they? They're

actually just gas power plants because

uh once you have something that can deal

with methane uh that has a lot of hes in

it, um it's not that difficult to switch

it to uh burning hydrogen. So, and and

hydrogen has has a lot of problems. Um,

the I don't want to go through the the

whole list, but besides the energy

efficiency, uh, I think that the the

next biggest problem might be what's

been called hydrogen in brittlement.

Like whenever you you want to store the

stuff or you you want to push it through

pipes, you have this issue that hydrogen

is the smallest of all atoms. uh and so

so it creeps into everything uh and

destroys the material sooner or later.

So and so um I think that this is going

to become uh a maintenance nightmare uh

which is also going to drive up the co

cost. Let me let me interrupt for one

second just because we we know it but

just so people know you might and in in

in light of the things you said already

people might say why hydrogen at all but

the whole point

initially is hydrogen burns with oxygen

to form water and releases energy and

therefore there's no carbon dioxide

there's no bad byproducts that's why

people were there no fossil fuel carbon

byproducts and that's that causes the

initial excitement about about hydrogen

and why there's so such verbiage and

I'll also say I used to speak out

against it because in the old days when

I did people didn't realize that you

know it sounds great except you get the

hydrogen from fossil fuels so you end up

in the old days you used to so it end up

it was a scam because you didn't save

any carbon or anything to get the

hydrogen but things have changed so

anyway just to preface where you're at

go on

yeah uh uh that's right um so um at the

moment we get almost all the hydrogen

actually from methane or coal in some

cases. Uh because exactly because there

are all these H's in there. But if you

have methane, that's CH4. Um you also

have the carbon. And so if you you split

off the hydrogen, you're left with

carbon dioxide, which is exactly what

you don't want. Um and so this is why

that's the idea. We produce it from

renewables like from solar or wind or

something. And that just isn't

happening. And even if we would manage

to get it done, uh it would be

ridiculously expensive. So I I doubt

it'll ever make sense. Um nevertheless,

uh the German government and also the uh

many European countries and also the

United States, by the way, have put a

lot of money into this so-called

hydrogen economy. Uh and and but so that

so there's one possibility to make sense

of this which is if we were to find

naturally occurring reservoirs of

hydrogen. This has been called white

hydrogen. Not actually referring to the

color of the stuff but just they have a

color scale that tells you where the

stuff comes from. And so this white

hydrogen uh is something that you can

find in the ground in some places like

like natural gas, like methane. uh and

so so it gets stuck in like porous rocks

uh and then it can drill into it and um

I actually don't know exactly how it

works but you know they get it out of

the ground and they can use it uh and

and so um the issue is just that um so

far there's been only one place where

they actually do this uh which I think

is in Marley or somewhere so it exists

but it's been believed to be really rare

and then like two years ago um there was

a group of French researchers uh we uh

that claimed they'd found a reservoir of

white hydrogen actually pretty close to

the border to um Germany and just like

two months ago they claimed they had

found a second one a little bit further

north um yeah so big of true like so we

haven't actually seen them extract

anything from the ground and then the

other thing that happened was a group of

geologists uh went and tried to

generally analyze ize like the the

possible abundance of the stuff in the

ground based on what we know about how

those rocks formed. There there are

different ways that hydrogen could be

formed underground. Uh for example,

because um water comes into contact with

some very hot stones that contain

certain minerals and so it gets split uh

into hydrogen and oxygen and the

hydrogen then embeds uh in the stone or

it could it could happen through

radioactive decay and and some some

other possibilities. and they just

produced this global map where they said

look in these places uh it's possible

that you can find something um and uh

they estimate that all these hydrogen

reserves combined could power the world

for 100,000 years. This is why I say big

if true, right?

So of course just because you put it on

a map doesn't mean you can actually

extract it. So there are lots of ifs and

thens there. But honestly to me it was

like I I got this feeling like we might

have underestimated

um if if not um

you know with an actual mathematical

estimate uh estimate but um

unconsciously how much of this stuff

might already be lying around. So if it

was true and we could just you know if

even if it's just a fraction of this and

I mean if it only if it lasts for a

hundred years or something that would be

a really big deal

and there are a lot of startups I mean

they're saying a lot is maybe is saying

too much but a couple of startups who

are trying to to look into this and are

actually trying to extract uh this white

hydrogen. So I think that's a very

interesting development.

Yeah, I think it is. I mean I learned

about it from you and I look I was

looking at that paper you mentioned it a

little bit and um I was intrigued

because it does make sense uh you know I

many years ago when I was trying I I

first proposed we look for antiutrinos

from the earth to do geoysics and now

people do it but this was 40 years ago

um there are lots of interesting

processes radioactive processes but

other processes and let's face it

there's lots you know when I think about

deep ocean vents

what comes out of there is hydrogen

sulfide. There's lots of hydrogen coming

out of those vents. So, when you think

about when you when I saw that, I

thought, you know, it makes sense. There

probably is a lot of a lot of hydrogen

that may come out. And, you know, but as

you point out, it's a long way from

maybe having it to making it practical

and useful. Um, but it does at least it

made me feel a little bad because I used

to make so much fun of the hydrogen

people by saying, "Yeah, first you get

gasoline, then you then you you burn it

to get hydrogen and then you use your

free clean hydrogen." It was just it it

was a it seemed like a scam. Plus, it's

a little bit dangerous to transport, but

but as you said, also it's very

invasive. But one thing that did occur

to me, I do want to ask you this. You

sort of mentioned it as a negative, but

but if if we had so much hydrogen, we

didn't know what to do with it. Then if

you had something like fishision or f I

mean obviously you had fusion, it' be

different. But even fishision something

that uh the problem with fishing plants

and fusion plants is you produce

electricity, which is not a problem, but

it's nice. Um, but you do sometimes un

until everyone has electric cars, you

you don't necessarily use electricity

and and

if you had hydrogen and combined with

oxygen, you would be you would be able

to power as fuel cars. And so it

occurred to me that um maybe if we had a

lot of some kind of um renewable or

carbon-f free power source then

splitting hydrogen oxygen from seawater

which is very energy intensive. If we

had a surplus of energy it might not be

completely crazy because not all forms

of energy we need are in the form of

electricity. uh you know, you may want

to have fuel. It's and and we don't have

these grids that allow you to move

electricity and and batteries that allow

you to store it. So, if you could

convert what would otherwise be

electricity to hydrogen that you could

use as fuel locally in your car or in a

power plant, it might not be so crazy. I

was thinking about it. What do you

think?

Yeah. Um I I I think that's that's uh

totally correct. In the end, I think

it'll come down to how how expensive is

it? But uh like this this is the most

practical application of hydrogen that

people have thought of. Uh you could use

it for transportation

um for example in big trucks or

something because you know these

hydrogen tanks are kind of heavy um so

for you know for for standard passenger

cars is maybe not so great. I mean they

do exist but um you know they have some

down size but you know if you if you

have a big truck that that weighs some

tons already it doesn't really matter

basically. Um and the other thing is

that you could use it to store energy.

Uh like so this is one of those things

um where for example in Germany there

was supposed to be a project where they

had these offshore wind turbines that

were supposed to produce hydrogen and

they wanted to pump this uh through some

pipes uh to somewhere on on land. Uh and

so if you have a surplus of energy from

there you can store it in the hydrogen

and then you can use it when the winters

blow. Basically, that was the idea. I'm

not sure what happened to this. The

issue with all these things is that um

you know,

they're too expensive. Uh and and and so

so especially Germany like we have all

that coal like and then it's that cheap

so why don't we just dig up some more

coal? Um yeah. So so it becomes a big

political problem.

Yeah. Yeah. Well, that's the problem. I

mean that was always my by the way my

concern about fishing. I have no

problem. I don't have any safety

concerns about nuclear reactors,

well-designed nuclear reactors, but

given all the regulations and everything

else, they're just damn expensive

compared to certain things. And so, it

was an economic question for me as much

as anything else. it. Yeah. Um but the

idea of storing it rather until we have

good batteries or a better a better

infrastructure nationwide to transport

electricity it, you know, if you

transport it as hydrogen and then turn

it into electricity here where you it

might it might work. On the other hand,

if I saw a truck that said hydrogen and

I think I've seen it, I tend to try and

steer clear of them having been

remembering the Zeppelin episodes that

from the old days. uh you know it is

kind of flammable. Anyway um okay so

much for terrestrial hydrogen in a sense

I want to move now to the cosmos cosmic

hydrogen which is one of the things and

cosmic hydrogen exists mostly in the

form of stars or a lot of star stars and

galaxies and it leads me to the uh news

of the month which I think is I think is

pro probably the most the biggest story

I I heard reported among the news of the

month was the the opening up of the Very

Ruben um telescope in Chile. um which is

named after a lovely woman I knew Vera

Rubin and and and her story is poignant

um as a as a young woman scientist at a

hard time to be a young woman scientist

but she never harped on that and she did

her work and and Vera Rubin was one of

the first people not the only one she

could sense to get most of the credit

now but not the only one to first argue

that there was dark matter in our galaxy

by looking at the rotation curves of our

galaxy and other galaxies looking at

hydrogen lines and and and hydrogen gas

clouds and um and um making an argument

which at the time many people didn't

believe that our galaxy was orbiting was

was uh rotating too fast and now we see

most other galaxies are and you and I

have different views of might what may

be the cause of that but um uh anyway

I'm very happy it's named after her she

was a lovely woman and a and a and a

good really good scientist but it's a

telescope in Chile

8.4 4 meter 8.4 meter telescope which is

pretty big and first of all it's nice to

see new telescope being built living in

a country I don't live in the United

States but living next to a country

which can't seem to build a big

telescope because it might offend some

deity of some weird indigenous group

that that have land on it and the the 30

meter telescope and is now a dead

project probably in the United States

and the Trump administration wants to

kill it anyway. It's nice to see a new

telescope being built and it's a very

diff different type of sculpt telescope.

It got a lot of press because it

produced some pretty pictures, which is

generally what telescopes do for people

on the ground. That's what people like

about them is some pretty pictures. It's

not really designed to produce these

pretty pictures. It's designed to scan

the entire southern sky every day, every

two or three days for the next 10 years

to make to make what if you were to look

at it might look like one of those most

boring 10-year long movies you could

imagine, but if you're a scientist, it

could be interesting. The idea is that

it's it's got this incredibly big field

of view. you it it essentially 45 moons

worth of the sky at any one time as

opposed to the James Webste telescope

which has also got a pretty big field of

view but it's like one moon. So and it's

it it's and it moves faster than any

other terrestrial telescope without

jittering and therefore it can it could

see the entire sky every three nights

and it and it and it has the largest

camera in the world a car-sized camera

3.2 2 gigapixel camera. Um, which you

know maybe you know which is always

amazing me to think about because I know

in 10 years you and our our iPhones will

have 3.2 gapixel cameras but right now

uh a 3.2 gigapixel camera is a thou you

know it's almost 100 times bigger than

the camera in your iPhone. it it uh it's

it produces a single image that would in

order to see it adequately require 400

ultra HD TVs

and each image each night it produces 20

terabytes of data which is more than the

information in all the world's books and

so it sounds great and it's because it's

fast moving you can see things but

because it has such a big field of view

and it looks night after night after

night um it can look for transient

events all over the sky time and it's

already in and one 10 the data they

released was only the first 10 hours

which already allowed them to see 10

million galaxies in that big image but

also discover all these asteroids

because they can actually see them

moving through the system and so

it it sounds great and it I'm sure it

will be great and it's touted to do many

things of course like it's almost a dur

regor it's almost required

constitutionally for every one of these

telescopes to say that they're going

learn about dark matter and dark energy

as a result. And I think you and I are

more skeptical sometimes. I think it'll

actually, to be fair, because it has

such a big field of view, it'll look at

something called gravitational lensing.

It'll be able to see lensing very, very

well. And that at least does map the

gravity of field of galaxies. So maybe

it'll reveal stuff about dark matter.

I'm still not convinced about dark

energy, but it actually may be most

useful for saving our lives in the long

run because it's going to discover

thousands of new asteroids, some of whom

may be some of which may be on a be

trajectory of on Earth,

an earth crossing trajectory, and maybe

give us enough time to save ourselves

one way or another. But but you know I

don't want to make too much light of it

because it is a new window on the

universe that particularly is good for

seeing a lot of things at once and

transient events supernovi pulsating

stars and it may be useful

in integrating with the with with

gravitational wave detectors because

it's going to be able to look in real

time very fast for odd signals. So when

when a gravitational wave detector sees

neutron stars collide, the gravitational

waves, it might be able to tell the the

the telescope where to quickly where to

move. And since such a big field of

view, even if you can't localize it very

well, this telescope might see it. So

it's going to it's going to improve

astronomy. And and I suspect what it

will discover are things we don't expect

because every time I open a new window

on the universe, you you're surprised.

and the the and the the usual suspects

like dark matter and dark energy I

expect won't be won't be changed

revolutionarily by it but it's really

exciting to have a nice new telescope

that does something very very different

with neat new technology um especially

if you like neat new new technology so

and and it and it has produced a few p

interesting pictures and if and and I

and all I for all I know there's going

to be some nerds out there in Germany

who buy themselves 400 HD

TVs, put them up on a big screen and and

and and see a picture from the from the

from the Reuben telescope someday. I

don't know. What do you think?

Um I think you've you've said almost

everything. Uh and I've I've thought

very hard about what I could possibly

add. Um so I have one thing to add,

which is that um they put out an app

which is called the Sky Viewer app that

allows you to So I don't know if that

that'll go through. So you can zoom in

in your high definition picture.

And I think they plan to,

you know, add some annotations, you

know, more details about what what's

going on in these pictures. Like this

one is very pretty. So you already

download it.

So actually I think it opens in the

browser. So like skyviewer.app. I think

that's just a website. Um Oh, I see. In

any case, so I mean you kind of said

this a little bit dismissively like it

puts out pretty pictures. Uh but you

know I think this is kind of really

important for for science communication

that you have something to show um

something that people can relate to and

and people just really like these

images. I don't see anything uh wrong

with it.

Oh it excites kids. Even if it excites

kids to study science in school, it's

great. I mean

Exactly. Exactly. So I I think it's

great they they put out something like

this. I hope they'll make a little bit

more of it like this seems to be just

you know the core of an idea where um

they're smart to come

and you know and if they did something

like that I I have always mixed feelings

but it's not that you know there's

citizen science too so if they have an

app and they've got these huge images

they can encourage people to look in the

images for things that they I mean I'm

sure their data analysis and their AI

you know mechanisms are are good but you

never know if you allow a few hundred

000 people to look at these images and

look at them. Maybe people will see fun

things that other people can't. So yeah,

I think it's it's wonderful and it does

illustrate a few things. I mean I I

space telescopes are very important and

I'm very sad that the next large space

telescope, the Roman space telescope,

which is essentially built and paid for,

may not fly because the Trump

administration has decided science is

bad. No science. No science makes the

world a better place. is apparently what

the Trump administration has decided. Um

because scientists are bad because

scientists are bad because of money

liberal therefore science is bad

therefore kill all the projects that

have already been done. But anyway,

that's a little button for me. But the

you know the next space telescope may

not fly because at least the current

administration hasn't doesn't want to

budget money for it. But it's lovely

when when you can see how much you can

do on Earth with with new telescopes and

it's and they're they're expensive. This

one was actually less expensive. I

thought it was under a billion dollars.

I think it was $800 million.

The uh the uh next large telescope uh

the I think the Mellan I think it's

called the large Mellan telescope.

Anyway, it's a it's probably more like 5

to 10 billion dollars, but uh which is a

which is a scale. This is 8.4 meters.

The next the larger telescopes are are

are tens of meters across like this, you

know, 15 to 20 meters across. and

they're incredible. And I'm still

amazed. You know, I had a telescope when

I was a kid. I still have a telescope. I

never do anything really great with it,

but it amazes me that here on Earth, you

can look through the atmosphere and see

all of that that it's all out there.

It's just just it is in so inspiring.

Anyway, yeah. So, if it does nothing

other than inspire young people to uh to

get interested in the universe, then I'm

all in favor of it. But

who knows? May maybe we'll we'll finally

discover extraterrestrial life.

Yeah. Well, you know, it's certainly

there for transient phenomena one way or

another. My my friend Jeff Marcy and his

colleagues are big on saying that the

best way to look for SETI is not radio

waves, but to look for optical

transients. Um, and you know, maybe this

maybe this will will allow us to do

that. There's certainly going to be a

lot of data, 20 terabytes a night. I

wouldn't I I wouldn't want to be the

data that the person's in charge of that

data set. It's amazing. Well, from the

from the um immediate and and um and uh

in front of your face science to the

esoteric science that the limits of

knowledge literally and metaphorically,

I'm told that you have a very

interesting paper that you read.

Yeah. Yeah. And so, so one of the

authors of the paper is also very in

your face. Um, so you you put out a

paper together with some co-authors

um that argues that a theory of

everything isn't logically possible. So

that's my brief summary

quoting some mathematical theorems and I

want to back up a little and and explain

I mean in which sense you use the phrase

theory of everything because the several

different things of this could mean. So

in the narrowest sense of the word uh

physicist talk of the theory of

everything as something that unifies all

the four known fundamental forces. So

the three forces in the standard model

electromagnetism the strong and weak

nuclear force and gravity. Um and that

in principle should explain everything

hence the name. So in practice of course

it's not going to you know allow us to

predict human behavior or something like

this because even though people are made

of particles that are

determined by those fundamental forces

and the interaction between the

particles and and everything we can't

actually calculate uh something as

complicated as that. But in in principle

it's it would be a theory of everything

we know. And of course, you know,

physicists hope that it would also

include uh dark energy and dark matter

and that might pan out not pan out.

Um so I I think that in your paper you

use the word in a somewhat more general

sense. Um which isn't just that it needs

to combine these four forces but is

actually also a final theory. um which

um I think a lot of physicists

believe uh and certainly hope um which

was maybe best captured in Steven

Weinberg's famous book dreams of a final

theory. Um there's this idea that we're

we're pretty close to the end of physics

in some sense because we the only thing

that's missing um is this unification of

all all the forces. And yeah, that

sounds very similar to what um Lord

Kelvin said like 150 years ago, which I

paraphrase. Uh we're almost done with

physics. They're just some little niggly

bits left to sort out that

decimal points,

quantum mechanics and the entire set,

right? Um so um yeah, so I don't really

believe the stuff with the final theory.

Um but u so this is why I read your

paper because it plays to my

confirmation bias. So basically um in a

nutshell I think um the the argument has

like three different points. The first

is good theorem that I think most people

have have already um heard about uh

which so we roughly speaking it says

that um any such theory if you can write

it down in innumerable axioms and so on

so that there are some

checkpoints that it needs to fulfill but

I think for you know those are actually

fulfilled for the theories that we use

in practice um then that theory will

have statements that are true but they

that you can't prove prove to be true.

So that's a little bit awkward. You

know, if if you if you want your theory

to give answers to everything because

it's a theory of everything that

actually uh you can prove uh that's not

possible. And then the second argument

is um is kind of related to this but it

builds on a on a different theorem that

says uh well for any such theory you

can't even construct uh a a function

that that would tell you what is true

and what isn't true. And then the the

third step it's uh is a uh complexity

bound. So whatever theory you have

there's a limit to the complexity of

statements that you can make roughly

speaking. Um and you you don't you can't

tell um exactly where it is. So so um

you don't even know and that's a little

bit awkward because it could be that

nature is just above that bound. So what

kind of theory of everything is that? uh

if it if it actually can't explain that

thing which is above the complexity bar.

So I think that's a very interesting

argument uh and and honestly I think

that um this is kind of an approach to

the mathematics that we use that I think

physicists should should use more often

like so so there's strictly logical

argumentation about what's even you know

theoretically possible where are the

limits of our methods. Uh that said, you

know, I I'm I'm not so sure that it's uh

in the end relevant for one thing

because I don't actually believe that

there's a final theory. I I already said

this. Um I just think like, you know, we

will find something wrong with quantum

mechanics basically and all this final

theory stuff will go out of the window.

And then there are things like um for

example in Google's theorem um I I I

always wonder like in practice how

relevant is it actually because suppose

you have this one statement that um you

know you you you can't prove to be true

what would you do as a physicist where

you you would go and see if it's true by

making a measurement

if you can make a measurement like if if

you can't measure it who cares. So um if

if you can measure it you check it and

so if it's true you add this as another

axion right so this how it works um so

you know I have to say that I think most

physicists won't be terribly impressed

by the paper but I I quite like

well I you know I have even though I'm

one of the authors I also have mixed

feelings but but I partly again it was

my confirmation bias I go back to It's a

1984 when I was old and you were a baby.

Um uh when when string theory was not

called string theory, it's called the

theory of everything by absolutely every

practitioner who was working on it. And

in fact, even if it were a theory of

everything, it would actually be a

theory of very little, as my as Frank

Wilch used to say, because you know,

quantum gravity is is incredibly

important for understanding the

beginning of the universe and the ends

of black holes, but not for

understanding how water boils or oatmeal

boils or or any of the you know, most of

the things or how how to build better

materials and many of the things that

matter people's lives. But but I did

like I I I I I

do like various aspects of this because

I'd like to hear people stop using the

term theory of everything. But the the

idea is that if you the way I think of a

theory of everything if you had it is

that you have a theory that any that can

allow you to calculate in advance

anything you could see. Calculate in

advance and predict any measurement you

could ever do. That would be a complete

theory. There's nothing that one could

ever see or do that this theory would

not allow you to predict. That's kind of

and on any scale in the universe. And

and there there are fundamental, as you

point out, mathematical arguments that

are quite profound in some cases that

basically say that's that's impossible.

And I think that's really interesting.

Uh as you say, I get your point about

Good is very is very correct. There are

things some things are true that you

can't prove to be true. It's also t it's

more explicit in Tarski's theorem.

There's three theorems girdle, tarski

and tradings but that we talk about but

they all are different aspects of

incompleteness. And then physicists

would add that as an axiom and all we're

saying is yes it would be we call that a

meta theory. So you can't

algorithmically start from a certain set

of rules and derive everything. You

basically there's some things you could

derive and then you want to add them and

there may be an infinite number of

things you have to add. Who knows? But

um and so that's interesting in a

general maybe I hate to use the word

philosophical sense but but in a general

sense for the claim limiting the claims

that that science will ever be over uh

in that sense. Now you may argue that

the direction science going in isn't

important anyway but but

uh but I like to think of as cosmic job

security that there'll never be a theory

that allows you to predict everything.

You'll have to do more experiments and

discover that you'll have to keep

looking and you know thinking alone

won't do it. they'll have to keep

looking. And that's great because

science at its best involves looking and

and discovering things about the world.

And and the thing I like about that last

part of this incompleteness theorem is

that and because it does relate to a lot

of the claims of quantum gravity and

there's so many claims and so much

hyperola as you know about string theory

and loop quantum gravity and all of the

rest. But one of the things aspects

that's relevant to black holes is

thermalization. And one of the

interesting things that I actually

hadn't I hadn't realized till we were

working on this paper and I didn't

didn't know enough about it. But the

process of thermalization has been shown

to basically be undecidable in a in a

fundamental mathematical way. When a

system will thermalize is kind of like

the same as knowing when this computer

program will shut itself off and you

can't you can't and that depends upon

the complexity that you don't know in

advance. you don't know how much

complexity you have to understand and

whether you'll that was the argument you

gave at the end and so it it does relate

in a in a to me at least in a real way

to some of the directions that people

are going in quantravity which is that

if thermalization in the neighborhood of

black hole vent horizons and all of that

is important this may actually come back

and bite you in the face that the theory

may not be able to actually actually te

tell you something you're claiming

happens that you may not actually be

able to ever really fundamentally

calculate from first principles. So it

may have a practical significance, it

may not. But you're absolutely right. It

from it's it's like everything related

to the theory of everything. When it

comes to something, it's probably

doesn't matter very much. As again, as

as Frank used to say, I don't want a

theory of everything. I want a theory of

something. And uh and and and theories

of something are particularly useful.

Theories of everything are are some are

fun for some people to talk about. And

like you, it was fun for me to come up

with to to write something and basically

argue that even if you like it, it can't

happen. Anyway, I don't know if we

illuminated anything. Any comments based

on what I said or no?

Uh, no, just a question. The paper that

you mentioned about the thermonization,

was this the recent cubit paper or

No, no, no. There's another paper about

undecidability theorems.

Interesting. Maybe maybe you can send me

a reference because I'm also

Yeah, I'll send you a reference because

it was I didn't know about this field uh

enough to to and and my colleagues did

and I and I learned about it in in and

we reference it in the in in in some

point in our paper.

Well, I've I've looked at all these

papers where they claim there's a

physical quantity that we can't compute.

uh and and it's always the same and in

all the papers that I've seen is that

some physical quantity becomes infinite

like it's an infinite number of um

particles or something like that. So, so

it's not actually a real system. So, so

far this has always been the case which

is why I want to look at this paper.

Okay. Well, well, it' be worthwhile

looking at this one together. I I have

to say I've written some papers of mine

some of which have had a great impact

which I didn't understand at the time I

wrote them and fully And uh and and this

one is one that I that is I've have had

to come to grips with a lot. I'm not

sure I even still fully understand it.

I'm not sure this one is going to have

the same kind of impact as some of the

other ones. But it was um it was fun to

learn some of these things because I

tend to ignore I mean good and these

incompleteness serums are very important

but because as you say from a practical

perspective I'm someone who likes to

just do things and it it it often seems

a little ephemeral. So I hadn't I I I

haven't thought about them in detail and

this this forced me at least to think

about them in detail a little bit. Okay.

Anyway, thank you for the advertisement

and I appreciate it.

Well, a lot of people asked me to to

discuss this. So, um yeah.

Well, I've now learned if you put theory

of everything in a paper, everyone all

the all these people in the public want

to hear about it. But um anyway, now I

will talk about something that's much

more grounded, perhaps the most grounded

calculation in elementary particle

physics. The thing that made quantum

field theory put it on the map. One of

the two calculations is something called

the G minus 2. In this case, G minus 2

of the muon. The magnetic moment of the

muon. Electric elementary particles are

charged and they behave like they're

spinning. And a spinning charge will

have a magnetic moment. Now, they're not

really spinning, but quantum mechanics

says they behave like they're spinning

and they and they have a magnetic

moment. And because they have a you know

their charge, it's which is quantized.

Um and you know their spin which is

quantized you can get a number for the

value in some in in terms of some

fundamental unit called the bore

magneton of the of the magnetic moment

of elementary particles and in and if

all things were just simple the number

would be two doesn't m in those units or

a half or depend what unit you pick. Um

but of course um the great thing about

quantum field theory is it says that

that there actually all sorts of virtual

processes going on where a muon isn't

just a muon. It's emitting a photon and

for a while it's two muons and a muon

and antimu and on small scales lots of

strange things are happening. Virtual

processes that make the vacuum of

elementary particle physics strange and

many people say well that can't be true.

But of course the beauty the real beauty

in the 19 1947 4849 was that you could

actually make prediction you could

actually calculate results in spite of

all the craziness that's going on at

small scales and make predictions of

what the quantum mechanical

uh corrections are to this number two.

And it is the best prediction in all of

science in the sense that from

fundamental principles you can make a

prediction of a number to almost 12

decimal places. And the other amazing

thing which I which I think is even more

amazing as a theorist in particular is

that experimentals can measure certain

quantities if they're really clever to

12 decimal places. And there's nowhere

else in science where a fundamental

prediction

that's that's all the way to 12 decimal

places can be compared to theory to 12

decimal places from fundamental

principles. It's it's really in my mind

one of the most beautiful

most beautiful things in physics that

you can actually make a fundamental

prediction and test it and it really

tells us that empty space is really

weird. And so the gus2 of the muon is

incredibly important because it

validated the whole computational scheme

we use to try and understand elementary

particles and fields. But what makes it

even more interesting and what's made it

more interesting for 40 years is because

the quantum corrections are sensitive to

weird processes. If there's any new

physics like super symmetry which has

new elementary particles which can be

emitted by muons and then absorbed at

very small levels giving very small

effects.

But if you can measure something to 12

decimal places maybe you can look at

those effects. In fact many people have

argued why build a big accelerator when

you can look for effects that are that

come in at the part per billion level in

a very sensitive experiment and then do

the experiment. It's not a tabletop

experiment. These are big, you know,

many meter long storage rings of muons

and but but it's but it it's still less

expensive than building a new particle

collider. And so people have argued,

hey, maybe this is a great place for new

physics. In fact, I I've forgotten this

till till till just now, but I actually

wrote a paper that's that's sort of

significant on the one of the first

calculations of the G minus 2 of the

muon and super symmetry theories.

totally forgot that. But anyway, um and

uh and so um it's it's incredibly

interesting. Now, what made it more

interesting, at least from a fundamental

perspective, is it looked like there was

a difference between the theoretical

prediction and the observations at in

the 12th or in the 10th or 11th or 12th

decimal place. And if it wasn't so

beautiful, you'd say, who cares at the

10th, 11th, 12th decimal place? How am I

going to believe something like that?

But you can. If there was a known

well-defined disagreement statistically

between the theory and the observation

of 12 decimal places, it would tell us

there's something new and super symmetry

predicted such a a difference. And for a

while it looked like there was such a

difference and people have been doing

these G minus 2 experiments colleagues

of mine back when I was at Yale their

whole careers and they're beautiful.

They really are beautiful and very

clever experiments. Basically it's it's

the idea is basically if you have a

magnetic moment and you have a magnetic

field then then a then a little magnet

will process

and and the and the frequency of its

procession will depend upon how big a

magnet it is. That's the basic simple

idea. And so they take muons and store

them and see how they process in

magnetic fields and look at the

radiation they emit more or less or the

and it's like it's like NMR. It's very

precise and allows you to do

unbelievably good experiments. And the

most recent and people say final G minus

2 experiment was just done at Fermy Lab.

I think the magnet was taken from you

need large magnets for this was taken

from Brook Haven National Laboratory and

other accelerator in the United States

to Fairmy Lab and built and and and it

stored 300 billion muons. Remember,

muons only live one second long, but if

you if you accelerate them enough, they

live a little longer. And so, it's hard

to store them. And and there are 300

billion million 300 billion of them in

this big magnetic field. and and they

finally measured the answer is and the

correction to this number too is point

and I have to say it because they work

so hard to get the number 01165920705

and that's known with an accuracy of 127

parts per billion. The prediction is

0.00116592033

with a theoretical uncertainty of 540

parts in the last three numbers

and um and and unfortunately or

fortunately it agrees the theory and the

experiment agree. Now having said that

there's one little additional

technicality which which is kind of neat

which is how do you calculate how what's

the uncertainty in the theory? How how

dare you say theory has an uncertainty?

Well, the uncertainty comes in the fact

that muons which are only measure only

only experience the weak force and the

electromagnetic force nevertheless can

emit quarks which experience the strong

force and we can't calculate with the

strong force. So we have to make

approximations

and for a long time people made

approximations based on measurements and

they kept giving the wrong answer. And

what's really fascinating is well one

way to calculate is put it on a computer

in a lattice and make space a lattice

and then let the computer do detail

calculations on a lattice. And it turns

out those lattice calculations give

answers that are different than the

other calculations and agree with the

observations. And initially that was a

big deal and people didn't believe the

lattice calculations but now more people

have done them. And now it looks like

the computers are smarter than we are in

that sense or at least they're giving an

answer that's in agreement with the

observations. And why extrapolating

results from other data from other

experiments gives the wrong answer is

not yet known. But the bottom line is at

the end of all of this of this amazing

bit of work and real solid work by

theorists and solid work by experiments,

we made the best measurement that's ever

been made and the theory agrees. And

many people are disappointed of course

because there's no evidence for super

symmetry or anything else that people

thought might come out of it. But even

though it gives a null result, I wanted

to herald it because I think it's in

some ways science at its best. Turning

it over to you.

Well, that was a very nice speech. Um,

yeah. Um, so I'm kind of super

unsurprised by this because one could

see this coming for like 10 years or

something. Um yeah uh and and I I think

there's a deeper lesson in this um which

is that you said um some people have

argued you could replace to some

external building big bigger colliders

with making high precision measurements

lower energies. Uh and this is the sort

of problem that you run into because you

you'll have to do these calculations

very very precisely. So you put the

burden on the theory

whereas if you go to higher energies uh

if there is a signal it tends to be in

your face uh in sense you know once once

you have collected enough uh data you

know you have a bump that's there you go

basically

so so it's it's it's a somewhat cleaner

signature

um but yeah I mean it's a great

achievement um so I mean I can remember

people discussing the muon anomaly since

since I've been in physics basically

and so I'm quite happy we can lay it to

rest though. I have to say that I mean

you only told like the

the good side. Um so I I think it also

has like a negative side which is that

um you know from from the people that

I've personally um known and talked to I

think no one actually believed it was

some physics beyond the standard model

stuff like every everyone maybe that's a

generational thing like um put the blame

on the theoretical calculation or maybe

it's something to do with the kind of

people that I tend to know uh in but um

the idea that is kind of a new particles

for poor or it's extra dimensions uh and

I I had a student who wrote a paper on

that so on which I'm a co-author so I'm

also guilty right

um so um that attracts much more

attention like in the headlines so when

uh you know when when a when a newspaper

writes about um the wrote about the um G

minus 2 anomaly they write something

about evidence for a fifth force or

something like that stand out model is

broken, stuff like that.

Uh and and that that attracts a lot of

attention not just uh in the general

population but also among businesses and

I think it inspires them to make up more

stuff and so it becomes this

self-running thing.

And so I think that there's this

community dynamics uh which is something

that I really don't like. So um I I'm

quite happy that we can finally put this

to rest.

Well, I don't know whether I I warn you

nothing ever gets put to rest finally.

But anyway, um but but look, I I'm

sympathetic with you there. But I look,

I'm telling you,

I'm of a generation where I attended

many meetings where certain group of

people would say this proves super

symmetry. This anomaly in the Gus 2, we

here's our calculation. Look at the

anomaly. It's it proves that super

symmetry is true. And I'd hear it over

and over again. I didn't believe it. But

uh and so it's it's so that so it has

been useful to to close certain doors.

But I'll also say you're right, but you

come from a generation of people saying,

"Oh, an alom let me invent some crap,

some nonsense crap." And and that's

that's true. But I let me play the other

side of the devil's advocate. When I did

the this was in 1985

84 85 that we did the calculation of G

minus 2 in in super symmetric theories

and I it was what I liked it was one of

I liked it because it was most most

honest and certainly difficult. It was

the most complicated calculation I'd

ever done and probably have ever done

and I did it with people I could never

have done it alone. I did it with a

colleague Niska Sakai and and others but

he's a a Japanese physicist and required

keeping so many terms that I I couldn't

even write them neatly on a piece of

paper but he he would store them but

what I loved about it was it was a

concrete theory low energy super

symmetry was a concrete theory if you

wanted it to explain data it was highly

constrained in certain ways and

therefore if you if you worked hard

enough you could make a concrete

prediction And it was for me that's the

most satisfying that was an extremely

satisfying thing was to dig a concrete

theory which was beyond the standard

model at the time and and make a

concrete prediction.

It just required a lot of work. And so

for me I think I was trying to convince

my colleagues that I could be an honest

hardworking theoretical physicist. I had

a number of colleagues like Joe

Bulchinsky and Mark Weiser at the time

at Harvard and who basically said you

got to you got to prove yourself by

doing this calculation. So but so anyway

I I think it's nice in an era where

where where science allowed you to make

concrete predictions even for new for

well motivated new physics and then

compare them to theory. The problem is

what's happened is things have become so

difficult and so the standard model is

so good that people now are of a

generation and your generation I think

that tend to say oh let me just invent

the wildest things in the world to

explain everything with no with no

constraints on parameters and every new

every new result going to be new

dimensions or this or that and yeah I

have the same kind of feeling about all

of that which is um that it's much to do

about nothing I don't know if I'm over

overstressing But so anyway, I think G

minus 2 is a is an example of maybe we

can put it this way. G minus 2 is an

example of physics at its best and maybe

physics at its worst. But uh but um it's

wonderful when good theories can be

tested with good experiments and one

hopes that's true in all areas of

science. Let let maybe we can end on a

positive note. Is that acceptable?

Yeah, let's let let's talk about the

future. Uh I I want to talk about I want

to talk about AI and um maybe more

generally um want to hear what your

senses what AI is going to do to

physics. So um in the past months there

have been several developments with

selfimproving

AIS. Uh maybe the biggest headline was

for um came from Google Google deep mind

alpha evolve. uh it's basically uh an

artificially intelligent system that um

mutates code uh and then tests the code

and improves it. uh and so in and uh so

I was quite impressed by how successful

it was at improving the code and also

that it made some actually useful

discoveries like one of the things that

they found was a better way to multiply

big matrices basically that's more

efficient than anything that was

previously known so no humans can come

up with it so far and this thing kind of

did it wi within I don't know a couple

of days or god knows I didn't look

exactly at the numbers because we can

quibble about do we do we actually count

the time that they put in to to

construct the thing or something like

that but it doesn't really matter so it

found something really new uh and and

and it's quite interesting because

matrix multiplication is one of the

operations that you need to do in a lot

of those models right so it in some

sense you know it did improve itself

um or at least other AIs. Um and and so

this is one of the steps towards what's

been called um the intelligence

explosion formerly known as the

singularity when AIs can improve

themselves. So it becomes this

self-running exponential process um of

which we don't know the end point. hence

uh the singularity.

Um so I'm not saying that that we're

anywhere close to that point. Um but

what we're seeing right now is that

people are trying to use these

evolutionary approaches to to do exactly

that. And of course that isn't a new

idea, but it's is kind of like this idea

of the neural networks, you know, also

isn't really a new idea. Uh so so in the

end, I think what was the inflection

point for making this work was just

computing power. Okay, so I'm greatly

oversimplifying and you know everyone

who works on the topic will hate me for

saying this because of course there have

been many uh you know improvements on

the code and insights of blah blah blah

blah but basically what's driving the

whole thing is computing power and that

more people can use it and if more

people can use it they can work with it

they have more more insights and so

naturally like this makes me wonder like

what's it going to do to science? Um so

one thing that we're we're already

seeing right now is that these models

are becoming really like not large

language models but some of the stuff

that Deep Mind's working on and actually

also some of the the large language

models like um 03 like GPT03

they're becoming really good at math um

and so I think some mathematicians are

actually spooked by this you know that

they see like these models have eaten so

many proofs And it's kind of a language,

right? And and but it's a very clean

language like it's not contaminated. Of

course, they're you know they're wrongs,

but you know, mostly I think maths is

fine. So So they can learn this very

well. Uh and they they're becoming

really good at um you know, covering

topics all over the place. So, and I

think it's it's quite possible that a

lot of what what you know pure

mathematicians do right now

will in the soon future be replaced by

AIS the same way that we once had human

calculators that were crunching the

numbers, you know, with logarithmic

tables and slide rules and stuff like

this. and it it doesn't exist anymore

because we now have we we now have

computers, right?

And and so I think that that's what's uh

going to happen to mathematics. I'm not

saying that all mathematicians are going

to be replaced by I think that's a

little bit more in the future.

But but what's what comes after

mathematics? I think that'll be physics.

And so that's the question like how much

how much can it do? So you know we want

to hear your opinion about that.

Well, okay. That was a great

introduction. I I looked at the paper

that you referred to me and I and I

confess there, you know, I didn't

understand it all, but it I found it

fascinating and terrifying. And I'm not

really often terrified by AI, but it

and in some sense in reflecting

it's not surprising

that um

that a code which is actually what it

is. If if you it's like fing you say

quantum computers will help us

understand quantum mechanics because

they use quantum mechanics and that's

right. And if you think of a large

language mod module or these these these

AI things, they are based on, you know,

matrix multiplication and and and and so

it's not too surprising that something

that

that that has huge data inputs for

learning would be able to prove the very

process

by which it's learning because that's

that's what it's doing. I mean, if it's

using if it's using these things and if

it's evolving

to to find to to find techniques that do

it faster, well, it's the first thing

it's going to find is the techniques

that it's using. You know, I mean, it's

not too surprising that the first area

that it kind of evolves in to improve is

computations

uh of the type that are central to its

existence. But I it is it is it is uh

fascinating me because of course if it

improves code and and what's really nice

is then you use empirical evidence you

run the code and see if it does a better

job. So it's not some sort of internal

calculation. Then you're then you're

sort of doing science that it's doing

its own experiments

and and and

self programming improving

AI for me is is was the future that I've

always

uh I've always said once you have a

self-improving

AI self-programmable improving AI then I

fully expect it to develop much faster

than biological systems because there's

no constraints on it that way. And this

is this is an example of it. And and

you're absolutely right. the fact that

it

it found a better method if you wish for

matrix multiplication or or looking for

values of matrices or whatever um is

interesting and as I point out what

makes it more interesting is just the

fact that it it will improve your

ability to do this because if you could

find a more efficient mechanism the

whole thing about LLMs and all this is

they require huge computing resources

that's why you know I don't know whether

it's uh um Amazon or whatever or or

Google, they want to have their own

nuclear reactors for power because

because they need so much power to the

do these damn things. But if you can

improve as they said I think I think

they improve some algorithm and made it

25% more efficient I think is what it

would say. I think that's the number

which turns but it what's interesting is

it turns into ultimately requiring 1%

less energy

I think was if I read the article right

and that may sound trivial but if you're

using a hell of a lot of energy 1% less

energy means or 1% fewer number of

computers that can actually be real

significant dollars and it makes and so

small small changes in efficiency can be

huge monetarily and that could mean that

these systems if they become more

efficient or more accessible because I

think one of the other areas that you

know I'm not as worried about

singularities as people are um

you know these are these are good for

doing certain things as you point out

matrix multiplication and proving

mathematical theorems um but they're not

good at other things. So they're really

good at certain things. They're not

they're not general intelligences that

are good at everything. But one thing

but one of the other limitations on them

is they're not tabletop.

You need incredible computing resources

to do this. And it's not the kind of

thing that can be done cheaply or

mass-produced and and therefore that's

good in one sense and bad in another.

It's bad because the only companies that

can do it are the ones that have rich

enough like Google or whatever. And if

they're the only ones who have smart

enough LLMs, then they have a monopoly

on producing money. And one worries

about what's going to happen there if AI

and I think it's quite likely that

that's exactly what's going to happen.

That AI is in the province of the

richest companies and those richest

companies are not are going to put a lot

of people out of work and make a lot of

money and that money won't be

distributed among society. And that's

that's going to be a huge problem. But

having aside from those societal things,

the fact that the self-improving

programs do are working so well is

really impressive and a little bit

scary. And um and I do and I was going

to ask you it I think the alpha evolve

group is part of the group that was

using these techniques in chemistry.

Were they not part of the group that

used these techniques in chemistry that

caused the people in in deep mind to win

the Nobel Prize in chemistry? I looking

for new I think that they I think they

were the ones who were looking for new

configurations of of large scale

molecules and new and new uh chemical

techniques to find certain things.

Anyway, I think it's the same the same

the same group within deep mind but I

could be wrong.

It's possible. I didn't look exactly at

the at the names of the people who were

involved but but they had a they they

have a lot of alpha things like they

have alpha proof, they have alpha

evolve. um uh you know alpha fold or

something with the proteins and there

were were a couple of other alphas. So

it's it's probably if not exactly the

same group than like uh you know a lot

of overlap. Yeah, there's probably a lot

of but I I think the point that you made

and the point that's inherent here is

quant both quantum computing and AI in

my mind will be

the most immediate benefits will be for

detailed scientific calculations and and

and and and that's what's happening and

it you know in terms of the implications

for society as a whole I think a lot of

people it's science fiction a lot of

that but but for improving the ability

to do calculations and and and

predictions and that that's a good thing

in general, I think. So, so um that's

why I'm not as afraid of AI as as some

people because I think it'll help us. I

I I hope in the long run it'll help us

more than hurt us except for those

socopolitical things I talked about in

the end where certain companies

basically become

so ultimately infinitely rich that that

they don't know what to do with

anything. Yeah. So I I totally agree uh

on the economic problem, but I want to

come back to your lack of worry about

what's going to happen to physics. Maybe

that's because you're not a student at

the moment because think about your your

calculation with the with this super

symmetric contributions to the the mus

2.

I I think it's totally totally realistic

that in like two three years or maybe

even sooner that calculation can be done

by an AI. So you you can already be

done. I bet I bet if I bet I bet it

could be done much better by an AI and

and even maybe you don't even need an

AI. You just needed a faster computer.

But yeah, you're right. That is a

problem. Except let me say why I'm not

worried there.

I'm not worried there. Because I believe

that science including physics is

empirical. And so since I think that

what drives physics is not theorists

like you and me for the most part, it's

experiments.

that as long as we keep having building

new machines that give us new windows on

the universe, then calculations

calculations aren't all of science. It's

looking out and being surprised. And so,

um, since you can't train a AI on things

we haven't yet seen,

I I'm I have I have hope for science

because as long as we keep looking,

we'll discover things we haven't yet

seen. And that means um that um that AI

and theorists are not going to take you

know that that means cosmic job

security. What do you think of that?

Yeah. Uh no I agree with this for for

the time being. Um what you've just

you've just discarded all theoretical

physicists.

Well in a sense I have but I I think um

yeah okay. Uh and that's okay. I mean I

don't have a problem in the world if if

in if if in 10 years the best

theoretical physicists in the world are

computers. I mean from the point of view

of of knowledge that's a good thing you

know and and I don't think it'll be 10

years by the way. So I think I think

it's still worth doing your PhD. But um

but but you know again my qu I I

remember having this discussion with a

mutual friend of ours again Frank um a

long time ago and and the question I and

it'll be fascinating because the

question will then be what physics

questions are of interest to this

computer?

What can you learn from it? What you

know what what will will its interest be

the same as as ours? And what can I

learn from it? And if I'm interested in

learning about the universe and I have a

better colleague that's a computer,

well, okay, you know, any I mean the

world is going to change and that's

okay. Technology changes the world. It's

doesn't always make the world worse.

That's my point. You know, having a a a

phone has changed my life. Having a

smartphone has changed my life. In many

ways, it make it worse, but in a hell of

a lot of ways that make it better. And

so um you know to to to um to uh

you know I guess to say the future's

always dystopic because of the new

technology is not always clear.

Nevertheless, I think the future's

miserable. So I guess it doesn't matter

in the long term. In the long term I

most of my science has been to say in

the very long term the future is

miserable. So enjoy it while you can.

Now speaking of enjoy it while you can.

The last topic I always like, you know,

I like thinking about the origin of

life, but I decided not to do any

biology. I want to do physicsish kind of

stuff in this, but it's still a

fascinating result. You know, I've said

over and over again that astrobiology is

the most hyped area of science. It's not

most of it isn't even science. It's just

pure speculation about things we know

nothing about. It's fascinating to think

about and talk about, but these people

who claim to talk about it with

authority are are are usually bogus.

What we need is data and experiment and

there's so much we don't understand. And

what I liked about this little result

which otherwise wouldn't have gotten a

lot of hype is is a article entitled how

organic molecules survive in space. And

it's important because I have little

doubt that in the origin of life the

basic building blocks came to earth from

space.

um and and not not like you know I don't

I'm not not talking about panspermia

where you know cells and living beings

came from elsewhere maybe that's true

but but the organic materials and in

particular these organic materials

called polyyclic aromatic hydrocarbons

which are ubiquitous at the basis of

ultimately DNA and and life and and and

and what's interesting is these

materials exist in interstellar space

and molecular clouds are produced in

stars and then emitted in supernova

explosions and but the point is what

they're not supposed to exist because a

supernova explosion and the subsequent

stuff that goes out is a pretty drastic

event with lots of ultraviolet radiation

and lots of molecules colliding and it's

not a nice place for simple molecules to

survive. They get torn apart. They get

they get banged into and they vibrate

and and so initial calculations said

these small

uh polyyclic aromatic hydroarbons P aes

like something called indine which I

think is is something like H A what is

you know it's C C9 AH C98 H8 there's

just hydrogen it's a great place to find

hydrogen carb nine carbons eight

hydrogens these kind of things if

they're small then collision will break

them apart. If they're large, it turns

out that they're so large that the

vibrations are so complicated that they

won't break them apart. But these small

ones are observed and the James Web

Space Telescope observed them. And and

that may be vitally important for the

origin of life because if they didn't if

they weren't didn't survive longer than

we expect, then maybe the conditions on

Earth wouldn't have been uh uh uh good

for life to form. And what's kind of

neat, it's a simple result and a simple

experiment. Yes, they're observed at a

higher abundance. And people actually

did an experiment on Earth with a with a

in in a in in Switzerland, I think it

is, um or Sweden, one of those two

places, Sweden, Stockholm, with a

storage ring that would would actually

be able to take these things and

recreate the conditions of molecular

clouds and try and find out when they're

when you shine ultraviolet light at them

and when they bang together why they

survive. and and they survive because of

of a process um called recurrent

fluoresence which is kind of neat to me

for as a physicist. It basically says

you have this molecule vibrating and if

it vibrates enough it can actually

translate that vibrational energy which

would otherwise break it apart into

exciting some of its eternal electric

which then relax by emitting a photon.

So it can get rid of the energy that

would break it apart by turning it into

a kind of energy that you can emit it at

the speed of light and get away. And the

new calculations suggest it's possible.

And then when they actually did the

experiment, they discovered that indeed

these these this energy that would

otherwise break it apart relaxes five

times faster than you'd imagine and the

thing could survive. So it's a little

teeny result but physics and science

grows on baby steps and that little

teeny result may in may result in the

reason why you and I are here which I

find it fun. I it's not some grand

theory of everything. It's a little baby

result but maybe that little baby result

which happens by an obscure chemical or

physical processes uh called recurrent

fluoresence which I'd never heard of

before. Maybe that little baby result

will result in you and me. And I think

that's a kind of a beauty of science. So

I thought I'd end with that one.

Yeah. No, I find that super interesting.

Um I I like these studies because uh I I

I keep thinking if we figure out how

life on our planet starts, that that

will tell us something about the way and

possibility

of life to start on other planets as

well. Um, and I certainly hope that at

some point we'll find life on other

planets. And I don't mean microbial

lives. I want something that can walk

around.

Well, it'll be qu it'll be an AI. But

anyway, um, but in the long run, maybe

it will be, but yeah, of course, we all

want that. But but I think it's

important is to get, and this I want to

stress over and over again, is to get

there. Don't listen to people who say,

"Oh, I have evidence for life on Venus,

which turns out to be garbage." It's

these little teeny experiments which

tell us more about the fundamental

science of how life may have evolved on

Earth that in the end, in my mind, will

help us much more as scientists know

where to look for life elsewhere in the

universe than people making grand claims

about this or that. And so I it's real

solid science and and and that's

science. Not much of what not much of

what you read about in astrobiology is

in my mind science. I'll say it again.

I love talking to you, Sabina. It's um

and I and I love listening to you.

Obviously, I love talking at you, but I

also love listening to you and it's

always illuminating. And I I this was

actually I think this went much better

than I thought it would. So, thank you

very much. I hope you had fun.

Yeah. No, it was great.

[Music]

Hi, it's Lawrence again. As the Origins

podcast continues to reach millions of

people around the world, I just wanted

to say thank you. It's because of your

support, whether you listen or watch,

that we're able to help enrich the

perspective of listeners by providing

access to the people and ideas that are

changing our understanding of ourselves

and our world and driving the future of

our society in the 21st century. If you

enjoyed today's conversation, please

consider leaving a review on Apple

Podcast or Spotify. You can also leave

us private feedback on our website if

you'd like to see any parts of the

podcast improved. Finally, if you'd like

to access ad free and bonus content,

become a paid subscriber at

originsspro.org.

This podcast is produced by the Origins

Project Foundation as a nonprofit effort

committed to enhancing public literacy

and engagement with the world by

connecting science and culture. You can

learn more about our events, our travel

excursions, and ways to get involved at

origins project.org.