back to indexAlien Debate: Sara Walker and Lee Cronin | Lex Fridman Podcast #279
link |
I don't know what it's like to be an alien.
link |
I would like to know.
link |
Two alien civilizations coexisting on a planet,
link |
what's that look like exactly?
link |
When you see them and they see you,
link |
you're assuming they have vision,
link |
they have the ability to construct in 3D and in time.
link |
That's a lot of assumptions they're making.
link |
What human level intelligence has done is quite different.
link |
It's not just that we remember states
link |
that the universe has existed in before,
link |
it's that we can imagine ones that have never existed
link |
and we can actually make them come into existence.
link |
So you can travel back in time sometimes.
link |
You travel forward in time to travel back.
link |
The following is a conversation
link |
with Sarah Walker and Lee Cronin.
link |
They have each been on this podcast once before
link |
individually and now for their second time,
link |
they're here together.
link |
Sarah is an astrobiologist and theoretical physicist.
link |
Lee is a chemist and if I may say so,
link |
the real life manifestation of Rick from Rick and Morty.
link |
They both are interested in how life originates
link |
and develops both life here on earth and alien life,
link |
including intelligent alien civilizations
link |
out there in the cosmos.
link |
They are colleagues and friends
link |
who love to explore, disagree and debate
link |
nuanced points about alien life.
link |
And so we're calling this an alien debate.
link |
Very few questions to me are as fascinating
link |
as what do aliens look like?
link |
How do we recognize them?
link |
How do we talk to them?
link |
And how do we make sense of life here on earth
link |
in the context of all possible life forms
link |
that are out there?
link |
Treating these questions with the seriousness
link |
and rigor they deserve is what I hope to do
link |
with this conversation and future ones like it.
link |
Our world is shrouded in mystery.
link |
We must first be humble to acknowledge this
link |
and then be bold and diving in
link |
and trying to figure things out anyway.
link |
This is the Lex Friedman podcast.
link |
To support it, please check out our sponsors
link |
in the description.
link |
And now, dear friends, here's Sarah Walker and Lee Cronin.
link |
First of all, welcome back Sarah, welcome back Lee.
link |
You guys, I'm a huge fan of yours.
link |
You're incredible people.
link |
I should say thank you to Sarah
link |
for wearing really awesome boots.
link |
We'll probably overlay a picture later on,
link |
but why the hell didn't you dress up, Lee?
link |
No, I'm just kidding.
link |
This is me dressed up.
link |
You were saying that you're pink,
link |
that your thing is pink.
link |
My thing is black and white, the simplicity of it.
link |
When did the pink, when did it hit you
link |
that pink is your color?
link |
I became pink about, I don't know, actually, maybe 2017.
link |
Did you know me when you first?
link |
I think I met you pre pink.
link |
Yeah, yeah, so about 2017, I think.
link |
I just decided I was boring
link |
and I needed to make a statement
link |
and red was too bright, so I went pink, salmon pink.
link |
Well, I think you were always pink.
link |
You just found yourself in 2017.
link |
There's an amazing photo of him
link |
where there's like everybody in their black gown
link |
and he's just wearing the pink pants.
link |
Oh, that was at the Waggonen University.
link |
It's totally nuts.
link |
100 year anniversary, they got me to give the plenary
link |
and they didn't find that outfit for me,
link |
so they were all wearing these silly hats and these gowns
link |
and there was me dressed up in pink
link |
looking like a complete idiot.
link |
We're definitely gonna have to find that picture
link |
and overlay it, big full screen, slow motion.
link |
All right, let's talk about aliens.
link |
We'll find places we disagree and places we agree,
link |
life, intelligence, consciousness, universe, all of that.
link |
Let's start with a tweet from Neil deGrasse Tyson
link |
stating his skepticism about aliens wanting to visit Earth.
link |
Quote, how egocentric of us to think that space aliens
link |
who have mastered interstellar travel across the galaxy
link |
would give, pardon the French,
link |
would give a shit about humans on Earth.
link |
So let me ask you, would aliens care about visiting Earth,
link |
observing, communicating with humans?
link |
Let's take a perspective of aliens, maybe Sarah first.
link |
Are we interesting in the whole spectrum
link |
of life in the universe?
link |
I'm completely biased, at least as far as I think right now
link |
we're the most interesting thing in the universe.
link |
So I would expect based on the intrinsic curiosity
link |
that we have and how much I think that's deeply related
link |
to the physics of what we are,
link |
that other intelligent aliens would want to seek out
link |
examples of the phenomena they are
link |
to understand themselves better.
link |
And I think that's kind of a natural thing to want to do.
link |
And I don't think there's any kind of judgment
link |
on it being a lesser being or not.
link |
It's like saying you have nothing to learn
link |
by talking to a baby.
link |
You have lots to learn, probably more than you do
link |
talking to somebody that's 90.
link |
So yeah, so I think they absolutely would.
link |
So whatever the phenomena is that is human,
link |
there will be an inkling of the same kind of phenomena
link |
within alien species and they will be seeking that same.
link |
I think there's gotta be some features of us
link |
that are universal.
link |
And I think the ones that are most interesting,
link |
and I hope I live in an interesting universe,
link |
are the ones that are driven by our curiosity
link |
and the fact that our intelligence allows us to do things
link |
that the universe wouldn't be able to do
link |
without things like us existing.
link |
We're gonna define a lot of terms.
link |
One of them is interesting.
link |
That's a very interesting term to try to define.
link |
Ali, what do you think?
link |
Are humans interesting for aliens?
link |
Well, let's take it from our perspective.
link |
We want to go find aliens as a species quite desperately.
link |
So if we put the shoe on the other foot,
link |
of course we're interesting.
link |
But I'm wondering and assuming
link |
that we're at the right technological capabilities
link |
to go searching for aliens, then that's interesting.
link |
So what I mean is,
link |
if there needs to be a massive leap in technology
link |
that we don't have,
link |
how will aliens prioritize coming to Earth and other places?
link |
But I do think that they would come and find us
link |
because they'd want to find out about our culture,
link |
what things are universal.
link |
I mean, I'm a chemist.
link |
I would say, well, is the chemistry universal, right?
link |
Are the creatures that we're going to find
link |
making all this commotion,
link |
are they made of the same stuff?
link |
What does their science look like?
link |
Are they off planet yet?
link |
I guess there's, so I think that Neil deGrasse Tyson
link |
is being slightly pessimistic
link |
and maybe trying to play the tune
link |
that the universe is vast
link |
and it's not worth them coming here.
link |
I don't think that,
link |
but I just worry that maybe we don't have
link |
the ability to talk to them.
link |
We don't have the universal translator.
link |
We don't have the right physics,
link |
but sure, they should come.
link |
We are interesting.
link |
I want to know if they exist.
link |
It would make it easier if they just came.
link |
So again, I'm going to use your tweets
link |
like it's Shakespeare and analyze it.
link |
thinking about aliens, thinking about aliens.
link |
So how much do you think aliens
link |
are thinking about other aliens, including humans?
link |
So you said, we humans want to visit.
link |
Like we're longing to connect with aliens.
link |
Can you introspect that?
link |
Is that an obvious thing that we should be,
link |
like what are we hoping to understand
link |
by meeting aliens exactly?
link |
Asking as an introvert, it's like,
link |
I ask myself this all the time.
link |
Why go out on a Friday night to meet people?
link |
What are you hoping to find?
link |
I think the curiosity, so when I saw Sarah put that tweet,
link |
I think I answered it actually as well,
link |
which was we are thinking about trying to make contact.
link |
So they almost certainly are,
link |
but maybe there's a number of classes.
link |
There are those aliens that have not yet made contact
link |
with other aliens like us.
link |
Those aliens have made contact with just one other alien
link |
and maybe it's an anticlimax and slime, right?
link |
And aliens have made contact
link |
with not just one set of intelligent species, but several.
link |
That must be amazing actually.
link |
Literally there are some place in the universe,
link |
there must be one alien civilization.
link |
Let's not make contact with not one,
link |
but two other intelligent civilizations.
link |
So they must be thinking about it.
link |
There must be entire degree courses on aliens,
link |
thinking about aliens and universal cultural norms.
link |
Do you think they will survive the meeting?
link |
And by the way, Lee did respond saying,
link |
that's all the universe wants.
link |
So Sarah said, thinking about aliens, thinking about aliens.
link |
Lee said, that's all the universe wants.
link |
And then Sarah responded, cheeky universe we live in.
link |
So cheeky is a cheeky version of the word interesting,
link |
all of which we'll try to define mathematically.
link |
Cheeky might be harder than interesting.
link |
Because there's humor in that too.
link |
I think there's a mathematical definition of humor,
link |
but we'll talk about that in a bit.
link |
Yeah, sure there is, yeah.
link |
So if you're a graduate student alien
link |
looking at multiple alien civilizations,
link |
do you think they survive the encounters?
link |
I think there's a tendency to anthropomorphize
link |
a lot of the discussions about alien life,
link |
which is a really big challenge.
link |
So usually when I'm trying to think about these problems,
link |
I don't try to think about us as humans,
link |
but us as an example of phenomenon
link |
that exists in the universe that we have yet to explain.
link |
And it doesn't seem to be the case
link |
that if I think about the features,
link |
I would argue are most universal about that phenomenon,
link |
that there's any reason to think
link |
that a first encounter with another lineage
link |
or example of life would be antagonistic.
link |
And I think there's this kind of assumption,
link |
I mean, going back to Neil deGrasse Tyson's quote,
link |
I mean, it kind of bothers me because there's a,
link |
I mean, I'm a physicist,
link |
so I know we have a lot of egos
link |
about how much we can describe the world,
link |
but that there's this like,
link |
because we understand fundamental physics so well,
link |
we understand alien life and we can kind of extrapolate,
link |
and I just think that we don't.
link |
And the quest there is really, you know,
link |
really to understand something totally new
link |
about the universe, and that thing just happens to be us.
link |
There's something else more profound.
link |
I think Neil was just being, again,
link |
he's just trying to stir the pot.
link |
I would say from a contingency point of view,
link |
I want to know how many ways
link |
does the universe build structures, build memories, right?
link |
And then I want to know if those memories
link |
can interact with each other.
link |
And if you have two different origins of life
link |
and then origins of intelligence,
link |
and then these things become conscious,
link |
surely you want to go and talk to them
link |
and figure out what commonalities you share.
link |
And it might be that we're just unable to conceive
link |
of what they're going to look like.
link |
They're just going to be completely different,
link |
you know, infrastructure,
link |
but surely we'll want to go and find out a map.
link |
And surely curiosity is a property
link |
that evolution has made on earth.
link |
And I can't see any reason
link |
that it won't happen elsewhere
link |
because curiosity probably exists
link |
because we want to find innovations in the environment.
link |
We want to use that information to help our technology.
link |
And also curiosity is like planning for the future.
link |
Are they going to fight us?
link |
Are we going to be able to trade with them?
link |
So I think that Neil's just, I don't know,
link |
maybe, you know, I mean, give a shit.
link |
That's really, I think that's really down on earth, right?
link |
How would aliens categorize humans, do you think?
link |
So let's put it the other way around.
link |
Maybe, no, no, no.
link |
Maybe we could, the thing is a bit odd, right?
link |
Look at Instagram, Twitter,
link |
all these people taking selfies.
link |
I mean, does the universe
link |
is the ultimate state of consciousness,
link |
thinking beings that take photographs themselves
link |
and upload them to an internet
link |
with other thinking beings looking at each other's photos.
link |
So I think that they will be.
link |
What's wrong with that?
link |
I did not say there was anything wrong with it.
link |
It's consciousness manifested at scale.
link |
Selfies, Instagram.
link |
It's like the mirror test at scale.
link |
Yeah, I do think that curiosity
link |
is really the driving force
link |
for why we have our technology, right?
link |
If we weren't curious, we wouldn't go out, left the cave.
link |
so I think that Neil's got it completely wrong, in fact.
link |
Actually, of course they'd want to come here.
link |
It doesn't mean they are coming here.
link |
We've seen evidence for that.
link |
I guess we can argue about that, right?
link |
But I think that we want,
link |
I desperately, and I know that Sarah does too,
link |
but I won't speak for you, you're here.
link |
I desperately want to have missions
link |
to look for life in the solar system right now.
link |
I want to map life over the solar system.
link |
And then I want to understand how we can go
link |
and find life as quickly as possible at the nearest stars,
link |
and also at the same time do it in the lab,
link |
just to compensate, you know?
link |
Yeah, just one more point on this.
link |
If you think about sort of what's driven
link |
the most features of our own evolution as a species,
link |
and try to map that to alien species,
link |
I always think like optimism is what's gonna get us furthest.
link |
And so I think a lot of people always think
link |
that it's like war and conflict is gonna be the way
link |
that alien species will expand out into the cosmos.
link |
But if you just look at how we're doing it
link |
and how we talk about it,
link |
so is our future in space is always, you know,
link |
built from narratives of optimism.
link |
And so it seems to me that if intelligence does get out
link |
in the universe, that it's gonna be more optimism
link |
and curiosity driving it than war and conflict,
link |
because those things end up crushing you.
link |
So there might be some selective filter.
link |
Of course, this is me being an optimist.
link |
I'm a half full kind of person, but.
link |
Is it obvious that curiosity, not obvious,
link |
but what do you think?
link |
Is curiosity a more powerful force in the universe
link |
than violence and the will to power?
link |
So, because you said you framed curiosity as a way
link |
to also plan on how to avoid violence,
link |
which is an interesting framing of curiosity.
link |
But I could also argue that violence
link |
is a pretty productive way to operate in the world,
link |
which is like, that's one way to protect yourself.
link |
The best defense is offense.
link |
I'm not qualified to answer this, but I'll have a go.
link |
I think violence, let's not talk about violence.
link |
That's the summary of this podcast.
link |
I would, yeah, maybe, I would, let's not call it violence,
link |
but I call it erasure.
link |
So if you think about the way evolution works,
link |
or the way, obviously talk about assembly theory,
link |
So if you say you build,
link |
curiosity allows you to open up avenues, new graphs, right?
link |
So new features you can play.
link |
The ability to erase those things allows you to start again
link |
and do some pruning.
link |
So the universe, I think curiosity gets you furthest.
link |
Curiosity gets you rockets that land.
link |
It gets you robots that can make drugs.
link |
It gets you poetry and art and communication.
link |
And then, I often think wouldn't it be great in bureaucracy
link |
to have another world war, not literally a world war now,
link |
please no world war, but the equivalent
link |
so we get remove all the admin bureaucracy, right?
link |
All the admin violence, get rid of it and start again.
link |
Do you know what I mean?
link |
Because you get layers and you get redundant systems built.
link |
So actually a reset, let's not call it violence,
link |
a reset in some aspects of our culture
link |
and our technology allows us to then build
link |
more important things about the,
link |
because how many cookies do I have to click on?
link |
How many extra clicks do I have in the future of my life
link |
that I could remove and a bit of a reset
link |
would allow us to start again.
link |
And maybe that's how I suppose our encounter aliens will be.
link |
Maybe they will fight with us and say,
link |
oh, we're not as excited by you as we thought.
link |
We'll just get rid of you.
link |
So they might want to reset Earth.
link |
To be like, let's see how the evolution runs again.
link |
This seems like they've, there's nothing new happening here.
link |
They're observing for a while.
link |
This is just not, let's keep it more fun.
link |
Let's start with the fish again.
link |
I like how you equated violence to resetting your cookies.
link |
I suppose that's the kind of violence.
link |
In this modern world where words are violence,
link |
resetting cookies is the kind of violence.
link |
I don't know where that came from.
link |
I'm completely, yeah.
link |
That's poetic, really.
link |
Okay, so let's talk about life.
link |
What is the line between life and non life?
link |
And maybe at any point,
link |
we can pull in ideas of assembly theory.
link |
Like how do we start to try to define life?
link |
And for people listening,
link |
so Sarah identifies as a physicist
link |
and Lee identifies as a chemist.
link |
Of course, they are very interdisciplinary in nature
link |
in general, but so what is life?
link |
I love asking that question
link |
because it's so absurdly big.
link |
I know, I love it.
link |
It's my absolute favorite question in the whole universe.
link |
So I think I have three ways of describing it right now.
link |
And I like to say all three of them
link |
because people latch onto different facets of them.
link |
And so the whole idea of what Lee and I are trying to work on
link |
is not to try to define life,
link |
but to try to find a more fundamental theory
link |
that explains what the phenomenon we call life.
link |
And then it should explain certain attributes
link |
and you end up having a really different framing
link |
than the way people usually talk.
link |
So the way I talk about it three different ways.
link |
Life is how information structures matter
link |
across space and time.
link |
Life is, I don't know, this one's from you actually,
link |
simple machines constructing more complex machines.
link |
And the other one is the physics of existence,
link |
so to speak, which is life is the mechanism
link |
the universe has to explore the space of what's possible.
link |
That's my favorite.
link |
So can I, yeah, yeah, can I add on to that?
link |
Okay, can you say the physics one again?
link |
Physics of existence.
link |
Yeah, the physics of existence.
link |
I don't know what to call it.
link |
If you think of all the things that could exist,
link |
only certain things do exist.
link |
And I think life is basically the universe's mechanism
link |
of bringing things into physically existing
link |
in the moment now.
link |
Yeah, and what's another one?
link |
So we were debating this the other day.
link |
So if you think about a universe that has nothing in it,
link |
that's kind of hard to conceive of, right?
link |
Because, and this is where physicists really go wrong.
link |
They think of a universe with nothing in it, they can't.
link |
But nonexistence is really hard to think.
link |
Nonexistence, yeah.
link |
And then you think of a universe with everything in it,
link |
that's really hard.
link |
And you just have this white blob, right?
link |
It's just everything.
link |
But the fact we have discrete stuff in the universe,
link |
Beyonce planets, so you've got stars, space, planet stuff,
link |
right, the boring stuff, but I would define life
link |
or say that life is where there are architectures,
link |
any architectures, and we should stop fixating
link |
on what is building the architectures to start with.
link |
And the fact that the universe has discrete things
link |
and it is completely mind blowing.
link |
If you think about it for one second,
link |
the fact there's any objects at all,
link |
and there's, because for me, the object is a proxy
link |
for a machine that built it,
link |
some information being moved around,
link |
actuation, sensing, getting resource,
link |
and building these objects.
link |
So for me, everyone's been obsessing about the machine,
link |
but I'm like, forget the machine, let's see the objects.
link |
And I think in a way that assembly theory,
link |
we realized maybe a few months ago
link |
that assembly theory actually does account for the soul
link |
and the objects, not mystically like say,
link |
Sheldrake's morphic resonance,
link |
or Leibniz's monadology, seeing souls in things.
link |
But when you see an object, and I've said this before,
link |
but this object is evidence of thought,
link |
and then there's a lineage of those objects.
link |
So I think what is fascinating is that,
link |
you put it much more elegantly,
link |
but the barrier between life and non life
link |
is accruing enough memories to then actuate.
link |
So what that means is there are contingency,
link |
there are things that happen in the universe get trapped,
link |
these memories then have a causal effect on the future.
link |
And then when you get those concentrated in a machine,
link |
and you're actually able in real time,
link |
able to integrate the past, the present with the future,
link |
and do stuff, that's when you are most alive.
link |
You being the machine.
link |
Wait a minute, why is the object,
link |
so one of the ways to define life,
link |
that Sarah said, is simple machines
link |
creating complex machines.
link |
So there's a million questions there.
link |
So how the hell does a simple machine
link |
create a complex machine?
link |
So this is what we were talking about at the beginning,
link |
is you have the minimum replicator, so a molecule.
link |
So this is what I was trying to convince Sarah
link |
of the mechanism get there years ago, I think,
link |
but then you've been building on it and saying,
link |
you have a molecule that can copy itself,
link |
but then there has to be some variability,
link |
otherwise it's not gonna get more functional.
link |
So you need to add bits on.
link |
So you have a minimum molecule that can copy itself,
link |
but then it can add bits on,
link |
and that can be copied as well,
link |
and those add ons can give you additional function,
link |
to be able to acquire more stuff to exist.
link |
So existence is weird,
link |
but the fact that there is existence is why there is life,
link |
and that's why I realized a few days ago
link |
that there must be, that's why alien life
link |
must be everywhere, because there is existence.
link |
Is there a conservation of cheeky stuff happening?
link |
So how can you keep injecting more complex things?
link |
Doesn't the machine that creates the object
link |
need to be as or more powerful than the things it creates?
link |
So how can you get complexity from simplicity?
link |
So the way you get complexity from simplicity
link |
is that you, I'm just making this up,
link |
but this is kind of my notion,
link |
and you have a large volume of stuff,
link |
so you're able to get seeds, if you like,
link |
random cues from the environment.
link |
So you just use those objects
link |
to basically write on your tape, ones and zeros, whatever.
link |
And that is necessarily rich, complex, okay?
link |
But it has a low assembliness,
link |
but even though it has a high assembly number,
link |
we can talk about that.
link |
But then when you start to then integrate
link |
that all into a smaller volume, as over time,
link |
and you become more autonomous,
link |
you then make the transition.
link |
I don't know what you think about that.
link |
I think the easiest way to think about it is actually,
link |
which I know is a concept you hate, but I also hate,
link |
which is entropy, but people are more familiar with entropy
link |
than what we talk about in assembly theory.
link |
And also the idea that, like, say physics as we know it
link |
involves objects that don't exist across time,
link |
or as we would say, low memory objects.
link |
So one of the key distinctions that is...
link |
Low memory objects.
link |
Yeah, so physics is all...
link |
Physicists are low memory objects.
link |
Low memory objects.
link |
Physicists are creators of low memory objects
link |
or manipulators of low memory objects.
link |
It's a very nice way of putting it.
link |
Okay, sorry, go ahead.
link |
Yeah, no, it's okay.
link |
Sorry to keep interrupting.
link |
No, no, no, it's fine.
link |
I like it too, it's very funny.
link |
But I think it's a good way of phrasing it
link |
because I think this kind of idea we have
link |
in assembly theory is that, you know,
link |
physics as we know it has basically removed time
link |
as being a physical observable of an object.
link |
And the argument I would make is that
link |
when you look at things like water bottles or us,
link |
we're actually things that exist
link |
that have a large extent in time.
link |
So we actually have a physical size and time,
link |
and we measure that with something called
link |
the assembly index in molecules,
link |
but presumably everyone should have sort of a,
link |
do you want to explain what assembly?
link |
Yeah, let's, you know what?
link |
Let's step back and start at the beginning.
link |
What is assembly theory?
link |
Lee sent me some slides.
link |
There's a big sexy paper coming out probably.
link |
Maybe, I don't know.
link |
We've almost finished it.
link |
Almost, almost finished it.
link |
That's also a summary of science.
link |
We're almost done.
link |
Well, no, no, we're almost done.
link |
It's the history of science.
link |
We are ready to start an interesting discussion
link |
You're the machine that created the object,
link |
and we'll see what the object takes us.
link |
All right, so what is assembly theory?
link |
Yeah, well, I think the easiest way
link |
for people to understand it is to think about
link |
assembly and molecules,
link |
although the theory is very general.
link |
It doesn't just apply to molecules.
link |
And this was really Lee's insight,
link |
so it's kind of funny that I'm explaining it, but.
link |
Okay, all right, I'm ready, I'm ready.
link |
You have to tell me where I get the check marks minus,
link |
It's your theory as well.
link |
But imagine a molecule,
link |
and then you can break the molecule apart
link |
into elementary building blocks.
link |
They happen to be bonds.
link |
And then you can think of all the ways,
link |
for molecular assembly theory,
link |
you can think of all the ways
link |
of building up the original molecule.
link |
So there's all these paths that you can assemble it.
link |
And the sort of rules of assembly is
link |
you can use pieces that have been generated already.
link |
So it has this kind of recursive property to it.
link |
And so that's where kind of memory
link |
comes into assembly theory.
link |
And then the assembly index is
link |
the shortest path in that space.
link |
So it's supposed to be the minimal amount of history
link |
that the universe has to undergo
link |
in order to assemble that particular object.
link |
And the reason that this is significant is
link |
we figured out how to measure that
link |
with a mass spec in the lab.
link |
And we had this conjecture
link |
that if that minimal number of steps
link |
was sufficiently large,
link |
it would indicate that you required a machine
link |
or a system that had information
link |
about how to assemble that specific object
link |
because the combinatorial space of possibilities
link |
is getting exponentially large
link |
as the assembly index is increasing.
link |
So it's just, sorry to interrupt,
link |
so that means there's a sufficiently high assembly index
link |
that if observed in an object
link |
is an indicator that something lifelike created it
link |
or is the object itself lifelike?
link |
But you might want to make the distinction
link |
that a water bottle is not life,
link |
but it would still be a signature
link |
that you were in that domain of physics
link |
and that I might be alive.
link |
So there will be potentially a lot of arguments
link |
about where the line, at which assembly index
link |
does interesting stuff start to happen.
link |
The point is we can make all the arguments,
link |
but it should be experimentally observable
link |
and Lee can talk more about that part of it.
link |
But the point I want to make about it is
link |
there was always this intuition that I had
link |
that there should be some complexity threshold
link |
in the universe above which you would start to say
link |
whatever physics governs life actually becomes operative.
link |
And I think about it a little bit
link |
like we have Planck's constant,
link |
and we have the fine structure constant.
link |
And then this sort of assembly threshold
link |
is basically another sort of potentially constant of nature.
link |
It might depend on the specific features of the system,
link |
but which we debate about sometimes.
link |
But then when you're past that,
link |
you have to have some other explanation
link |
than the current explanations we have in physics,
link |
because now you're in high memory.
link |
The things actually require time for them to exist
link |
and time becomes a physical variable.
link |
The path to the creation of the object is the memory.
link |
So you need to consider that.
link |
Yeah, but the point is that's a feature of the object.
link |
So when I think of all the things in this room,
link |
we see the projection of them as a water bottle,
link |
but assembly theory would say that this is a causal graph
link |
of all the ways the universe can create this thing.
link |
That's what it is as an object.
link |
And we're all interacting, a causal graph.
link |
And most of the creativity in the biosphere
link |
is because a lot of the objects that exist now
link |
are huge in their structure across time.
link |
Four billion years of evolution to get to us.
link |
Is it possible to look at me
link |
and infer the history that led to me?
link |
If you, you as an individual might be hard.
link |
You as a representative of a population of objects
link |
that have high assembly
link |
with similar causal history and structure
link |
that you can communicate with, i.e. other humans,
link |
you can infer a lot probably.
link |
Yeah, also with them.
link |
Which we do genomically even.
link |
I mean, it's not like,
link |
we have a lot of information in us,
link |
we can reconstruct histories from.
link |
Assembly is saying something slightly deeper.
link |
Yeah, one thing to add.
link |
I mean, it's not just about the object,
link |
but the objects occur
link |
and not just objects with high assembly number,
link |
because you can have random things
link |
that have a high assembly number,
link |
but they must have,
link |
there must be a number of identical copies.
link |
So you know you're getting away from the random,
link |
because you could take a snapshot.
link |
This is why, it's not I hate entropy,
link |
I love entropy when used correctly,
link |
but it's about the problem of entropy,
link |
you have to have a labeler.
link |
And so you can label the beginning and the end,
link |
the start and the finish, you know,
link |
where what you can do in assembly is say,
link |
oh, I have a number of objects in abundance.
link |
They all have these features.
link |
And then you can infer.
link |
And one of the things that we debated a lot,
link |
particularly during lockdown,
link |
because I almost went insane trying to crush the,
link |
produce the assembly equation.
link |
So we came up with the assembly equation.
link |
I had, just imagine this.
link |
So you have a string where,
link |
actually it makes me sick trying to remember it.
link |
It was so, it did my head in for a long time.
link |
Yeah, because I couldn't,
link |
so if you just have a string of say words,
link |
say, you know, a series of words, series of letters.
link |
So you just have A, A, A, B, B, B, C, C, C, D, D, D.
link |
And you find that object and you just have four A's,
link |
four B's, four C's, four D's together, boom.
link |
Then, and that you measured that.
link |
So you physically measured that string of letters.
link |
Then what you could do is you can infer sub graphs
link |
of maybe the four A's, the four B's,
link |
the four C's and the four C's,
link |
but you don't see them in the real world.
link |
You just infer them.
link |
And I really got stuck with that
link |
because there's a problem to try and work out
link |
what's the difference between a long,
link |
you know, a physical object
link |
and this assembly space of the objects
link |
that we realized the best way to put that is infer in time.
link |
So although we can't infer your entire history,
link |
we know at some point the four A's were made,
link |
the four B's was made, the four C's were made,
link |
the four D's were made and they all got added together.
link |
And that's one really interesting thing
link |
that's come out of the theory,
link |
but the killer when we knew we were going beyond
link |
beyond standard complexity theories,
link |
but incredibly successful is that we realized
link |
we could start to measure these things
link |
for real across domains.
link |
So the assembly index is actually an intrinsic property
link |
of all stuff that you can break into components,
link |
particularly molecules are good
link |
because you can break them up
link |
into smaller molecules, into atoms.
link |
The challenge will be making that more general
link |
across all the domains,
link |
but we're working on it right now
link |
and I think the theory will do that.
link |
So components, domains,
link |
so you're talking about basically measuring
link |
the complexity of an object in what,
link |
biology, chemistry, physics,
link |
that's what you mean by domains.
link |
Complexity of tests.
link |
Sociology. Complexity of computers.
link |
Complexity of memes, you know.
link |
What is that, ideas?
link |
Yeah, I mean, so one of the.
link |
Ideas are objects in assembly theory.
link |
They're physical things.
link |
They're just features of the causal graph.
link |
I mean, the fact that I can talk to you right now
link |
is because we're exchanging structure
link |
of our assembly space.
link |
So conversation is the exchanging structures
link |
in assembly space.
link |
What is assembly space?
link |
When I started working on origins of life,
link |
I was writing about something called top down causation,
link |
which a lot of like philosophers are interested in
link |
and people that worry about the mind body problem.
link |
But the whole idea is, you know,
link |
if we have, you know, the microscopic world of physics
link |
is causally complete,
link |
it seems like there's no room for higher level causes
link |
like our thoughts to actually have any impact on the world.
link |
And that didn't, that seems problematic
link |
when you get to studying life and mind
link |
because it does seem that quote unquote,
link |
emergent properties do matter to matter.
link |
And then there's this other sort of paradoxical situation
link |
where information looks like it's disembodied.
link |
So we talk about information,
link |
like it can just move from any physical system
link |
to any other physical system.
link |
And it doesn't require,
link |
like you don't have to specify anything about the substrate
link |
to talk about information.
link |
And then there's also the way we talk about mathematics
link |
is also disembodied, right?
link |
Like the platonic world of forms
link |
and I think all of those things are hinging
link |
that we really don't know how to think about abstractions
link |
as physical things.
link |
And really, I think what assembly theory is pointing to
link |
is what we're missing there is the dimension of time.
link |
And if you actually look at an object
link |
being extended across time,
link |
what we call information and the things that look abstract
link |
are things that are entangled
link |
in the histories of those objects.
link |
They're features of the overlapping assembly space.
link |
So they look abstract because they're not
link |
part of the current structure,
link |
but they're part of the structure
link |
if you thought about it as like the philosophical concept
link |
of a hyperobject, an object that's too big in time
link |
for us to actually to resolve.
link |
And so I think information is physical.
link |
It's just physical in time, not in space.
link |
Too hyperobject, too difficult for us to resolve.
link |
So we're supposed to think about of life
link |
as this thing that stretches through time
link |
and there's a causation chain that led to that thing.
link |
And then you're trying to measure something
link |
with the assembly index about properties of that.
link |
The assembly index is the ordering,
link |
like you could think of it as like a partial ordering
link |
of all the things that can happen.
link |
So in thermodynamics, we coarse grain things
link |
by temperature and pressure.
link |
In assembly theory, we coarse grain
link |
by the number of copies of an object
link |
and the assembly index, which is basically,
link |
if you think of the space of all possible things,
link |
it's like a depth of how far you've gone into that space
link |
and how much time was required to get there.
link |
In the shortest possible version.
link |
The shortest possible version.
link |
Not average, because can't you just 3D?
link |
You're gonna kill me with that question.
link |
Not 3D, can't you always 3D print the thing?
link |
Let's like stab him in the heart.
link |
No, because I had such a fight.
link |
So Sarah's team and my team are writing this paper
link |
at the moment and.
link |
I think we kind of share the, at the beginning,
link |
you were like, no, that's not right.
link |
Oh yeah, that's right.
link |
And we're doing this for a bit.
link |
And then the problem is when you build a theory
link |
and build the intuition,
link |
there's some certain features, right,
link |
of the theory that almost felt like
link |
I was being religious about saying,
link |
right, you have to do this.
link |
A good assembly theorist does this, does this, does this.
link |
And Sarah's postdoc, Daniel, and my postdoc, Abhishek,
link |
and they were both.
link |
They're both brilliant.
link |
They're brilliant, but they were like,
link |
no, we don't buy that.
link |
And I was like, it is, they were like,
link |
well, Lee, actually, I thought you were the first
link |
to say that, you know, you can't,
link |
if you can't explain it, it doesn't,
link |
and you can't do an experiment that doesn't exist.
link |
And that saved me.
link |
And I said to Abhishek,
link |
Abhishek's my postdoc in Glasgow,
link |
Daniel is Sarah's postdoc in ASU.
link |
I was like, I have the experimental data.
link |
So when I basically take the molecules
link |
and chop them up in the mass spec,
link |
the assembly number is never the average.
link |
It's always the shortest.
link |
It's an intrinsic property.
link |
And then the penny drop for Abhishek said, okay.
link |
So I had these things that we had to believe
link |
to start with or to trust,
link |
and then we'd done the math and it comes out.
link |
And they now have the shortest path, actually.
link |
It's up, it explains why the shortest path.
link |
Here's why the shortest path is important, not the average.
link |
The shortest path needs you to identify
link |
when the universe has basically got a memory,
link |
So what you want to be able to do is to say,
link |
what is the minimum number of features
link |
that I want to be able to see in the universe?
link |
When I find those features,
link |
I know the universe has had a coherent memory
link |
and is basically alive.
link |
And so that gives you the lower bound.
link |
So that's like, of course there's going to be other paths.
link |
We can be more ridiculous, right?
link |
We can have other parts, but it's just the minimum.
link |
So probabilistically at the beginning,
link |
because assembly theory was built
link |
as a measure for biosignatures, I needed to go there.
link |
And then I realized it was intrinsic.
link |
And then Sarah realized it was intrinsic
link |
and these hyperobjects were coming.
link |
And we were kind of fusing that notions together.
link |
And then the team were like, yeah,
link |
but if I have enough energy and I have enough resources,
link |
I might not take the shortest path.
link |
I might go a bit longer.
link |
I might take a really long path
link |
because it allows me then to do something else.
link |
So what you can do is, let's say
link |
I've got two different objects, A and B,
link |
and they both have different shortest paths to get them.
link |
But then if you want to make A and B together,
link |
they will have a compromise.
link |
So in the joint assembly space, that might be an average,
link |
but actually it's the shortest way
link |
you can make both A and B
link |
with a minimum amount of resource in time.
link |
So suddenly you then layer these things up.
link |
And so the average becomes not important,
link |
but as you literally overlap those sets,
link |
you get a new shortest path.
link |
And so what we realized time and time again
link |
when we're doing the math,
link |
the shortest path is intrinsic, is fundamental,
link |
and is measurable, which is kind of mind blowing.
link |
So what we're talking about, some basic ingredients,
link |
maybe we'll talk about that, what those basic ingredients
link |
could be and how many steps, when you say shortest path,
link |
how many steps it takes to turn those basic ingredients
link |
into the final meal.
link |
So how to make a, what's the shortest way to make a pizza?
link |
And a pizza and a pie together.
link |
So there's a lot of ways.
link |
There's the shortest way,
link |
and then you take the full spectrum of ways
link |
and there's probably an average duration
link |
for a noob to make an apple pie.
link |
Is the average interesting still?
link |
If you measure the average length of the path
link |
to assemble a thing, does that tell you something
link |
about the way nature usually does it?
link |
Versus something fundamental about the object,
link |
which I think is what you're aiming at
link |
with the assembly index.
link |
Yeah, I mean, look, we all have to quantify things.
link |
The minimum path gives you the lower bounds.
link |
You know you're detecting something.
link |
You know you're inferring something.
link |
The average tells you about really how the objects
link |
are existing in the ecosystem or the technology.
link |
And there has to be more paths explored
link |
because then you can happen upon other memories
link |
and then condense them down.
link |
I'm not making too much sense, but if you look and say,
link |
let's just say, I mean, maybe we're gonna get
link |
to alien civilizations later, right?
link |
But I would argue very strongly
link |
that alien civilization A and alien civilization B,
link |
they're different assembly spaces.
link |
So they're kind of gonna be a bit messed up
link |
if they happen to come one another,
link |
only when they find some joint overlap in their technology,
link |
because if aliens come to us and they don't share
link |
any of the causal graph we've showed,
link |
but hopefully they share the periodic table
link |
and bonds and things,
link |
that we're gonna have to really think about the language
link |
to talk to us aliens by inferring,
link |
by using assembly theory to infer their language,
link |
their technology, and other bits and bobs.
link |
And the shortest path will help you do that quickly.
link |
All right, so all aliens in this causality graphs
link |
have a common ancestor in the...
link |
If the building blocks are the same,
link |
which means they live in the same universe as us.
link |
So this is the assumption.
link |
It depends on how far back in time you go, though.
link |
But the universe has all the same building blocks.
link |
And we have to assume that.
link |
So at least there's not different classes
link |
of causality graphs, right?
link |
The universe doesn't just say like,
link |
here you get the red causality graph,
link |
and you get the blue one.
link |
These basic ingredients,
link |
and they're geographically constrained,
link |
or constrained in space or time, or something like that.
link |
They're constrained in time
link |
because only by the virtue of the fact
link |
that you need enough time to have passed
link |
for some things to exist.
link |
So the universe has to be big enough in time
link |
So just one point on the shortest path
link |
versus the average path,
link |
which I think we'll get to this,
link |
is you had a nice way of saying it's like
link |
the minimal compression is the shortest path
link |
for the universe to produce that.
link |
But it's also like the first time
link |
in the ordering of events
link |
that you might expect to see that object.
link |
But the average path tells you something
link |
about the actual steps that were realized,
link |
and that becomes an emergent property
link |
of that object's interaction with other objects.
link |
So it's not an intrinsic feature of that object.
link |
It's a feature of the interactions with other things.
link |
And so one of the nice features of assembly
link |
is you've basically gotten rid of,
link |
you just look at the things that exist,
link |
and you've gotten rid of the mechanisms
link |
for constructing them in some sense.
link |
Like the machines are not as important
link |
in the current construction of the theory,
link |
although I would like to bridge it
link |
to some ideas about constructors.
link |
But then you can only communicate with things
link |
as Lee was saying,
link |
if you have some overlap in the past history.
link |
So if you had an alien species
link |
that had absolutely no overlap,
link |
then there would be no means of communication.
link |
But as we progress further and further in time
link |
and more things become possible
link |
because the assembly spaces are larger,
link |
because you can have a larger assembly space
link |
in terms of index and also just the size of the space,
link |
because it's exponentially growing,
link |
then more things can happen in the future.
link |
And the example I like to give is actually
link |
when we made first contact with gravitational waves,
link |
because that's an alien phenomenon
link |
that's been permeating our planet,
link |
not alien in life phenomenon,
link |
but alien like something we had never knew existed.
link |
It's been like there's gravitational waves
link |
rippling through this room right now,
link |
but we had to advance to the level of Einstein
link |
writing down his theory of relativity
link |
and then 100 years of technological development
link |
to even quote unquote see that phenomena.
link |
So the, okay, to see that phenomena,
link |
our causal graph have to start intersecting.
link |
Yeah, we needed the idea to emerge first,
link |
the abstraction, right?
link |
And then we had to build the technology
link |
that could actually observe features of that abstraction.
link |
So the nice promising thing is over time,
link |
the graph can grow so we can start overlapping eventually.
link |
Yeah, so the interesting feature of that graph
link |
is there was an event 1.4 billion years away
link |
of a black hole merger that we detected on our detector.
link |
And now suddenly we're connected
link |
through this communication channel
link |
with this distant event in our universe
link |
that if you think about 1.4 billion years ago,
link |
what was happening on this planet
link |
or even further back in time,
link |
that there's common physics underlying all of those events,
link |
but even for those two events to communicate.
link |
Now I understand what you were going on about
link |
Yeah, I'm sorry, this is a really abstract example,
link |
Your causal graphs are now overlapping.
link |
Well, let's just say now our causal graphs
link |
are overlapping in the deep past.
link |
No, I like it, so you made it.
link |
I totally missed it.
link |
Oh, the 1.6 billion.
link |
You made a connection with it.
link |
No, I do like that.
link |
No, no, you can tell me what your epiphany is now.
link |
And I should get the jokes before 30 seconds after, so.
link |
No, it's all right.
link |
The joke from two minutes ago.
link |
I'm slow on the uptake here.
link |
I wasn't able to comprehend
link |
what you were talking about when saying
link |
the channel communicating to the past.
link |
But what you're saying is we were able to infer
link |
what happened 1.4 billion years ago.
link |
We detected the gravity wave.
link |
I mean, I think it's amazing that at that time,
link |
we weren't even, we were just becoming multicellular,
link |
And then we progressed from multicellularity
link |
through to technology and built the detector
link |
and then we just extrapolate backwards.
link |
So, although we didn't do anything back to the graph
link |
back in time, we understood this existence
link |
then overlapped going forward.
link |
Well, that's because our graphs are larger.
link |
Yeah, but that means that has a consequence.
link |
One of the things I was trying to say is I think,
link |
I don't know, Sarah might be, she can correct me,
link |
information first and I'm a object first kind of guy.
link |
So, I mean, there's things that get constructed.
link |
There has to be this transition in random constructions.
link |
So when the object that's being constructed by the process
link |
bakes in that memory and those memories then add on
link |
and add on and add on.
link |
So as it becomes more competent and life is about
link |
taking those memories and compressing them,
link |
increasing their autonomy.
link |
And so I think that, like the cell that we have
link |
in biology on earth is our way of doing that,
link |
that really the maximum ability to take memories
link |
and to act on the future.
link |
Oh, I think that's mathematics.
link |
No, mathematics doesn't exist.
link |
No, but that's the point.
link |
The point is that abstractions do exist.
link |
They're real physical things.
link |
We call them abstractions.
link |
But the point about mathematics that I think is,
link |
so I don't disagree that I think you're object first
link |
and I'm information first, but I think I'm only
link |
information first in the sense that I think the thing
link |
that we need to explain is what abstractions are
link |
and what they are as physical things
link |
because of all of human history,
link |
we've thought that there were these properties
link |
that are disembodied, exist outside of the universe
link |
and really they do exist in the universe
link |
and we just don't understand what their physics is.
link |
So I think mathematics is a really good example.
link |
We do theoretical physics with math,
link |
but imagine doing physics of math
link |
and then thinking about math as a physical object.
link |
And math is super interesting.
link |
I think this is why we think it describes reality so well
link |
because it's the most copyable kind of information.
link |
It retains its properties
link |
when you move it between physical media,
link |
which means that it's very deep.
link |
And so it seems to describe the universe really well,
link |
but it probably is because it's information
link |
that's very deep in our past.
link |
And it's just, we invented a way of communicating it
link |
very effectively between us.
link |
Isn't math more fundamental?
link |
Isn't the assembly of the graph,
link |
isn't basically, I sound completely boring.
link |
It's like math, assembly theory invented math, but it did.
link |
So what is math exactly?
link |
It's a nice simplification,
link |
a simple description of what?
link |
So we have a computer scientist,
link |
a physicist, and a chemist here.
link |
I think the chemist is gonna define math
link |
and you guys can correct me.
link |
Lay in on us, Lee.
link |
I think the ability to label objects
link |
and place them into classes
link |
and then do operations on the objects is what math is.
link |
what does it mean to be object first
link |
versus information first?
link |
So what's the difference between object and information
link |
when you get to that low fundamental level?
link |
Well, I might change my view.
link |
So I'm stuff first, the stuff.
link |
And then when stuff becomes objects,
link |
it has to invent information.
link |
And then the information acts on more stuff
link |
and becomes more objects.
link |
So I think there is a transition to information
link |
that occurs when you go from stuff to objects.
link |
You mean time though, I think.
link |
Information is emergent.
link |
Information is actionable memories from the universe.
link |
So when memories become actionable, that's information.
link |
But there's always memory, but it's not actionable.
link |
And then it's not information.
link |
And actionable is what you can create.
link |
If you can't use it, then it's not information.
link |
If you can't transmit it,
link |
if it doesn't have any causal consequence.
link |
Falls in the forest.
link |
I don't understand.
link |
Why is that not information?
link |
It's not information.
link |
It's stuff happening, but it's not causal.
link |
Yeah, yeah, we can.
link |
But it's happening.
link |
Happening requires information.
link |
No, no, no, no, no.
link |
Stuff is always happening.
link |
No, this is where the physicists
link |
and the math petitions get themselves in a loop
link |
because I think the universe, I mean,
link |
I think say Max Tegmark is very playful
link |
and say like the universe is just math.
link |
Well, the universe is just math.
link |
Then we might as well not bother having any conversation
link |
because the conversation already written,
link |
we just might as well go to the future and say,
link |
can you just give us the conversations happened already?
link |
So I think the problem is that math petitions
link |
are so successful at labeling stuff
link |
and so successful understanding the stuff
link |
through those labels,
link |
they forget that actually those labels had to emerge
link |
and that information had to be built on those memories.
link |
So memory in the universe, so constraints, graph,
link |
when they become actionable
link |
and the graph can loop back on itself
link |
or interact with other graphs and they can intersect,
link |
those memories become actionable
link |
and therefore they're information.
link |
And I think you just changed my mind
link |
on something pretty big, but I don't have a pen.
link |
So I can't write, I'm gonna write it down later,
link |
but roughly the idea is like you've got these two graphs
link |
of objects of stuff that you have memories
link |
and then when they intersect
link |
and then they can act on each other,
link |
that's maybe the mechanism by which information is then,
link |
so then you can then abstract.
link |
So when one graph can then build another graph and say,
link |
hey, you don't have to go through the nonsense
link |
we had to go through.
link |
Here's literally the way to do it.
link |
Stuff always comes first,
link |
but then when stuff builds the abstraction,
link |
the abstractions can be then teleported
link |
And the abstractions is the looping back power.
link |
Am I making, I don't know, I got stuck.
link |
Yeah, so first, God made stuff.
link |
Then after that, when you start to be able
link |
to form abstractions, that's when the information.
link |
God is the memory the universe can remember.
link |
God is the memory the universe can remember.
link |
Otherwise, there's no, wait, did you?
link |
Someone's gonna be deciphering that statement
link |
hundreds of years from now, what the hell does that mean?
link |
What does the humans mean by this?
link |
Look, don't diss my one liners.
link |
I took me 15 seconds to come up with that.
link |
I don't know what it means.
link |
What does it mean?
link |
Okay, wait, we need to, how do we get onto this?
link |
We were time, causality, mathematics.
link |
So what is mathematics in this picture of stuff,
link |
objects, memory, and information?
link |
What exactly is mathematics?
link |
It's the most efficient labeling scheme
link |
that you can apply to lots of different graphs.
link |
Labeling scheme doesn't make it sound useful.
link |
Yep, sure, please.
link |
Have you rejected my definition of mathematics?
link |
Yeah, no, it's all right.
link |
No, I mean, I think we have a problem, right?
link |
Cause we can't not be us,
link |
like we're stuck in the shells we are
link |
and we're trying to observe the world.
link |
And so mathematics looks like it has certain properties.
link |
And I guess the thought experiment I find is useful
link |
is to try to imagine if you were outside of us
link |
looking at us as physical systems using mathematics,
link |
what would be the specific features you associate
link |
to the property of understanding mathematics
link |
and being able to implement it in the universe, right?
link |
And when you do that,
link |
mathematics seems to have some really interesting properties
link |
relative to other kinds of abstraction
link |
we might talk about like language or artistic expression.
link |
One of those properties is the one I mentioned already
link |
that is really easy to copy between physical media.
link |
So if I give you a mathematical statement,
link |
you almost immediately know what I mean.
link |
If I tell you the sky is blue,
link |
you might say, is it gold ball blue?
link |
What color blue do you mean?
link |
And you have a harder time visualizing what I actually mean.
link |
So mathematics carries a lot of meaning with it
link |
when it's copied between physical systems.
link |
It's also the reason we use it to communicate with computers.
link |
And then the second one is it retains its property
link |
of actually what it can do in the universe when it's copied.
link |
So the example I like to give there
link |
is think about like Newton's law of gravitation.
link |
It's actually, it's a compressed regularity
link |
of a bunch of phenomena that we observe in the universe,
link |
but then that information actually is causal in a sense
link |
that it allows us to do things we wouldn't be able to do
link |
without that particular knowledge
link |
and that particular abstraction.
link |
And in this case, like launch satellites to space
link |
or send people to Mars or whatever it is.
link |
So if you look at us from the outside and you say,
link |
what is it for physical systems
link |
to invent a thing called mathematics
link |
and then to use and then it to become a physical observable,
link |
mathematics is kind of like
link |
the universally copyable information
link |
that allows new possibility spaces
link |
to be open in the future
link |
because it allows this kind of ability
link |
to map one physical system to another
link |
and actually understand that the general principles.
link |
So is it helping the overlap of causal graphs then
link |
Oh, I think that's the explanation for what it is
link |
in terms of the physical theory of assembly
link |
would be some feature of the structure
link |
of the assembly spaces of causal graphs
link |
and their relationship to each other.
link |
So for example, and I mean, this is things
link |
that we're gonna have to work out over the next few years.
link |
I mean, we're in totally uncharted conceptual territory
link |
here, but as is usual, diving off the deep end,
link |
but I would expect that we would be able to come up
link |
with a theory of like, why is it that some physical systems
link |
can communicate with each other?
link |
Like language, language is basically
link |
because we're objects extended over time
link |
and some of the history of that assembly space
link |
actually overlaps.
link |
And when we communicate,
link |
it's because we actually have shared structure
link |
in our causal history.
link |
Let me have another quick go at this, right?
link |
So I think we all agree.
link |
So I think we take mathematics for granted
link |
because we've gone through this chain, right?
link |
Of, you know, we all share a language now, okay?
link |
And we can, well, we share language,
link |
so we have languages that we can make interoperable.
link |
And so whether you're speaking, I don't know,
link |
all the different dialects of Chinese,
link |
all the different dialects of English, French, German,
link |
whatever, you can interconvert them.
link |
The interesting thing about mathematics now
link |
is that everybody on planet Earth, every human being
link |
and computers share that common language.
link |
That language was constructed by a process in time.
link |
So what I'm trying to say is assembly invented math
link |
is those, right from the, you know,
link |
mathematics didn't occur, it didn't exist before life.
link |
Abstraction was invented by life, right?
link |
That doesn't mean that the universe wasn't capable
link |
of mathematical things.
link |
Wait a minute, can we just ask that old famous question,
link |
is math invented or discovered?
link |
So when you say assembly invented, or whatever.
link |
It means it's just.
link |
Well, someone might just say assembly
link |
is a mathematical theory, but sorry.
link |
Are we arguing? Exactly.
link |
Are we arguing now?
link |
That's what it sounds like.
link |
Are we discovering mathematics?
link |
No, well, yes and no.
link |
And you call mathematics a language
link |
that we're developing. I would say that,
link |
look, I'm pretty sure that there are some very common
link |
seeds of mathematics in the universe, right?
link |
But actually not the mathematics that we are finding now
link |
is not discovered, it's invented.
link |
And, but even though I think those two terms
link |
are very triggering, and I don't think
link |
they're necessarily useful,
link |
because I think that what people do,
link |
the mathematicians that say, oh, mathematics was discovered
link |
because they live in a universe where there is no time
link |
and it just all exists.
link |
But what I'm saying is, and I think in the same way
link |
you can create, let's say I'm gonna go and create
link |
and make a piece of art, did I make that piece of art
link |
or did I discover it?
link |
Like inventing the aeroplane.
link |
Did I invent the aeroplane?
link |
Let's stick with the aeroplane.
link |
The aeroplane is a good one.
link |
Let's say, did I discover the aeroplane?
link |
Well, in a way, the universe discovered the aeroplane
link |
because it's just chucked a load of atoms together
link |
and a load of random human beings want to do stuff
link |
and then we discover the aeroplane
link |
in the space of all the possibilities.
link |
But here's the thing, when the space of possibilities
link |
is so vast, infinite almost, and you're able to actualize
link |
one of those in an object, then you are inventing it.
link |
So in mathematics, because there are infinite number
link |
of theorems, the fact you're actually pulling,
link |
there's no difference between inventing
link |
a mathematical structure and inventing the aeroplane.
link |
They're the same thing, but that doesn't mean
link |
that now the aeroplane exists in the universe,
link |
it's something weird about the universe.
link |
So I think that the more, this is the thing
link |
that you probably, the more memory required
link |
for the object, the more invented it is.
link |
So when a mathematical theorem needs more bytes
link |
to store it, the more invented it is,
link |
and the less bytes, the more discovered it is.
link |
But everything then is invented.
link |
It's just more or less invented.
link |
The universe has to generate everything as it goes.
link |
And it wasn't there in the beginning.
link |
And the way we're thinking it,
link |
when you're thinking about the difference
link |
between invented and discovered,
link |
is because we're throwing away all the memory.
link |
So if you start to think in terms of causality and time,
link |
then those things become the same.
link |
Everything is invented.
link |
And the idea is to make everything intrinsic
link |
So I think one of the features of assembly theory
link |
is we don't wanna have external observers.
link |
There's been this long tradition in physics
link |
of trying to describe the universe from the outside
link |
and not the inside.
link |
And the universe has to generate everything itself
link |
if you do it from the inside.
link |
Assembly theory describes how the universe builds itself.
link |
Did it take you 15 seconds to say that?
link |
And to come up with that also?
link |
No, I've thought of that before.
link |
That's a good line.
link |
It's a, it's like.
link |
Oh, you're making fun of me.
link |
No, I'm not making fun, I'm having fun.
link |
There's a difference.
link |
She's inventing fun.
link |
I'm not all intimidated.
link |
And there's a causal history to that fun.
link |
You mentioned that there's no way to communicate
link |
with aliens until there's overlap in the causal graph.
link |
Communication includes being able to see them.
link |
And like, what are we, this is the question is,
link |
is communication any kind of detection?
link |
And if so, what do aliens look like
link |
as you get more and more overlap on the causal graph?
link |
You're assuming, let's assume that aliens,
link |
so when you see them and they see you,
link |
you're assuming they have vision,
link |
they have the ability to construct in 3D and in time.
link |
That's a lot of assumptions they're making.
link |
All right, let's step back.
link |
So yes, okay, you're right.
link |
So when, in the English language,
link |
when we say the word see, we mean visually,
link |
they show up to a party and it's like,
link |
oh, wow, that's an alien.
link |
That's visual, that's 3D, that's, okay.
link |
And that's also assuming scale,
link |
spatial scale of something that's visible to you.
link |
So it can't be microscopic or it can't be so big
link |
that you don't even realize that's an entity.
link |
Okay, but other kinds of detection too.
link |
I would make it more abstract and go down.
link |
I was thinking this morning about how to rewrite
link |
the Arecibo message in assembly theory
link |
and also to abandon binary.
link |
Because I don't think aliens necessarily,
link |
why should they have binary?
link |
Well, they have some basic elements
link |
with which to do information exchange.
link |
Let's make it more fundamental, more universal.
link |
So we need to think about what is the universal way
link |
of making a memory and then we should
link |
reencode Arecibo in that way.
link |
What's more basic than zeros and ones?
link |
Well, it's really difficult to get out
link |
that causal chain because we're so,
link |
so let's erase the idea of zero for a moment.
link |
It took human beings a long time
link |
to come up with the idea of zero.
link |
Now you got the idea of zero, you can't throw it away.
link |
To discover the idea of zero.
link |
To discover or invent.
link |
I don't know, but it took a long time
link |
so it was invented, that's right.
link |
Yeah, I think zero was invented, exactly.
link |
So it's not a given that aliens know what zero is.
link |
That's a massive assumption.
link |
It's a useful discovery.
link |
You're saying if you break the causal chain
link |
there might be some other more efficient way of representing.
link |
That's why I wanna meet him and ask him.
link |
But you won't be able to ask him until.
link |
So I interrupted you and I think you're making good point.
link |
I was just gonna say, well look.
link |
Rather than saying.
link |
Please internet, tweet at him for the rude interruptions.
link |
Oh, go ahead, I'm sorry.
link |
Maybe it's change.
link |
How do we, so, oh, I don't know what it's like
link |
I would like to know.
link |
What is the full spectrum of what aliens
link |
might look like to us?
link |
Now that we've laid this all on the table of like,
link |
all right, so there has to be some overlap
link |
in this causal chain that led to them.
link |
What are we looking for?
link |
What do you think we should be looking for?
link |
So you mentioned mass spec.
link |
Measuring certain objects that aliens could create
link |
or are aliens themselves.
link |
We show up to a planet or maybe not a planet.
link |
Or maybe, what the hell is the basic object
link |
we're trying to measure the assembly index of?
link |
Let's cut ourselves a break.
link |
Let's assume that they are metabolized.
link |
They've got an energy source.
link |
And they're a size that we can recognize.
link |
Let's give ourselves a break.
link |
Because there could be aliens that are so big
link |
we won't recognize we're seeing them.
link |
There might be aliens that are so small
link |
we don't yet have the ability to,
link |
we don't have microscopes that can see far enough away
link |
that just wouldn't be able to see them.
link |
So what's a good range?
link |
So let's just make a range.
link |
Let's just be very anthropocentric and say,
link |
we're gonna look for aliens roughly our size
link |
and technology our size.
link |
Because we know it's possible on Earth, right?
link |
I mean, a reasonable thing to do would be
link |
to find exoplanets that are in the same zone as Earth
link |
in terms of heat and stuff.
link |
And then say, hey, if there's that same kind of gravity,
link |
same kind of stuff, we could reasonably assume
link |
that alien life there might use a similar kind
link |
of physical infrastructure.
link |
And then we're good.
link |
So then your question becomes really relevant.
link |
Say, right, let's use vision, sound, touch.
link |
So okay, that's really nice.
link |
So that if there's a lot of aliens out there,
link |
there's a good likelihood if you match to the planet
link |
that they're going to be in the same spatial
link |
and temporal, operating in the same spatial
link |
and temporal domain as humans.
link |
Okay, within that, what do they look like visually?
link |
What do they sound like?
link |
What do they, oh god, this sounds creepy.
link |
Tastes like, what do they, oh, smell like, smell like.
link |
That sounds like our clubhouse.
link |
We was like, can we have sex with aliens?
link |
Which was basically me saying.
link |
Passionate, passionate love.
link |
But it wasn't actually about sex.
link |
It was about, is our chemistry compatible, right?
link |
Yeah, can we, yeah.
link |
Are they edible too?
link |
They could be very edible.
link |
They could be delicious.
link |
That's why I want to see some aliens, right?
link |
Because I think, are there, I think evolution,
link |
I mean, evolution exploits symmetry, right?
link |
Because why generate memory?
link |
Why generate storage, the need for storage space
link |
when you can use symmetry?
link |
So, and symmetry is quite, may be quite effective
link |
in allowing you to mechanically design stuff, right?
link |
So maybe alien, you could be reasonable to assume
link |
that aliens could have, they could be bipedal.
link |
They could be symmetric in the same way.
link |
Might have a couple of eyes or a couple of senses.
link |
We can make them, perhaps there's this whole zoo
link |
of different aliens out there.
link |
And we'll never get to be able to classify
link |
some of the weird aliens we can't interact with
link |
because they have made such weird stuff.
link |
But we are just going to look at,
link |
we're going to find aliens that look most like us.
link |
Because those are the first ones we're likely to see.
link |
But I think it's really hard to imagine
link |
what the space of aliens is because the space is huge.
link |
Because, you know, like one of the arguments
link |
that you can make about why life emerges in chemistry
link |
is because chemistry is the first scale
link |
in terms of like, you know,
link |
building up objects from elementary objects.
link |
That the number of possible things that could exist
link |
is larger than the universe can possibly make all at once.
link |
So imagine you have two planets
link |
and they're cooking some geochemistry.
link |
You know, our planet invented one kind of biochemistry.
link |
And presumably as you start building up
link |
the complexity of the molecules,
link |
the chances of the overlap in those trajectories,
link |
those causal chains being built up is probably very low.
link |
And it gets lower and lower as it gets further advanced
link |
along its evolutionary path.
link |
So I think it's very difficult to imagine
link |
predicting the technologies that aliens are gonna have.
link |
I mean, it's so, you're looking at basically
link |
planets have kind of convergent chemistry,
link |
but there's some variability.
link |
And then you're looking basically at the outgrowth
link |
into the possibility space for chemistry.
link |
So do you think we would detect the technology,
link |
the objects created by aliens before we detect the aliens?
link |
So when you're talking about measuring assembly index,
link |
don't you think we would detect the garbage first?
link |
Like at the outskirts of alien civilizations,
link |
isn't this just gonna be trash?
link |
I think I would come back to Arecibo.
link |
The Arecibo message sent from the Arecibo telescope
link |
built by Drake, I think, and Sagan.
link |
How's Arecibo spelled?
link |
And there we go, they've got that up there.
link |
That's the telescope that sent the message
link |
that you're talking about.
link |
So that message was sent where?
link |
It was beamed at a star, a specific star,
link |
and it was sent out many years ago.
link |
And what they did, so this is why I was pushing on binary,
link |
it's a binary message.
link |
I think it's a semi prime length number of characters.
link |
So I think 73 by 23, I think.
link |
And it basically represents human bit proton,
link |
binary, human beings, DNA, male and female.
link |
And it's really cool.
link |
But I'm just wondering if it could be done
link |
cause it made assumptions that aliens speak binary.
link |
Why make that assumption?
link |
Why not just assume that if the difference between physics,
link |
chemistry and biology is the amount of memory
link |
that's instead that's recordable by the substrates,
link |
then surely the universal thing,
link |
I'm gonna make some sacrilegious statement,
link |
which I think is pretty awesome for people to argue with.
link |
So this is, we're looking at an image
link |
where it's the entirety of the message encoded in binary.
link |
And then there's a probably interpretation
link |
of different parts of that image.
link |
There's a person, there's green parts.
link |
It looks like for people just listening,
link |
like a game of Tetris.
link |
So it's encoding in minimal ways,
link |
a bunch of cool information, probably.
link |
Representing all of us.
link |
So the topic's kind of teaching us how to count
link |
and then it all goes all the way down
link |
teaching you chemistry and then just says,
link |
but it makes so many assumptions.
link |
And I think if we can actually,
link |
so look, I think, I mean, Sarah's much more eloquent
link |
in expressing this, but I'll have a go
link |
and you can correct it if you want,
link |
which is like one of the things that Sarah has had
link |
a profound effect on the way I look at the origin of life.
link |
And this is one of the reasons why we're working together
link |
because we don't really care about the origin of life.
link |
We wanna make life, make aliens and find aliens.
link |
Make aliens, find aliens.
link |
I think we might have to make aliens in the lab
link |
before we find aliens in the universe, right?
link |
I think that would be a cool way to do it.
link |
So what is it about the universe that creates aliens?
link |
Well, it's selection through assembly theory,
link |
creating memories, because when you create memories,
link |
you can then command your domain.
link |
You can basically do stuff.
link |
You can command matter.
link |
So we need to find a way by understanding what life is
link |
of how the minimal way to command matter,
link |
how that would emerge in the universe.
link |
And if we want to communicate,
link |
I mean, maybe we don't want
link |
to necessarily uniformly communicate.
link |
What I would do perhaps if I had,
link |
is I would send out lots of probes away from Earth
link |
that have this magic way of communicating with aliens,
link |
get them quite a far away from Earth, plausibly deniable,
link |
and then send out the message
link |
that would then attract all the aliens,
link |
and then basically work out if they were friend or foe
link |
and how they wanna hang out.
link |
The messages being something has to do with the memories.
link |
Yes, like the assembly version of Arecibo,
link |
so that everyone in the universe
link |
that has been understands what life is.
link |
So aliens need to work out what they are.
link |
Once they've worked out what they are,
link |
they then can work out how to encode what they are,
link |
and then they can go out and send messages.
link |
It's like the universal, the Rosetta Stone
link |
for life in the universe is working out
link |
how the memories are built.
link |
I don't know, Sarah, you have any, well,
link |
whether you would agree with that.
link |
No, I wanted to raise a different point,
link |
which is about the fact that we can't see the aliens yet
link |
because we haven't gotten the technology.
link |
And presumably we think assembly theory
link |
is the right way of doing it,
link |
but I don't think that we know how to go
link |
from the kind of data you're describing, Lex,
link |
like visual data or smell
link |
to construct the assembly spaces yet.
link |
And in some ways, I think that the problem
link |
of life detection really is the same problem
link |
at the foundations of AI that we don't understand
link |
how to get machines to see causal graphs,
link |
to see reality in terms of causation.
link |
And so I think assembly and AI
link |
are gonna intersect in interesting ways, hopefully,
link |
but the sort of key point,
link |
and I've been trying to make this argument more recently,
link |
I might write an essay on it,
link |
is people talk about the great filter, right?
link |
And which is, again, this doomsday thing
link |
that people wanna say there's no aliens out there
link |
because something terrible happened to them.
link |
And it matters whether that's in our past or our future
link |
as to the longevity of our species, presumably,
link |
which is why people find it interesting.
link |
But I think it's not a physical filter.
link |
It's not like things go extinct.
link |
I think it's literally,
link |
we don't have the technology to see them.
link |
And you could see that with microscopes.
link |
I mean, we didn't know there were microbes on this table
link |
or tables for thousands of years or telescopes.
link |
Like there's so much of the universe we can't see.
link |
And then basically what we have done as a species
link |
is outsource our physical perceptions to technology,
link |
building microscopes based on our eyes,
link |
and building seismometers based on our sense of feelings,
link |
like feel earthquakes and things.
link |
And AI is basically we're trying to outsource
link |
what's actually happening in our thinking apparatus
link |
into machines now and to technological devices.
link |
And maybe that's the key technology
link |
that's gonna allow us to see things like us
link |
and see the universe in a totally different way.
link |
But you kind of mentioned the great filter.
link |
Do you think there's a way through technology
link |
to stop being able to see stuff?
link |
So can you take a step backwards?
link |
Did you imply that with the great, so like?
link |
Well, no, I mean, I think there's a great perceptual filter
link |
in the sense that a example of life evolving on a planet
link |
over billions of years has to acquire
link |
a certain amount of knowledge and technology
link |
to actually recognize the phenomena that it is.
link |
Well, that's the sense I have is,
link |
I mean, you talk with physicists, engineers in general,
link |
there's this kind of idea that we have
link |
most of the tools already to hear the signal.
link |
But to me, it feels like we don't have any of the tools
link |
to see the signal.
link |
No, we don't know what we're doing, yeah, I agree.
link |
That's the biggest, like to hear.
link |
We don't have the tools to really hear, to see.
link |
Aliens are everywhere.
link |
We just don't have the, yeah, well, oh, that's.
link |
I mean, I got this in part, actually,
link |
because you were like, you know,
link |
last time I was here, he was like, look at the carpet.
link |
You know, if you had an alien detector
link |
where the carpet be aliens.
link |
I mean, I think we really don't.
link |
I think it would be.
link |
But the aliens would nevertheless have a high assembly index
link |
or produce things of high assembly index.
link |
And those things of high assembly index,
link |
you have to have a detector that can recognize
link |
high assembly index in all its forms.
link |
That's it, that's it.
link |
Take data, construct assembly space.
link |
Yeah. Those patterns, basically.
link |
So one way to think about high assembly index
link |
is interesting patterns of basic ingredients.
link |
I can give you an example,
link |
because I mean, in molecules,
link |
we've been talking about in objects,
link |
but we're also trying to do it in spatial trajectories.
link |
Like, imagine you're just,
link |
like, I always get bothered by the fact that, like,
link |
when you look at birds flocking,
link |
you can describe that with like a simple Boyd's model,
link |
or like, you know, people use spin glass
link |
to describe animal behavior.
link |
And those are like really simple physics models.
link |
Yet you're looking at a system that you know has agency
link |
and there's intelligence in those birds.
link |
And basically, like, you can't help but think
link |
there must be some statistical signatures
link |
of the fact that they're,
link |
that's a group of agents versus, you know, like,
link |
I don't know, you know, the physics example,
link |
maybe like, I don't know, Brownian motion or something.
link |
And so what we're trying to do
link |
is actually apply assembly to trajectory data
link |
to try to say there's a minimal amount of causal history
link |
to build up certain trajectories for observed agents
link |
that's like an agency detector for behavior.
link |
Do you think it's possible to do some like Boyd's
link |
or those kinds of things, like artificial,
link |
like cellular automata, play with those ideas
link |
with assembly, with assembly theory?
link |
Have you found any useful, really simple mathematical,
link |
like, simulation tools that allow you
link |
to play with these concepts?
link |
So like one, of course, you're doing mass spec
link |
in this physical space with chemistry,
link |
but it just seems, well, I mean,
link |
computer science person, maybe,
link |
it seems easier to just.
link |
It seems even sexier in terms of tweeting visual information
link |
on Twitter or Instagram, more importantly,
link |
to play like, here's an organism of a low assembly index
link |
and here's an organism of a high assembly index
link |
and let's watch them create more and more memories
link |
and more and more complex objects.
link |
And so like, in mathematics,
link |
you get to observe what that looks like
link |
to build up an intuition what assembly index is like.
link |
We are building a toolkit right now.
link |
So I think it's a really good idea,
link |
but what we've got to do is I'm kind of still obsessed
link |
with the infrastructure required.
link |
And one of the reasons why I was pushing on information
link |
and mathematics when human beings,
link |
when human beings, we take a lot of the infrastructure
link |
And I think we have to strip that back a bit
link |
for going forward, but you're absolutely right.
link |
I would agree that I think the fact that we exist
link |
in the universe, this is like,
link |
I can see that lots of people would disagree
link |
with the statement, but I don't think Sarah will,
link |
The fact that objects exist,
link |
I don't think anyone on earth will disagree
link |
that objects can exist elsewhere, right?
link |
But they will disagree that life can exist elsewhere.
link |
But what perhaps I'm trying to say is that
link |
the acquisition, the universe's ability to acquire memory
link |
is the very first step for building life.
link |
And that must be, that's so easy to happen.
link |
So therefore alien life is everywhere
link |
because all alien life is,
link |
is those memories being compressed and minimalized
link |
and the alien equivalent of the cell working.
link |
So I think that we will build new technologies
link |
to find aliens, but we need to understand what we are first
link |
and how we go from physics to chemistry to biology.
link |
The most interesting thing,
link |
as you're saying to these two organisms,
link |
different assemblies, there's one you get into biology.
link |
Biology gets more and more weird,
link |
more and more contingent.
link |
Physics is, chemistry is less weird
link |
cause the rules of chemistry are smaller
link |
than the rules of biology.
link |
And then going away to physics where you have a very
link |
nicely tangible number of ways of arranging things.
link |
And I think assembly theory just helps you appreciate that.
link |
And so once we get there,
link |
my dream is that we are just gonna be able to suddenly,
link |
I mean, I'm maybe just being really arrogant here.
link |
I don't mean to be arrogant.
link |
It's just, I've got this hammer called assembly
link |
and everything's a nail.
link |
But I think that once we crack it,
link |
we'll be able to use assembly theory plus telescopes
link |
Do you have, Sarah, do you have disagreements with Lee
link |
on the number of aliens that are out there?
link |
I do actually, yeah, well.
link |
And what they look like.
link |
So any of the things we've been talking about,
link |
is there nuanced, it's always nice to discover wisdom
link |
through nuanced disagreement.
link |
Yeah, I don't wholly disagree, but I think,
link |
but I do think I disagree.
link |
It's kind of, there's nuance there.
link |
It is nuanced, right?
link |
So you made the point earlier that you think,
link |
once we discover what life is,
link |
we'll see alien life everywhere.
link |
And I think I agree on some levels in the sense
link |
that I think the physics that governs us is universal.
link |
But I don't know how far I would go to say,
link |
to say that we're a likely phenomenon
link |
because we don't understand all of the features
link |
of the transition at the origin of life,
link |
which we would just say in assembly,
link |
as you go from the no memory physics
link |
to there's like a critical transition
link |
around the assembly index
link |
where assembliness starts to increase.
link |
And that's what we call the evolution of the biosphere
link |
and complexification of the biosphere.
link |
So there's a principle of increasing assembliness
link |
where that goes back to what I was saying
link |
at the very beginning about the physics of the possible,
link |
that the universe basically gets in this mode
link |
of trying to make as much possibilities as possible.
link |
Now, how often that transition happens
link |
that you get the kind of cascading effect
link |
that we get in our biosphere, I think we don't know.
link |
If we did, we would know the likelihood of life
link |
And a lot of people wanna say life is common,
link |
but I don't think that we can say that yet
link |
till we have the empirical data,
link |
which I think you would agree with.
link |
But then there's this other kind of thought experiment I have
link |
which I don't like, but I did have it,
link |
which is if life emerges on one planet
link |
and you get this real high density of things
link |
that can exist on that planet,
link |
is it sort of dominating the density of creation
link |
that the universe can actually generate?
link |
So like if you're thinking about counting entropy, right?
link |
Like the universe has a certain amount of stuff in it.
link |
And then assembly is kind of like an entropic principle.
link |
But the idea is that now transformations among stuff
link |
or the actual physical histories of things
link |
now become things that you have to count
link |
as far as saying that these things exist
link |
and we're increasing the number of things that exist.
link |
And if you think about that cosmologically,
link |
maybe Earth is sucking up all the life potential
link |
of the whole universe, I don't know.
link |
How's that, can you expand that a little bit?
link |
Why can any one geographical region
link |
suck up the creative capacity of the universe?
link |
Just like, I know it's a ridiculous thought.
link |
I don't actually agree with it,
link |
but it was just a thought experiment.
link |
I love that you can have thoughts
link |
that you don't like and don't agree with,
link |
but you have to think through them anyway.
link |
The human mind is fascinating.
link |
Yeah, I think these sort of counterfactual
link |
thought experiments are really good
link |
when you're trying to build new theories
link |
because you have to think through all the consequences.
link |
And there are people that want to try to account for,
link |
say, the degrees of freedom on our planet
link |
in cosmological inventories of talking about
link |
the entropy of the universe.
link |
And when we're thinking about cosmological
link |
arrow of time and things like that.
link |
Now, I think those are pretty superficial proposals
link |
as they stand now, but assembly would give you
link |
a way of counting it.
link |
And then the question is if there's a certain
link |
maximal capacity of the universe's speed
link |
of generating stuff, which Lee always has this argument
link |
that assembly is about time.
link |
The universe is generating more states.
link |
Really what it's generating is more assembly possibilities.
link |
And then dark energy might be one manifestation of that,
link |
that the universe is accelerating its expansion
link |
because that makes more physical space.
link |
And what's happening on our planet is it's accelerating
link |
in the expansion of possible things that exist.
link |
And maybe the universe just has a maximal rate
link |
of what it can do to generate things.
link |
And then if there is a maximal rate,
link |
maybe only a certain number of planets
link |
can actually do that.
link |
Or there's a trade off about the pace of growth
link |
on certain planets versus others.
link |
I have a million questions there,
link |
but do you have thoughts on?
link |
Just a quick, yeah, I'll just say something very quick.
link |
It's a thought experiment.
link |
No, it's good, I think I get it.
link |
So what I want to say is when I mean aliens are everywhere,
link |
I mean memories are the prerequisite
link |
for aliens via selection
link |
and then concentration of selection
link |
when selection becomes autonomous.
link |
So what I would love to do is to build,
link |
say a magical telescope that was a memory,
link |
a magical one, or a real one,
link |
that would be a memory detector to see selection.
link |
So you could get to exoplanets and say that exoplanet
link |
looks like there's lots of selection going on there.
link |
Maybe there's evolution and maybe there's going to be life.
link |
So what I'm trying to say is narrow down
link |
the regions of space where you say
link |
there's definitely evidence of memory as high assembly there
link |
or not high assembly, because that would be life,
link |
but where it's capable of happening.
link |
And then that would also help us frame the search for aliens.
link |
I don't know how likely it is to make the transition
link |
to cells and all the other things.
link |
I think you're right.
link |
But I think that we just need to get more data.
link |
Well, I didn't like the thought experiment
link |
because I don't like the idea
link |
that if the universe has a maximal limit
link |
on the amount it can generate per unit time
link |
that our existence is actually precluding the existence
link |
Well, I'll just say one thing.
link |
But I think that's probably true anyway
link |
because of the resource limitations.
link |
So I don't like your thought experiment
link |
because I think it's wrong.
link |
Well, no, no, I do like the thought experiment.
link |
So what you're trying to say is like,
link |
there is a chain of events that goes back
link |
that's manifestly culminated with life on Earth.
link |
And you're not saying that life isn't possible elsewhere.
link |
You say that there has been these number of things,
link |
contingent things that have happened
link |
that have allowed life to merge here.
link |
That doesn't mean that life can't emerge elsewhere,
link |
but you're saying that the intersection of events
link |
may be concentrated here, right?
link |
And I think there's...
link |
It's more like if you look at,
link |
say the causal graphs are fundamental,
link |
maybe space is an emergent property,
link |
which is consistent with some proposals on quantum gravity,
link |
but also how we talk about things in assembly theory.
link |
Then the universe is causal graphs generating
link |
more structure in causal graphs, right?
link |
So this is how the universe is unfolding.
link |
And maybe there's a cap on the rate of generation.
link |
Like there's only so much stuff
link |
that gets made per update of the universe.
link |
And then if there's a lot of stuff being made
link |
in a particular region that happens
link |
to look the same locally, spatially,
link |
that's an after effect of the fact
link |
that the whole causal graph is updating.
link |
Yeah, I don't know that.
link |
I think that that doesn't work.
link |
I don't think it works either,
link |
but I don't have a good argument in my mind about.
link |
But I do like the idea of the capacity that universe,
link |
cause you've got the number of states.
link |
Yeah, we can come back to it.
link |
Let me ask real quick.
link |
Like why does different like local pockets
link |
of the universe start remembering stuff?
link |
How does memory emerge exactly?
link |
So at the origin of the universe, it was very forgetful.
link |
That's when the physicists were happiest.
link |
It was low memory objects,
link |
which is like ultra low memory objects,
link |
which is what the definition of stuff.
link |
Okay, so how does memory emerge?
link |
How does the temporal stickiness of objects emerge?
link |
I'm gonna take a very chemocentric point of view
link |
because I can't imagine any other way of doing it.
link |
You could think of other ways maybe.
link |
But I would say heterogeneity in matter
link |
is where the memory...
link |
So you must have enough different ways
link |
of rearranging matter for there to be a memory.
link |
So what that means,
link |
if you've got particles colliding in a box,
link |
let's just take some elements in a box.
link |
Those elements can combine in a combinatorial set of ways.
link |
So there's a combinatorial explosion
link |
of the number of molecules or minerals or solid objects,
link |
Because there's such a large number,
link |
the population of different objects that are possible,
link |
this goes back to assembly theory
link |
where assembly theory, there's four types of universes.
link |
So you've got basically, and this is what one was up earlier
link |
where one universe where you've just got
link |
everything is possible.
link |
So you can take all the atoms
link |
and combine them and make everything.
link |
Then you've got basically what is the assembly combinatorial
link |
where you basically have to accrue information in steps.
link |
Then you've got assembly observed,
link |
and then you've got the object assembly going back.
link |
So what I'm trying to say is like,
link |
if you can take atoms and make bonds,
link |
let's say you take a nitrogen atom and add it
link |
to a carbon atom, you find an amino acid,
link |
then you add another carbon atom on
link |
in a particular configuration,
link |
then another one, all different molecules.
link |
They all represent different histories.
link |
So I would say for me right now,
link |
the most simple route into life seems to be
link |
through recording memories and chemistry.
link |
But that doesn't mean there can't be other ways
link |
and can't be other emergent effects.
link |
But I think if you can make bonds
link |
and lots of different bonds,
link |
and those molecules can have a causal effect on the future.
link |
So imagine a box of atoms,
link |
and then you combine those atoms in some way.
link |
So you make molecule A from load of atoms,
link |
and then molecule A can go back to the box
link |
and influence the box.
link |
Then you make A prime or AB or ABC.
link |
And that process keeps going,
link |
and that's where the memories come from,
link |
is that heterogeneity in the universe from bonding.
link |
I don't know if that makes any sense.
link |
And it's beginning to flourish at the chemistry level.
link |
So the physicists have no, like not enough.
link |
They're like desperately begging
link |
for more freedom and heterogeneous components to play with.
link |
Yeah, that's exactly it.
link |
What do you think about that, Sarah?
link |
I mentioned already, I think it's significant
link |
that whatever physics governs life
link |
emerges actually in chemistry.
link |
It's not relevant at the subatomic scale
link |
or even at the atomic scale.
link |
It's in, well, atomic scale because chemistry.
link |
But like when you get into this combinatorial diversity
link |
that you get from combining things on the periodic table,
link |
that's when selection actually matters
link |
or the fact that some things can exist
link |
and others can't exist actually starts to matter.
link |
So I think of it like you don't study gravity
link |
inside the atomic nucleus.
link |
You study it in terms of large scale structure
link |
of the universe or black holes or things like that.
link |
And whatever we're talking about as physics of information
link |
or physics of assembly becomes relevant
link |
at a certain scale of reality.
link |
And the transition that you're talking about,
link |
I would think of as just when you get a sufficient density
link |
in terms of the assembly space
link |
of like the relationship of the overlap
link |
and the assembly space,
link |
which is like a feature of common memory,
link |
there is this transition to assembly dominated physics,
link |
Oh, like when we're talking about,
link |
and we're trying to map out exactly
link |
what that transition looks like.
link |
We're pretty sure of some of its features,
link |
but we haven't done all of the...
link |
Do you think if you were there in the early universe,
link |
you would have been able to predict
link |
the emergence of chemistry and biology?
link |
And I ask that because at this stage as humans,
link |
do you think we can possibly predict the length of memory
link |
that might be able to be formed later on
link |
in this pocket of the universe?
link |
Like how complex is, what is the ceiling of assembly?
link |
I think as much time as you have in the past
link |
is how much you can predict in the future.
link |
Because it's actually physical in the system
link |
and you have to have enough time
link |
for features of that structure to exist.
link |
Wait, let me push back on that.
link |
Isn't there somewhere in the universe
link |
that's like a shortest path that's been,
link |
that stretches all the way to the beginning?
link |
That's building some giant monster?
link |
So you can't predict.
link |
The universe has as much memory
link |
as the largest assembly object in the universe.
link |
But so you can't predict.
link |
You can't predict any deeper than that, no.
link |
So like that, I guess what I'm saying is,
link |
like what intuition do you have about complexity
link |
living in the world that you'd have today?
link |
Right, because you just, you can,
link |
I mean I guess how long does it get more fun?
link |
Like isn't there gonna be at some point,
link |
because there's a heat death in the universe,
link |
isn't there going to be a point of the most,
link |
of the highest assembly of object,
link |
with the highest probability being generated?
link |
When is the universe gonna be the most fun,
link |
and can we freeze ourselves and then live then?
link |
And will you know when you're having the most fun
link |
that this is the best time, you're in your prime?
link |
Are you going to do what everyone does,
link |
which is deny that you're in your prime,
link |
and the best years are still ahead of you?
link |
What option do you have?
link |
I don't, I mean the problem is there's lots of,
link |
lots of really interesting features here.
link |
I just wanna mention one thing that might be,
link |
is I do think assembly theory applies all the way back
link |
to subatomic particles.
link |
And I also think that cosmological selection
link |
might've been actually, there might've been,
link |
I would say it's a really boring bit,
link |
but it's really important for a cosmologist
link |
that universes have gone through.
link |
Was it Lee Smolin who proposed this?
link |
Maybe that there is this,
link |
that basically a universe evolves,
link |
you've got the wrong constants, we'll start again.
link |
And the most productive constants
link |
where you can allow particles to form in a certain way,
link |
propagate to the next universe, and we go again.
link |
So actually selection goes all the way back,
link |
and there's this cycle of universes.
link |
And now this universe has been selected
link |
because life can occur, and it carries on.
link |
But I've really butchered that.
link |
There is a much more.
link |
So this is some aspect where through the selection process
link |
there's parameters that are being fine tuned,
link |
and we happen to be living in one
link |
where there's some level of fine tuning.
link |
Is there, given that, can you steel man the case
link |
that we humans are alone in the universe?
link |
We are the highest assembly index object in the universe.
link |
Yeah, I can, I guess.
link |
I mean, so from a.
link |
Yes, it's possible.
link |
I mean, it's possible.
link |
So let me, so okay, so there is a particular
link |
set of elements on Earth in a particular ratio,
link |
and the right gravitational constant,
link |
and the right viscosity, you know,
link |
of stuff being able to move around,
link |
the right distance from our sun,
link |
right number of events where we have a moon,
link |
the Earth is rotating.
link |
The late heavy bombardment produced a lot of,
link |
brought in the right stuff.
link |
And Mars was cooking up, you know,
link |
the right molecules first.
link |
So it was habitable before Earth.
link |
It was actually doing the combinatorial search.
link |
And before Mars kind of became uninhabitable,
link |
it seeded Earth with the right molecular replicators.
link |
And there was just the right stuff on Earth,
link |
and that's how the miracle of life occurred.
link |
Although I find I'm very uncomfortable with that
link |
because actually, because life came so quickly
link |
in the Earth's past.
link |
But that doesn't mean that life is easy elsewhere.
link |
It just might mean that,
link |
because chemistry is actually not a long term thing.
link |
Chemistry can happen quickly.
link |
So maybe going on with the steel manning of the argument
link |
to say actually, the fact that life emerged quickly
link |
doesn't mean that life is easy.
link |
It just means that the chemistry was right on Earth,
link |
and Earth is very special.
link |
And that's why there's no life
link |
anywhere else in the universe.
link |
Yeah, so Sarah mentioned this kind of cascading thing.
link |
So what if that's the reason we're lucky,
link |
is that we got to have a rare cascading of,
link |
like an accelerating cascading effect
link |
in terms of the complexity of things.
link |
So like, maybe most of the universe
link |
is trying to get sticky with the memory,
link |
and it's not able to really form it.
link |
And then we got really lucky in that.
link |
And it has nothing,
link |
like there's a lot of Earth like conditions, let's say,
link |
but it's just you really, really have to get lucky on this.
link |
But I'm doing experiments right now.
link |
In fact, experiments that Sarah and I are working on,
link |
because we have some joint funding for this,
link |
where we're seeing that the universe
link |
can get sticky really quickly.
link |
Now, of course, we're being very anthropocentric,
link |
we're using laboratory tools, we're using theory,
link |
but actually, the phenomena of selection,
link |
the process of developing heterogeneity,
link |
we can do in the lab.
link |
We're just seeing the very first hints of it.
link |
And wouldn't it be great if we can start to pin down
link |
a bit more precisely becoming good Bayesianists for this,
link |
for the origin of life and the emergence of life,
link |
to finding out what kind of chemistries
link |
we really need to look for.
link |
And I'm becoming increasingly confident
link |
we'll be able to do that in the next few years.
link |
Make life in the lab or make some selection in the lab
link |
from inorganic stuff, from sand, from rocks,
link |
from dead stuff, from moon.
link |
Wouldn't it be great to get stuff from the moon,
link |
put it in our origin of life experiment,
link |
and make moon life?
link |
And restrict ourselves to interesting self replicating
link |
stuff that we find on the moon.
link |
Sarah, what do you think about this approach
link |
of engineering life in order to understand life?
link |
So building life in the machine.
link |
Yeah, so, I mean, Lee and I are trying right now
link |
to build a vision for a large institute
link |
or experimental program, basically, to do this problem.
link |
But I think of it as like, we need to simulate a planet.
link |
So like the Large Hadron Collider was supposed
link |
to be simulating conditions just after the Big Bang.
link |
Lee's built a lot of technology in his lab
link |
to do these kind of selection engines.
link |
But the question you're asking is,
link |
how many experiments do you need to run?
link |
What volume of chemical space do you need to explore
link |
before you actually see an event?
link |
And I like to make an analogy
link |
to one of my favorite particle physics experiments,
link |
which is Super Kamiakande that's looking
link |
for the decay of the proton.
link |
So this is something that we predicted theoretically,
link |
but we've never observed in our universe.
link |
And basically what they're doing is every time
link |
they don't see a proton decay event,
link |
they have a longer bound on the lifetime of a proton.
link |
So imagine we built an experiment with the idea in mind
link |
of trying to simulate planetary conditions,
link |
physically simulate.
link |
You can't simulate origin life in a computer.
link |
You have to do it in an experiment.
link |
Simulate enough planetary conditions
link |
to explore the space of what's possible
link |
and bound the probability for an origin life event.
link |
Even if you're not observing it,
link |
you can talk about the probability.
link |
But we, hopefully, life is not exponentially rare
link |
and we would then be able to evolve
link |
in an automated system alien life in the lab.
link |
And if we can do that, then we understand the physics
link |
as well as we understand what we can do
link |
in particle accelerators.
link |
So keep expanding physically the simulation,
link |
the physical simulation, until something happens.
link |
Yeah, or just build a big enough volume
link |
of chemical experiments and evolve them.
link |
So if you say volume, you mean like literally volume?
link |
I mean physical volume in terms of space,
link |
but I actually mean volume in terms
link |
of the combinatorial space of chemistry.
link |
How do you nicely control the combinatorial exploration,
link |
the search space, such that it's always like
link |
you keep grabbing the low hanging fruit?
link |
Yeah, how do you build a search engine for chemistry?
link |
It's like for aliens. I think you explained it really well.
link |
We should carry on doing this.
link |
I should pretend the physics, be the physicist,
link |
you be the chemist.
link |
So the way to do it is I will always play a joke.
link |
Cause I like writing grants to ask for money
link |
But years ago I started wanting to build.
link |
So I actually wanted the weather.
link |
So I built this robot in my lab called the computer,
link |
which is this robot you can program to do chemistry.
link |
I made a programming language for the computer
link |
and made it operate chemical equipment.
link |
Originally I wrote grants to say,
link |
Hey, I want to make an origin of life system.
link |
And no one would give me any money for this.
link |
They said, what this is ridiculous.
link |
Why are you wanting to make, oh, it's really hard.
link |
You're not a very good origin of life chemist anyway.
link |
Why would we give you any money?
link |
And so I turned it around and said, can you,
link |
can instead, can you give me money to make robots,
link |
to make molecules are interesting.
link |
And everyone went, yeah, okay, you can do that.
link |
And that's, so actually the funny thing is the computer
link |
project, which I have in my lab, which is very briefly,
link |
it's just basically, it's like literally an automated
link |
And we've made a programming language for the test tube,
link |
which is cool, has come as literally came from this.
link |
I went to my lab one day.
link |
So I want to make a search engine to get origin of life
link |
because they don't have a planet.
link |
And I thought about doing in a microfluidic format.
link |
So microfluidic is very nano, very small channels
link |
in device where you can basically have all the pipes
link |
lit dump produced by lithography.
link |
And you can have a chamber, maybe say between say 10
link |
and a hundred microns in volume.
link |
And we slot them all together like Lego,
link |
and we can make an origin of life system.
link |
And I could never get it to work.
link |
And I realized I had to make, do chemistry at the kind of
link |
test tube level and what you want to be able to do.
link |
Yeah, it goes back to that tweet in 1981.
link |
1981, the computer, we're looking at a tweet from Lee.
link |
In 1981, the computer was a distant dream in,
link |
oh wow, this is the scientist looking back.
link |
It is the young boy who dreamed.
link |
In 2018, it was realized, spelled in a British way,
link |
realized, which is the wrong way.
link |
Yeah, I'm starting with Z, but not.
link |
So now there's a system that does the physical
link |
manifestation or whatever the programming language,
link |
the spec tells you to do.
link |
Yeah, well in 1981, I got my first computer, ZX81.
link |
What was the computer?
link |
It was, and I got a chemistry set.
link |
And I liked the chemistry set and I liked the computer
link |
and I just wanted to put them together.
link |
I thought, wouldn't it be cool if I could just use
link |
the computer to control the chemistry set.
link |
And obviously that was insane.
link |
And I was like, you know, eight years old, right?
link |
Nine years old, going on nine years old.
link |
And then I invented the computer
link |
just because I wanted to build this origin of life grid.
link |
Which is like literally a billion test tubes
link |
connected together in real time and real space,
link |
basically throwing a chemical dice.
link |
Throw dice, throw dice, throw dice.
link |
You're gonna get lucky.
link |
And that's what we, I think Sarah and I
link |
have been thinking very deeply about.
link |
Because, you know, there's more money being spent
link |
on the origin of the gravity
link |
or looking at the Higgs boson than the origin of life, right?
link |
And the origin of life is the, I think the biggest question
link |
or not the biggest question, it is a big question.
link |
Let's put it that way.
link |
It is the biggest question.
link |
You're okay saying that.
link |
Isn't it possible once you figure out the origin of life
link |
that that's not going to solve,
link |
that's not actually gonna solve
link |
the question of what is life?
link |
Because you're kind of putting a lot of.
link |
Yeah, I think they're the same problem.
link |
But you're putting, is it possible
link |
that you're putting too many,
link |
too much bets into this origin part?
link |
Maybe the origin thing isn't,
link |
isn't there always a turtle underneath the turtle?
link |
Isn't there a stack of turtles?
link |
Because then if you create it in the lab,
link |
maybe you need some other stuff.
link |
Well, let's not think about the origin.
link |
Like in the lab, there's still memory.
link |
So the experiment is already the product of evolution.
link |
Right, in some maybe really deep way,
link |
not an obvious way, in some very deep way.
link |
So maybe the haters are always going to be like,
link |
well, you have to reconstruct the fold.
link |
You have to build a new script.
link |
Fortunately for us, the haters are not aware
link |
Well, no, I know, I just.
link |
We're the one making that argument usually, but yeah.
link |
I just think that if we create life in the lab,
link |
it's not obvious that you'll get
link |
to the deep, deep understanding of necessarily,
link |
what is the line between life and non life?
link |
No, I think, so there's so much here.
link |
I'm just like playing devil.
link |
So much here, but let me play devil's advocate
link |
back in a previous conversation, right?
link |
And say, yeah, I will.
link |
Cellular automata.
link |
Cellular automata, these very, very simple things
link |
where you color squares black or white
link |
and implement rules and play them in time.
link |
And you can get these very, very complex patterns coming out.
link |
You know, there's nice rules.
link |
There are Turing complete rules and I would argue
link |
that cellular automata don't really exist on their own.
link |
They have to exist in a computing device.
link |
If that, well, that's computing devices,
link |
a piece of paper and abstraction,
link |
a mathematician drawing a grid or a framework.
link |
Now, so I would argue CAs are beautiful things,
link |
simple, going complex, but the complexity is all borrowed
link |
from the lithography, the numbers.
link |
Right, now let's take that same argument
link |
with the chemistry experiment origin of life.
link |
What you need to be able to do is go,
link |
and I'm inspired to do this,
link |
to go out and look for CAs occur in nature.
link |
You know, let's kind of, let's find some CAs
link |
that just emerge in our universe and.
link |
For people just, sorry to interrupt,
link |
for people just listening and in general,
link |
I think what we're looking at is a cellular automata
link |
where again, as Lee described,
link |
there is just binary black or white squares
link |
and they only have local information
link |
and they're born and they die.
link |
And you would think nothing interesting would emerge,
link |
but actually what we're looking at is something
link |
that I believe is called glider guns
link |
or a glider gun, which is moving objects
link |
in this multi cell space that look like they're organisms
link |
that have much more information,
link |
that have much more complexity
link |
than the individual building components.
link |
In fact, look like they have a long term memory
link |
while the individual components don't seem like
link |
they have any memory at all, which is fascinating.
link |
The argument here is that has to exist
link |
on all this layer of infrastructure, right?
link |
And though it looks simple.
link |
And then what I would make, the argument I would make
link |
if I were you, say, well, I think CAs are really simple
link |
and everywhere, is say, show me how they emerge
link |
Now this goes to the origin of life, machine.
link |
I don't think we want to do the origin of life,
link |
just any origin is good.
link |
So we do, so we literally have our sand shaker,
link |
shake the sand like massive grid of chemistry experiments,
link |
shaking sand, shaking whatever.
link |
And then because we know what we've put in,
link |
so we know how we've cheated and the same way with CA,
link |
we know how we've cheated,
link |
we know the number of operations needed,
link |
we know how big a grid we want to get this.
link |
If we could then say, okay,
link |
how can we generate this recipe in the lab
link |
and make a life form?
link |
What contingency did we need to put in?
link |
And we're upfront about how we cheated, okay?
link |
Say, oh, you had to shake it, it was a periodic,
link |
planet rotates, it's tried, comes in and out.
link |
So, and then we can start to basically say,
link |
okay, how difficult is it for these features to be found?
link |
And then we can look for extra planets and other features.
link |
So I think Sarah is absolutely right.
link |
We want to explain to people we're cheating.
link |
In fact, we have to cheat.
link |
No one has given, I'm good at writing grants,
link |
well, I used to be, I'm not very good right now,
link |
I keep getting rejected,
link |
but writing a grant for a planet in 100 million years,
link |
no grant fund there is going to give me that,
link |
but maybe money to make a kind of a grid,
link |
a computer grid, origin of life computer grid.
link |
In physical space.
link |
In physical space, and just do it.
link |
So Sarah said something which is,
link |
you can't simulate the origin of life in a computer,
link |
so like in simulation, why not?
link |
What are your, you said it very confidently,
link |
so is it possible and why would it be very difficult?
link |
Like what's your intuition there?
link |
I think it's very difficult right now
link |
because we don't know the physics,
link |
but if you go based on principles of assembly theory
link |
and you think every molecule is actually
link |
a very large causal graph, not just the molecule,
link |
then you have to simulate all the features
link |
of those causal graphs,
link |
and I think it becomes computationally intractable.
link |
You might as well just build the experiment.
link |
Because you have, in the physical space,
link |
you have all the objects with all the memories.
link |
In the computer, you would have to copy them
link |
or reconstruct them.
link |
Yeah, that's a beautifully put,
link |
and I would say that lots of people,
link |
you just don't have enough resources.
link |
It's easier to actually do the physical experiment
link |
because we are literally,
link |
I would view the physical experiment
link |
almost like a computational experiment.
link |
We're just outsourcing, it's just basically,
link |
we're just outsourcing all the matrix algebra.
link |
And on your point about the experiment
link |
being also an example of life,
link |
it's almost like you want to design,
link |
it's like all of us are lineages
link |
of propagating information across time,
link |
and so everything we do becomes part of life
link |
because it's part of that causal chain.
link |
So it's like you want to try to pinch off
link |
as much as you can of the information
link |
from your causal chain that goes into the experiment,
link |
but you can't pinch off all of it
link |
to move it to a different timeline.
link |
It's always going to be part of your timeline.
link |
But at least if you can control
link |
how much information you put in,
link |
you can try to see how much does that particular trajectory
link |
you've set up start generating its own assembly.
link |
So you know where it starts,
link |
and then you want to try to see it take off on its own
link |
when you try to pinch it off as much as possible.
link |
Quick pause, bathroom break.
link |
And now we're back.
link |
We talked about the early days of the universe
link |
when there was just stuff and no memory,
link |
not even causality.
link |
I think Lee at least implied
link |
that causality is immersion somehow.
link |
We could discuss this.
link |
What happened before this all originated?
link |
What's outside the universe?
link |
Okay, so it's not relevant, not understandable.
link |
Is it useful to even ask the question?
link |
Just because it's so hard?
link |
No, it's not hard.
link |
It's just not a question.
link |
If I can't do an experiment or even think of an experiment,
link |
the question doesn't exist.
link |
Well, no, you can't think of a lot of experiments,
link |
What I mean is I can't.
link |
Your causality graph is like,
link |
this is what we're talking about.
link |
It's like there is limits to your ability
link |
to construct experiments.
link |
I agree, but I was trying to be facetious
link |
and I'm trying to make a point
link |
because I think that if there is a causal bottleneck
link |
through which information can't propagate in principle,
link |
then it's very hard to think of an experiment,
link |
even in principle, even one that's beyond
link |
my mediocre intellect, right?
link |
I'm happy to accept that.
link |
But this is one of the things I actually do think
link |
there was something before the Big Bang
link |
because I would say that I think the Big Bang
link |
just couldn't occur and create time.
link |
Time created the Big Bang.
link |
So there was time before the Big Bang.
link |
There was no space, but there was time.
link |
But I mean, I'm just making that stuff up
link |
just to make all the physicists happy,
link |
but I think it's...
link |
Do you think that would make them happy
link |
because they would be quite upset, actually.
link |
And why would they be upset?
link |
Because they would say that time can't exist
link |
before the Big Bang.
link |
Yeah, I mean, this goes back to an argument
link |
that you might not want to have the argument here.
link |
I was talking to Sarah earlier today
link |
about an argument we had about time a long time ago.
link |
A long time in time.
link |
And what I would, it's like, I think there is this thing
link |
called time or state creation.
link |
The universe is creating states and it's outside of space,
link |
but they create space.
link |
So what I mean is you can imagine there are states
link |
being created all the time.
link |
And there is this thing called time.
link |
Time is a clock, which you can use to measure
link |
when things happen, but that doesn't mean,
link |
because you can't measure something,
link |
that states aren't being created.
link |
And so you might locally refer to the Big Bang
link |
and the Big Bang occurred at some point
link |
when those states were there.
link |
Probably there had to be enough states
link |
for the Big Bang to occur.
link |
And then, but I think that there is something wrong
link |
with our conception of how the universe was created
link |
and the Big Bang because we don't really get time.
link |
Because again, I don't want to become boring
link |
and sound like a broken record, but time is a real thing.
link |
And until I can really explain that more elegantly,
link |
I'm just gonna get into more trouble.
link |
Well, we're gonna talk about time
link |
because time is a useful measuring device for experiments,
link |
but also time is an ideal, okay.
link |
But let me first ask Sarah, what do you think?
link |
Is it a useful question to ask what happened
link |
before the Big Bang?
link |
Is it a useful question to ask what's outside the universe?
link |
So I would think about it as the Big Bang
link |
is an event that we reconstructed
link |
as probably happening in the past of our universe
link |
based on current observational data.
link |
And so the way I like to think about it
link |
is we exist locally in something called the universe.
link |
So, and going back to like the physics of existence,
link |
we exist locally in the space of all things that could exist
link |
and we can infer certain properties of the structure
link |
of where we exist locally.
link |
And one of the properties that we've inferred in the past
link |
is that there is a thing we call the Big Bang.
link |
There's some signatures of our local environment
link |
that indicate that there was a very low information event
link |
that started our universe.
link |
I think that's actually just an artifact
link |
of the structure of the assembly space
link |
that when you start losing all the memory in the objects,
link |
it looks like what we call a Big Bang.
link |
So I think it makes sense to talk about
link |
where you are locally.
link |
I think it makes sense to talk about
link |
counterfactual possibilities,
link |
what could exist outside the universe
link |
in the sense that they become part of our reasoning
link |
and therefore part of our causal chain
link |
of things that we can do.
link |
So like the multiverse in my mind exists,
link |
but it doesn't exist as a multiverse
link |
of possible universes.
link |
It exists as an idea in our minds
link |
that allows us to reason about how physics works
link |
and then to do physics differently
link |
because we reason about it that way.
link |
So I always like to recenter it on things exist,
link |
but they don't always exist like we think they exist.
link |
So when we're thinking about things outside the universe,
link |
they absolutely exist because we're thinking about them,
link |
but they don't look like the projections in our minds.
link |
They're something else.
link |
And something you said just gave an idea
link |
to go back to your question.
link |
If there was, I mean, if something caused the Big Bang,
link |
if there was some memory or some artifact of that,
link |
then of course, to answer your question,
link |
it's worth going back to that
link |
because that would imply there is something
link |
beyond that barrier, that filter.
link |
And that's what you were saying, I guess, right?
link |
I'm agnostic to what exists outside the universe.
link |
I just don't think that.
link |
I think the most interesting things for us to be doing
link |
are finding explanations that allow us to do more,
link |
like that optimism.
link |
So I tend to draw the boundary on questions I ask
link |
as being scientific ones because I find
link |
that that's where the most creative potential is
link |
to impact the future trajectory
link |
of what we're doing on this planet.
link |
It's an interesting thing about the Big Bang
link |
is basically from our current perspective
link |
of what we're able to detect,
link |
it's the time when things were forgotten.
link |
It's the time to reset from our limited perspective.
link |
And so the question is, is it useful to ever study
link |
the thing that was forgotten?
link |
Or should we focus just on the memories
link |
that are still there?
link |
Well, the point I was trying to make about the experiment
link |
is I was trying to say both things.
link |
And I think perhaps yes, from the portfolio point of view,
link |
if you could then imagine what was forgotten
link |
and then work forwards,
link |
you will have different consequences.
link |
So then it becomes testable.
link |
So as long as we can find tests,
link |
then it's definitely worth thinking about.
link |
What I don't like is when physicists say
link |
what happened before the Big Bang
link |
and before, before, before,
link |
without giving me any credible conjecture
link |
about how would we know the difference?
link |
But the way you framed it is quite nice.
link |
It's like, what have we forgotten?
link |
Is there room for God in assembly theory?
link |
I like arguments for a necessary being better than God.
link |
Well, I think I said it earlier.
link |
What's a necessary being?
link |
What's a necessary?
link |
Like something that has to exist.
link |
Oh, so you like, I mean, you like the shortest path.
link |
Like does God need?
link |
I mean, well, you can go back to like Thomas Aquinas
link |
and arguments for the existence of God.
link |
But I think most of the interesting theological arguments
link |
are always about whether something has to exist
link |
or there was a first thing that had to exist.
link |
But I think there's a lot of logical loopholes
link |
in those kind of arguments.
link |
Well, so God here, meaning the machine
link |
that creates, that generates the stuff.
link |
But God, so what I was trying to say earlier is that.
link |
Isn't that just the universe though?
link |
Well, yeah, well, but there's a difference between,
link |
I sort of imagine like a black box, like a machine.
link |
I mean, I would be more comfortable calling that God
link |
because it's a machine.
link |
You go into a room and there's a thing with a button.
link |
Yeah, I don't like the great programmer in the sky version.
link |
Yeah, but if it's more kind of like,
link |
I don't like to think of, if you look at a cellular automata,
link |
if it's the cells and the rules,
link |
that doesn't feel like God
link |
that generates a bunch of stuff.
link |
But if there's a machine like that does,
link |
that runs the cellular automata and set the rules,
link |
then that feels like God.
link |
That sort of, in terms of terminology.
link |
So I wonder if there's like a machine
link |
that's required to generate this universe.
link |
That's very sort of important for running this in the lab.
link |
So as I said earlier, I think I said this earlier,
link |
that I can't remember the phrase, but something like,
link |
I mean, does God exist in our universe?
link |
Where does God exist?
link |
God at least exists in abstraction in our minds,
link |
particularly of people who have religious faith
link |
But let's then take, but you're talking a little bit more
link |
about generics, say, well,
link |
is there a mechanism beyond the universe you're calling God?
link |
I would say God did not exist at the beginning,
link |
but he or she does now.
link |
Because I'm saying the mechanism.
link |
Well, you don't know that he didn't exist in the beginning.
link |
So like this could be us in our minds trying to,
link |
like just listening to gravitational waves,
link |
detecting gravitational waves.
link |
It's the same thing.
link |
It's us trying to go back further and further
link |
into our memories to try to understand the machines
link |
that make up, that make up us.
link |
And so it's possible that we're trying to grasp
link |
at possible kind of what kind of machines could create.
link |
There's always a tweet.
link |
There's always a tweet.
link |
If the universe is a computer, then God must have built it
link |
because computers need creators.
link |
And then Joshe Bach replied,
link |
since there's something rather than nothing,
link |
perhaps existence is the default.
link |
If existence is the default, then many computers exist.
link |
Creator gods are necessary computers,
link |
unnecessarily computers too.
link |
I'm very confused by that, but that's an interesting idea
link |
that existence is the default versus nonexistence.
link |
I agree with that, but the rest is not.
link |
And then Lee responds,
link |
perhaps this reasoning is incomplete.
link |
That's how scientists talk trash each other
link |
on Twitter apparently.
link |
Which part don't you agree with?
link |
When he said if existence is default,
link |
then many computers exist,
link |
this comes back to the inventor and discovery argument.
link |
I would say the universe at the beginning wasn't capable
link |
of computation because there wasn't enough technology,
link |
So what you're saying is if God is a mechanism,
link |
so I might actually agree,
link |
but then the thing is lots of people see God
link |
as more than a mechanism.
link |
For me, God could be the causal graph in assembly theory
link |
that creates all the stuff that the memories we know.
link |
And the fact that we can even relate to each other
link |
is because we have the same, we share that heritage.
link |
And why we love each other
link |
or we like to see God in each other
link |
is it's just we know we have a shared existence.
link |
So if the God is the mechanism
link |
that created this whole thing,
link |
I think a lot of people see God in a religious sense
link |
as that mechanism also being able to communicate
link |
with the objects it creates.
link |
And if it's just the mechanism,
link |
we won't be able to communicate with the objects it creates.
link |
It can only create.
link |
You can't interact with the...
link |
Well, there's versions of God that create the universe
link |
For some religions.
link |
The first spark, yeah.
link |
But I think I liked your analogy
link |
of the machine and the rules, right?
link |
But I think part of the problem is,
link |
I mean, we have this conception
link |
that we can disentangle the rules
link |
from the physical substrate, right?
link |
And that's the whole thing about software and hardware
link |
being separate or the way Newton wrote his laws
link |
that there was some,
link |
like they exist outside the universe.
link |
They're not actually a feature of the universe.
link |
They don't have to emerge out of the universe itself.
link |
So I think if you merge your two views,
link |
then it gets back to the God is the universe.
link |
And then I think the deeper question
link |
is why does it seem like there's meaning and purpose?
link |
And if I think about the features of the universe
link |
that give it the most meaning and purpose,
link |
those are what we would call
link |
the living components of the universe.
link |
So if you wanted to say God is a physically real thing,
link |
which you were saying
link |
is like an emergent property of our minds,
link |
but I would just say the way the universe
link |
creates meaning and purpose,
link |
there is really a physics there.
link |
It's not like a illusory thing.
link |
And that is just what the physics of life is.
link |
Is it possible that we've forgotten
link |
much of the mechanisms that created the universe?
link |
So like, so basically, you know,
link |
whatever, if God is that mechanism,
link |
we just leave parts of that behind.
link |
Well, but the universe is constantly generating itself.
link |
So if God is that mechanism,
link |
it would be that that would still be active today.
link |
I don't, like, I'm agnostic,
link |
but if I recall the things I believe in God
link |
in the way that some people talk about God,
link |
I would say that God is, you know,
link |
like in the universe now, it's not an absent thing.
link |
So I think there's a mislabeling here
link |
because you're, I mean, I mean, I'm a professional idiot,
link |
actually, but, but, um.
link |
You should put that on your CV.
link |
Professionally, not recreationally or amateur,
link |
but professionally, you're paid for it.
link |
I would say if you were talking about God,
link |
I mean, again, I'm way out, way out of my depth here,
link |
and I almost feel uncomfortable.
link |
Yeah, but I feel quite uncomfortable articulating,
link |
For me, a lot of people that think of God as a consciousness
link |
and a reasoning entity that actually has causal power,
link |
and you're, and so you're, it's like,
link |
then you're saying like gravity could be God
link |
or time could be God.
link |
I mean, I think for me, for my conception of time
link |
is probably as fundamental as God
link |
because it gave rise to human intelligence and consciousness
link |
in which we can have this abstract notion of God.
link |
So I think that you're maybe talking about God
link |
in a very mechanistic, kind of unsophisticated sense,
link |
whereas other people say that God is more sophisticated
link |
and got all this, you know, feelings and love
link |
and, you know, and this abstracting ability.
link |
So is that what, or do you mean that?
link |
Do you mean God as in this conscious entity
link |
that decided to flick the universe into existence?
link |
Well, one of the features that God would have
link |
is the ability to flick the universe into existence.
link |
I, you know, like Windows 95,
link |
I don't know if God is Windows 95 or Windows XP
link |
or Windows 10, I don't know the full feature set.
link |
So at the very least, you have to flick the universe
link |
into existence, and then other features might include
link |
ability to interact with that universe in interesting ways,
link |
and then how do you interact with the universe
link |
in interesting ways?
link |
You have to be able to speak the language
link |
of its different components.
link |
So in order to interact with humans,
link |
you have to know how to act humanlike.
link |
So I don't know, but it seems like
link |
whatever mechanism created the universe
link |
might want to also generate local pockets of mechanisms
link |
that can interact with that.
link |
Like God was lonely?
link |
Yeah, I mean, it could be just a teenager
link |
and another just playing a video game.
link |
Well, I was gonna say, I mean, I don't,
link |
so this is referring to our origin of life engine.
link |
It's like, I don't believe in God,
link |
but that doesn't mean I don't wanna be one.
link |
I don't wanna make a universe and make a life form,
link |
but that may be rude to people who have
link |
dear religious beliefs.
link |
What I mean by that is if we are able to create
link |
an entirely new life form, different chemistry,
link |
different culture, what does it make us?
link |
By that definition, it makes us gods, right?
link |
I mean, like when you have children,
link |
you're like one of the magical things of that
link |
is you're kind of mini gods.
link |
I mean, first of all, from a child's perspective,
link |
parents are gods for quite a while.
link |
And then, I mean, in the positive sense,
link |
there's a magic to that.
link |
That's why I love robotics,
link |
is you instill life into something,
link |
and that makes you feel godlike in a sort of positive way.
link |
Being a creator is a positive feeling.
link |
Creator, yeah, exactly, on a small scale.
link |
And then God would be a creator
link |
at the largest possible scale, I suppose.
link |
Okay, you mentioned offline the Assembletron.
link |
What's an Assembletron?
link |
This is an early idea of something you're thinking about.
link |
So Sarah's team, well, I think Sarah's team
link |
are interested in using AI to understand life.
link |
My team is, and I'm wondering if we could apply
link |
the principles of assembly theory,
link |
that is the causal structure that you get
link |
with assembly theory, and hybridize it,
link |
and make a new type of neuron, if you like.
link |
I mean, there are causal neural networks out there,
link |
but they are not quite the architecture
link |
of what I would like.
link |
I would like to associate memory bits with,
link |
basically, I'd like to make a,
link |
rather than having an ASIC for neural networks,
link |
I wanna make an ASIC for assembly networks, right?
link |
So can you say that again?
link |
Assembly networks.
link |
So what is a thing with an input and an output,
link |
and it's like a neural network type of thing,
link |
what does it do exactly?
link |
What's the input, what's the output?
link |
So in this case, so if you're talking about
link |
a general neural network, I mean,
link |
in general neural network, you can train it
link |
on any sort of data, right, depending on the framework,
link |
whether it's like text, or image data, or whatnot.
link |
And that's fine, but there's no causal structure
link |
associated with that data.
link |
Now just imagine, rather than, you know,
link |
let's say we're gonna classify a difference
link |
between cat and dog, right?
link |
Classic cat and dog neural network.
link |
What about if the system understood the assembly space,
link |
it created the cat and the dog,
link |
and rather than guessing what was happening,
link |
and training on those images,
link |
and not understanding those features,
link |
you almost like, you could imagine doing a,
link |
going back a step and doing, and training,
link |
going back a step and doing the training,
link |
going back a step, back a step, back a step,
link |
and I wonder if that is actually the origin of intelligence,
link |
or how we'll crack intelligence.
link |
Because we need to, because we'll create
link |
the entire graph of events, and be able to kind of
link |
look at cause and effect across those graphs.
link |
I'm explaining it really badly,
link |
but it's a gene of an idea, and I'm guessing
link |
very smart, very rich people in AI are already doing this.
link |
Trying to not generate cats and dogs,
link |
but trying to generate things of high assembly index.
link |
Yeah, and I think, and also using causal graphs
link |
in neural networks, and machine learning, and deep learning,
link |
maybe building a new architecture.
link |
I'm just wondering, is there something
link |
we can get out of assembly theory
link |
that allows us to rebuild current machine learning
link |
architectures to give causation more cheaply?
link |
I mean, I don't know if that's what you,
link |
we've been inventing this for a little while,
link |
but we're trying to finish the theory paper first
link |
before we do anything else.
link |
Yeah, you also wanna have, say, goal directed behavior
link |
in neural networks, then assembly theory
link |
is a good framework for doing that.
link |
Daniel's been thinking about that a lot.
link |
And I think it's a really interesting idea
link |
that you can map concepts from how neural networks learn
link |
to thinking about goal directed behavior
link |
as a learning process.
link |
That you're learning a specific goal.
link |
The universe is learning a goal
link |
when it generates a particular structure,
link |
and that you can map that physical structure
link |
in a neural network.
link |
Well, in a neural network, you're designing the goal.
link |
In biology, I mean, people are not supposed to use
link |
teleological language in biology, which is ridiculous,
link |
but, because goals are real things.
link |
They're just post selected.
link |
So you can talk about goals after the fact.
link |
Once a goal emerges in the universe,
link |
that physical entity has a goal.
link |
But Lee and I came up with a test for,
link |
like a Turing test for goal directed behavior
link |
based on the idea of assembly.
link |
We have to formalize it still,
link |
but I would like to write a paper on it.
link |
But the basic idea is if you had two systems
link |
that were completely equivalent,
link |
you know, like in the instantaneous,
link |
like physical experimental setup.
link |
So Lee has to figure out how to do this.
link |
But there was something that would be different
link |
And there was a symmetry breaking you observe
link |
in the present based on that possibility
link |
of that future outcome.
link |
Then you could say that that system
link |
had some representation of some kind of goal in mind
link |
about what it wanted to do in the future.
link |
And I, so goals are interesting
link |
because they don't exist as instantaneous things.
link |
They exist across time,
link |
which is one of the reasons that assembly theories
link |
may be more naturally able to account
link |
for the existence of goals.
link |
So goals are, they only exist in time,
link |
or they manifest themselves in time
link |
through, you said, symmetry breaking.
link |
So it's almost like, imagine,
link |
like if representations in your mind are real, right?
link |
And you can imagine future possibilities,
link |
but imagine everything else is physically equivalent.
link |
And the only thing that you actually change
link |
your decision based on is what you model
link |
as being the future outcome.
link |
Then somehow that representation in your mind
link |
of the future outcome becomes causal
link |
to what you're doing now.
link |
So there's kind of like retro causal effect,
link |
but it's not actually retro causal.
link |
It's just that your assembly space
link |
is actually includes those possibilities
link |
as part of the structure.
link |
It's just, you're not observing all the features
link |
of the assembly space in the current moment.
link |
Or the possibilities exist, but they don't become a goal
link |
until they're realized.
link |
So one of the features of assembly space
link |
that's super interesting,
link |
and it's easier to envision with like Legos, for example,
link |
is if you're thinking about an assembly space,
link |
you can't observe the entire assembly space
link |
in any instant in time.
link |
So if you imagine a stack of Legos,
link |
and you wanna look at the assembly space of a stack of Legos,
link |
you have to break the Legos apart,
link |
and then you look at all the possible ways
link |
of building up the original object.
link |
So now you have in your mind
link |
the goal of building that object,
link |
and you have all the possible ways of doing it.
link |
And those are actual physical features of that object,
link |
but that object doesn't always exist.
link |
What exists is the possibility of generating it.
link |
And the possibilities are always infinite.
link |
Well, for that particular object,
link |
like it has a well defined assembly space.
link |
And I guess what I'm saying is that object
link |
is the assembly space,
link |
but you actually have to unpack that object across time
link |
to view that feature of it.
link |
It's only an observable across time.
link |
The term goal is such a important
link |
and difficult to explain concept, right?
link |
Because what you want is a way is like,
link |
I think only conscious beings can have conscious goals.
link |
Everything else is doing selection.
link |
And but selection does invent goals.
link |
And in a way that the way that biology reinterprets the past
link |
in the present is kind of helped you to understand
link |
there was a goal in the past now, right?
link |
It's kind of like goals only exist back in time.
link |
So first of all, only conscious beings
link |
can have conscious goals.
link |
I'm not even gonna touch that one.
link |
Go for it, come on.
link |
The line between conscious goals
link |
and non conscious goals, exactly.
link |
And also maybe just on top of that,
link |
you said a Turing test for goal directed behavior.
link |
What does a Turing test potentially look like?
link |
So if you've got two objects, we were thinking about this.
link |
So we actually got some funding
link |
to work to go on two teams.
link |
So I'm trying to do, and part of this
link |
is I'm trying to do a bit of theory
link |
and Sarah is teaching me a bit of theory
link |
and Sarah is trying to design experiments
link |
and I'm teaching experiments.
link |
Cause I think it's really good for us to have that.
link |
So say, when would a, so that's good.
link |
I like this, I'm sure we're using Dan Dennett essay.
link |
Yeah, and I can explain why we wouldn't want to call it
link |
a Turing test after, but Dan Dennett.
link |
So Dan Dennett wrote this really nice essay
link |
about herding cats and freewill inflation.
link |
The title is so brilliant.
link |
That's the actual title?
link |
That's the title, yeah.
link |
Herding cats and freewill inflation.
link |
Yeah, something like that.
link |
I mean, it's not, maybe not.
link |
No, I think that's right.
link |
So if you've got a, let's imagine you've got two objects
link |
on a hillside, okay?
link |
And it just happens to be a snowy hill
link |
and let's just say you see an object
link |
get rolling down the hill.
link |
Or you, and the rock rolls down the hill
link |
but the start goes to the end.
link |
How do you know that object's had a goal?
link |
Now you unveil the object
link |
and you'll see it's actually a skier.
link |
And the skier starts at the top and goes down the bottom.
link |
Then you look at the rock.
link |
Rock rolls down the hill and goes to the bottom.
link |
How can you tell the difference between the two?
link |
So, and what Dan says is like, well, this is clear.
link |
The skier's in control.
link |
And because they're adjusting the trajectory,
link |
so there's some updating going on.
link |
Then the only way you could really do that
link |
is if you put the skier back to the top of the hill again,
link |
they would tend to start roughly in the same space
link |
and probably take all that complex set of trajectories
link |
and end up pretty much at the same finish point, right?
link |
With plus or minus a few meters.
link |
Whereas if it was just a random rock
link |
going down to a random trajectory, that wouldn't happen.
link |
And so what Sarah and I were kind of doing
link |
when we were writing this grant,
link |
we were like, we need to somehow instantiate the skier
link |
and the rock in an experiment.
link |
And then say, okay, when does the object,
link |
so for an object to have a goal, it has to have an update.
link |
It has to have some sensing and some kind of,
link |
you know, inbuilt actuation to respond to the environment.
link |
And then we just have to iterate on that.
link |
And maybe Sarah, you can then fill in the Turing test part.
link |
Well, yeah, I guess the motivation for me
link |
was slightly different.
link |
So I get really frustrated about conversations
link |
about consciousness as most people do.
link |
You know, a lot of people are,
link |
which is not necessarily related to free will directly
link |
or to this goal directed behavior.
link |
But I think there's a whole set of bundled
link |
and related topics here.
link |
But I think for me, I was, you know,
link |
everybody's always interested in explaining
link |
intrinsic experience and quantifying intrinsic experience.
link |
And there's all sorts of problems with that
link |
because you can never actually be another physical system.
link |
So you can't know what it's like
link |
to be another physical system.
link |
So I always thought there must be some way
link |
of getting at this problem about
link |
if an agent or an entity is conscious
link |
or at least has internal representations
link |
and those are real physical things,
link |
that it must have causal consequences.
link |
So the way I would ask the question of consciousness
link |
is not, you know, what is it like intrinsically?
link |
But if things have intrinsic experience,
link |
is there any observable difference from the outside
link |
about the kind of causation
link |
that that physical system would enact in?
link |
And for me, the most interesting thing that humans do
link |
is have imagination.
link |
So like we can imagine rockets centuries
link |
before we build them.
link |
They've become real physical things
link |
because we imagine them.
link |
And people might disentangle that from conscious experience,
link |
but I think a lot of the sort of imagination we do
link |
is actually a conscious process.
link |
So then this becomes a question of
link |
if I were observing systems and I said,
link |
one had an internal representation,
link |
which is slightly different
link |
than a conscious experience, obviously.
link |
So I'm entangling some concepts,
link |
but it's a loose set of thought experiments.
link |
Then how, and I set them up
link |
in a physically equivalent situation.
link |
Would it be the case that
link |
there would be experimental observables associated with it?
link |
And that became the idea of trying to actually
link |
measure for internal representation and conscious.
link |
So Turing basically didn't wanna do that.
link |
You just wanted the machine that could emulate
link |
and trick you into having the behavior,
link |
but never dealt with the internal experience
link |
because he didn't know how to do that.
link |
And I guess I was wondering,
link |
is there a way to set up the experiment
link |
where you could actually test for that?
link |
For imagination that led to the thing.
link |
That there was something internal going on,
link |
some kind of inner world, as people say,
link |
or you could say, it actually is an agent,
link |
it's making decisions, it has an internal representation.
link |
And whether you say that's experience or not
link |
is a different thing, but at least the feature
link |
that there's some abstraction it's doing
link |
that's not obvious from looking at the physical substrates.
link |
Do you think it's possible to do that kind of thing?
link |
One of the compelling things about the Turing test
link |
is that defining intelligence,
link |
defining any complicated concept
link |
as a thing like observing it from the surface
link |
and not caring about what's going on deep inside
link |
because how do you know?
link |
So the idea is exactly that.
link |
So what we're trying to do,
link |
the Turing test for goal directedness
link |
is literally take some objects
link |
that clearly don't have any internal representation,
link |
grains of sand blowing on the beach or something,
link |
and I don't know, a crab wandering around on the beach
link |
and then generating an experiment
link |
where literally the experiment generates an entity
link |
that literally has no internal representation to sand,
link |
these are oil droplets actually,
link |
what we've got in mind, a robot that makes oil droplets.
link |
But then what we wanna try and do
link |
is train the oil droplets to be like crabs,
link |
give them an internal representation,
link |
give them the ability to integrate information
link |
from the environment so they remember the past,
link |
are in the present, and can imagine a future.
link |
And in a very limited way, their kind of game engine,
link |
their limited simulation of the world
link |
allows them to then make a decision.
link |
They're objects across time.
link |
So then you would run a bunch of crabs
link |
like over and over and over and over?
link |
How many crabs, Lee?
link |
How many, is there, what's,
link |
because you have to have a large number of crabs,
link |
what does your theory say, is there a mathematical?
link |
We're working on it, I mean, this is literally.
link |
Limit, crab limit.
link |
There's literally a.
link |
There's literally a.
link |
What's the herding cats have to do?
link |
Oh, that's random, wait, what's cats,
link |
in the title by Daniel Dennett,
link |
Herding Cats and the Free Will Inflation.
link |
What does herding cats mean,
link |
what does free will inflation mean?
link |
So this, I love this essay,
link |
because it explained to me
link |
how I can live in a deterministic universe,
link |
but have, not free will, but have freedom.
link |
And also it helped me explain
link |
that time needed to be a real thing in this universe.
link |
So what basically Dan was saying here is like,
link |
how do you, how do these cats appear
link |
to just do what they want, right?
link |
And if you live in a deterministic universe,
link |
why do the cats do these things?
link |
You know, aren't they just, isn't it all obvious?
link |
And how does free will inflate the universe?
link |
And for me, I mean, probably I love the essay
link |
because my interpretation of the essay
link |
in assembly theory makes complete sense.
link |
Because you need an expanding universe
link |
in assembly theory to create novelty
link |
that you search for,
link |
that then when you find something interesting
link |
and you keep doing it because it's cool
link |
and it gives you an advantage,
link |
then it appears in the past to be a goal.
link |
So what does, in assembly theory,
link |
the expansion of the universe look like?
link |
What are we talking about?
link |
Why does the expansion of the universe
link |
give you more possibilities of novelty and cool stuff?
link |
So for me, I don't think about the universe
link |
in terms of big bang and space.
link |
I think about it in terms of the big memory expansion.
link |
That you have one, you only have the ability
link |
to store one bit of information,
link |
so then you can't do very much.
link |
So what the universe has been doing since forever,
link |
it's been creating more,
link |
it's been increasing the size of its RAM, okay?
link |
So it's like one megabyte, two megabyte,
link |
three megabyte, four megabytes, all the way up.
link |
And so the more RAM you have,
link |
the more you can remember about the past,
link |
which allows you to do cooler things in the future.
link |
So if you can remember how to launch a rocket,
link |
then you might be able to imagine how to land a rocket,
link |
and then relaunch, reland, and carry on.
link |
And so you're able to expand the space
link |
and remember the past.
link |
And so that's why I think it's very important.
link |
But not a perfect memory.
link |
It's an interesting question,
link |
whether there's some forgetting that happens
link |
that might increase.
link |
Is the expansion of the forgetting, at some point,
link |
accelerate faster than the remembering?
link |
I think that that's a very important thing
link |
that probably intelligence does,
link |
and we're gonna learn in machine learning about,
link |
because you want machine learning right now,
link |
or artificial intelligence right now,
link |
doesn't have memory right,
link |
but you want the ability to,
link |
or not for, if you want to get to human like consciousness,
link |
you need to have the ability, I suppose, to remember stuff
link |
and then to selectively forget stuff
link |
so you can re remember it and compress it.
link |
Arguably, the way that we come up with new physical laws.
link |
I think that there is a great deal to be gained
link |
from having the ability to remember things,
link |
but then when you forget them,
link |
you can then have a,
link |
you can basically do the simulation again
link |
and work out if you get to that compressed representation.
link |
So that's in cycles.
link |
So cycles of remembering and forgetting
link |
are probably important,
link |
but there shouldn't be excuse to have a universe
link |
with no memory in it.
link |
The universe is gonna remember that it forgot,
link |
but just not tell you.
link |
I'm looking at this paper
link |
and it's talking about a puppet controlling a puppet
link |
controlling a puppet controlling a puppet controlling
link |
a puppet controlling a puppet,
link |
conceptually easy to understand,
link |
but physically impossible,
link |
as physically impossible as predicting a fair coin toss.
link |
I don't know what he's talking about,
link |
but there's pictures of puppets controlling puppets.
link |
there's a few things I wanna ask,
link |
but we brought up time quite a bit.
link |
You guys tweet about time quite a bit.
link |
What is time in all of this?
link |
We kind of mentioned it a bunch.
link |
Is it not important at all in terms of,
link |
is it just a word?
link |
Should we be talking about causality mostly?
link |
Like Sarah, what do you think?
link |
Is, we've talked about like memories.
link |
Is that the fundamental thing
link |
that we should be thinking about?
link |
And time is just a useful measurement device or something
link |
Well, there's different concepts of time, right?
link |
So I think in assembly theory,
link |
when we're talking about time,
link |
we're talking about the ordering of things.
link |
So that's the causal graph part.
link |
And so then the fundamental structure of the universe
link |
is that there is a certain ordering
link |
and certain things can't happen till other things happen.
link |
But usually when we colloquially talk about time,
link |
we're talking about the flow of time.
link |
And I guess Lee and I were actually debating
link |
about this this morning.
link |
So in talking on it, walking on the river here,
link |
which is a very lovely spot for talking about time,
link |
but that when the universe is updating,
link |
it's transitioning between things that exist now
link |
and things that exist now.
link |
That's really the flow of time.
link |
So you have to separate out those concepts at bare minimum.
link |
And then there's also an arrow of time
link |
that people talk about in physics,
link |
which is that time doesn't appear to have a directionality
link |
in fundamental physics, but it does to us, right?
link |
Like we can't go backwards in time.
link |
And usually that would be explained in physics
link |
in terms of, well, there's a cosmological arrow of time,
link |
but there's also the thermodynamic arrow of time
link |
of increasing entropy.
link |
But what we would say in assembly theory
link |
is that there is a clear directionality.
link |
The universe only runs in one direction,
link |
which is why some things, it's easy to make,
link |
if the universe runs in one direction,
link |
it's easy to make processes look reversible.
link |
For example, if they have no memory,
link |
they're easy to run forward and backwards,
link |
which is why the laws of physics that we have now
link |
look the way they do, because they involve objects
link |
that have no memory.
link |
But when you get to things like us,
link |
it becomes very clear that the universe
link |
has a directionality associated to it.
link |
So it's not reversible at all.
link |
It's the no man ever steps in the same river.
link |
I just have to bring that out
link |
because you walked on the river.
link |
No man ever steps in the same river twice
link |
before he's not the same river and he's not the same man.
link |
So it's not reversible, any of the same.
link |
No, but reversibility is an emergent property, right?
link |
So we think of the reversibility laws as being fundamental
link |
and the irreversibility as being emergent.
link |
But I think what we would say from how we think about it,
link |
and it certainly seems to be the case
link |
for our perception of time,
link |
but also what's happening in biological evolution,
link |
you can make things reversible,
link |
but it requires work to do it.
link |
And it requires certain machines
link |
to run it forward and backward.
link |
And Chiara Marletto is working some interesting ideas
link |
on constructor theory related to that,
link |
which is a totally different set of ideas.
link |
You can travel back in time sometimes.
link |
You can't travel actually back in time,
link |
but you could reconstruct things
link |
that have existed in the past.
link |
You're always moving forward in time,
link |
but you can cycle through.
link |
Like, I mean, I can...
link |
Can I clarify what you just said?
link |
Yeah, yeah, go for it.
link |
Quickly, you travel forward in time to travel back.
link |
Thank you, that really clarified it.
link |
What Sarah's saying is you don't go back in time,
link |
you recreate what happened in the past in the future
link |
and inspect it again.
link |
So in that local pocket of time,
link |
it's as if you travel back in time.
link |
How is that not traveling back in time?
link |
Because you're not going back
link |
to your same self back in time.
link |
You're creating that in the future.
link |
But everything else is the same as it was in the past.
link |
No, no, no, no, it's not in registry.
link |
I mean, it goes back to the big question I'm saying.
link |
I mean, this is something I was trying to look up today
link |
when we first had this discussion
link |
and I was talking to Sarah on Skype and said,
link |
by the way, time is the fundamental thing in the universe.
link |
She almost hung up on me.
link |
Right, but you can even...
link |
I mean, if you wanna make an analogy to computation,
link |
and I think Charles Bennett actually has a paper on this
link |
like about reversible computation
link |
and reversible Turing machines.
link |
In order to make it reversible,
link |
you have to store memory to run the process backwards.
link |
So time is always running forward in that
link |
because you have to write the memory.
link |
You can't erase the memory.
link |
You can erase the memory,
link |
but the point when you go back to zero, right?
link |
But the whole point is that in order to have a process
link |
that even runs in both directions,
link |
you have to start talking about memory
link |
to store the information to run it backwards.
link |
So you can't really then...
link |
You can't have it exactly how it was in the past.
link |
You have extra stuff, extra baggage always.
link |
A really important thing that I want to say on this,
link |
I think if I try and get it right,
link |
I have to say that if you can think
link |
that the universe is expanding
link |
in terms of the number of boxes
link |
that it has to store states, right?
link |
And this is where the directionality of the universe
link |
comes from, everything comes from.
link |
You could erase what's in those boxes,
link |
but the fact you've now got so many boxes at time now
link |
in this present, there's more of those boxes
link |
than there were in the past.
link |
See, but the boxes aren't physical boxes.
link |
They're not space or time.
link |
Why is the number of boxes always expanding?
link |
It's very hard to imagine this
link |
because we live in space.
link |
So what I'm saying, which is I think probably correct,
link |
is that we just, let's just imagine for a second,
link |
there is a nonlocal situation,
link |
but there are these things called states
link |
and that the universe irrespective
link |
of whether you measure anything,
link |
there is a universal, let's call it a clock
link |
or a state creator.
link |
Maybe we can call it, that's why maybe you can call it God,
link |
but let's call it a state creator
link |
where the universe is expanding
link |
in the number of states it has.
link |
Why are you saying it's expanding though?
link |
Is that obvious that it's expanding?
link |
It's obvious because that's where the,
link |
That's a source of novelty.
link |
It's a source of novelty
link |
and it also explains why the universe is not predictable.
link |
How do you know it's not predictable?
link |
I just like interrupting you.
link |
Sorry, it's fun, because you're struggling.
link |
I'm struggling because I'm trying to be
link |
as concrete as possible and not sound like I'm insane.
link |
And I'm not insane.
link |
It's obvious because you,
link |
So as a chemist, I grew into the world
link |
understanding irreversibility.
link |
Irreversibility is all I knew.
link |
And when people start telling me
link |
the universe is actually reversible, it's a magic trick.
link |
We can use time to do it.
link |
So what I mean is that the second law
link |
is really the magical.
link |
But why does it need to be magical?
link |
The universe is just asymmetric.
link |
All I'm saying is the universe is asymmetric
link |
in the state production and we can erase those states,
link |
but we just have more computational power.
link |
So what I'm saying is that the universe's
link |
deterministic horizon,
link |
this is one of the reasons we can't live in a simulation,
link |
by the way, you can't live in a simulation.
link |
The irreversibility.
link |
Yeah, yeah, so basically every time
link |
you try and simulate the universe,
link |
in this, you know, I live in a simulation,
link |
the universe is expanded in states.
link |
You're like, oh damn it,
link |
I need to make my computer bigger again.
link |
And every time you try and contain the universe
link |
in the computation,
link |
because it's got bigger in number of states.
link |
And so I'm saying the fact the universe has novelty in it
link |
is going to turn out experimentally to be proof
link |
that time, as I've labeled it, is fundamental
link |
and exists as a physical thing that creates space.
link |
Okay, so if you can prove that novelty
link |
is always being created,
link |
you're saying that it's possible to also then prove
link |
that it's always expanding in the state space.
link |
Those are things that have to be proven.
link |
That's what we're working on experiments for, yeah.
link |
And you're trying to, like by looking at the sliver
link |
of reality, show that there's always novelty being generated.
link |
Yeah, because if we go live in a universe
link |
that conventional physicists would live in,
link |
it's a big lookup table of stuff and everything exists.
link |
I want to prove that that book doesn't exist,
link |
it's continuously being added pages on.
link |
So all I'm saying, if the universe is a book,
link |
we started, the universe at the beginning only had no pages
link |
and had one page, another page, another page.
link |
Whereas a physicist would now say all the pages exist
link |
and we could in principle access them.
link |
I'm saying that is fundamentally incorrect.
link |
Do you know what's written in this book?
link |
The free will question.
link |
Is there room for free will in this view
link |
of the universe is generating novelty
link |
and getting greater and greater assembly structures built?
link |
Why, what's the source of free will in this?
link |
Well, I think it depends on what you mean by free will, but.
link |
Yeah, well, please, there's a lot.
link |
I think what I'm interested in
link |
as far as the phenomena of free will is,
link |
do we have individual autonomy and agency?
link |
And when I do things, is it really me
link |
or is it my atoms that did it?
link |
And that's the part that's interesting to me.
link |
I guess there's also the determinism versus randomness part.
link |
But the way I think about it is like each of us
link |
are like a thread or like an assembly space
link |
through this giant possibility space.
link |
And it's like we're moving on our own trajectory
link |
through that space and that is defined by our history.
link |
So we're sort of causally contingent on our past,
link |
but also because of the sort of intersection
link |
of novelty generation, it's not completely predetermined
link |
And so then you have the causal control
link |
of the determinism part that you are your causal history
link |
and there's some determinism from that past,
link |
but there's also room for creativity.
link |
And I think it's actually necessary
link |
that something like free will exists
link |
if the universe is gonna be as creative as possible.
link |
Because if I were all intelligent being,
link |
inventing a universe, and I wanted it to have
link |
maximum number of interesting things happen,
link |
again, we should come up with the metric of interesting,
link |
but generating, yes, I know,
link |
generating maximal possibilities,
link |
then I would want the agents to have free will
link |
because it means that they're more individual,
link |
like each entity actually is a different causal force
link |
And it's intrinsic and local property of that system.
link |
There's a greater number of distributed agents?
link |
Like are you always creating more and more individuality?
link |
I would say you're creating more causal power, but.
link |
So causal power, the word consciousness,
link |
is the causal power somehow correlated with consciousness?
link |
I mean, that's why I have this conception
link |
of consciousness being related to imagination,
link |
because the more that we can imagine can happen
link |
and the more counterfactual possibilities you have in mind,
link |
the more you can actually implement.
link |
And somehow free will is also at the intersection
link |
of the counterfactual becoming the actual.
link |
So can you elaborate on that a little bit,
link |
that consciousness is imagination?
link |
I don't know exactly how to articulate it,
link |
and I'm sure people will aim at certain things I'm saying,
link |
but I think the language is really imprecise,
link |
so I'm not the best way to.
link |
It's really interesting, like what is imagination
link |
and what role does it play in the human experience,
link |
in the experience of any agent?
link |
Yes, I love imagination.
link |
I think it's like the most amazing thing we do.
link |
But I guess one way I would think about it is,
link |
we talked about the transition to life
link |
being the universe acquiring memory,
link |
and life does something really interesting.
link |
You just think about biology generally.
link |
It remembers states of the past to adapt
link |
to things that happen in the future.
link |
So the longer life has evolved on this planet,
link |
the deeper that past is, the more memory we have,
link |
the more kinds of organisms and things.
link |
But what human level intelligence has done
link |
is quite different.
link |
It's not just that we remember states
link |
that the universe has existed in before,
link |
it's that we can imagine ones that have never existed,
link |
and we can actually make them come into existence.
link |
And I think that's the most unique feature
link |
about the transition to whatever we are
link |
from what life on this planet has been doing
link |
for the last four billion years.
link |
And I think it's deeply related to the phenomenon
link |
we call consciousness.
link |
Yeah, I was gonna, I mean, just agree with that.
link |
I think that consciousness is the ability
link |
to generate those counterfactuals.
link |
Now, whether you can say, you know,
link |
are there degrees of consciousness?
link |
I mean, I'm sorry, panpsychist,
link |
but electrons don't have counterfactuals,
link |
although they do have some kind of,
link |
they are able to search a space and pathways.
link |
But I think that there is a very concrete,
link |
or concrete, there's a very specific property
link |
that humans have, and I don't know if it's unique to humans.
link |
I mean, maybe dogs can do it, and birds can do it, right?
link |
And where they are basically solving a problem,
link |
because consciousness was invented,
link |
or this abstraction was invented by evolution
link |
for a specific reason.
link |
And so look, one of the reasons why I came to the conclusion
link |
that time was fundamental was actually because
link |
Sarah and I had a completely different.
link |
The most heated debate on Skype chat ever.
link |
No, no, no, we had to, I was like.
link |
No, no, it goes back to the free will thing.
link |
So I think that, although I've changed my view a bit,
link |
because there's some really interesting physicists
link |
out there who talks about how the measurement problem
link |
in Newtonian space, but I don't want to go there just now,
link |
because I think I'll mess it up.
link |
But briefly, I could not see how the universe,
link |
how we can have free will.
link |
And I mean, this is really boring,
link |
because this is like, this is a well trodden path,
link |
but I mean, not so boring.
link |
I suppose it's kind of, we just want to be precise.
link |
If the universe is deterministic,
link |
how can we have free will, right?
link |
So Sarah's a physicist.
link |
I think she, not believe, can show that most of the laws
link |
we have are deterministic to some degree.
link |
Quantum mechanics onto Newtonian stuff.
link |
And yet there's Sarah telling me she believes in free will.
link |
And I'm like, your belief system's broken here, right?
link |
Because you're demanding free will
link |
in a deterministic universe.
link |
And then I realized that I agreed with her
link |
that I do think that free will is a thing
link |
because we are able to search for novelty.
link |
And then that's where I came to the conclusion
link |
that time, the universe is expanding in terms of novelty.
link |
And it goes back to that Dan Dennett essay
link |
that we're talking about, the free will inflation.
link |
Free will, so you have, so the past,
link |
it did not exist in the past.
link |
The past exists in the present.
link |
What I mean is like, you are the, there was no past.
link |
There is only present.
link |
So that means you are the sum total.
link |
Everything that's occurs in the past
link |
is manifestly here in the present.
link |
And then you have this little echo state
link |
in your consciousness because you're able to,
link |
you're able to imagine something without actualization.
link |
But the fact you imagine it, that occurs in electrons
link |
and potassium ion flows in your neural network
link |
Maybe consciousness is just the present.
link |
So somehow you imagine that, and then by imagining,
link |
oh, that's good, yeah.
link |
I'm gonna make a robot to do this thing and program it.
link |
And then you physically then go and do it.
link |
So that then changes the future, sorry.
link |
What's imagination?
link |
Does it require the past?
link |
Does it require the future?
link |
Does it require memory?
link |
Does it only exist in the moment?
link |
So imagination is, well, yeah, probably
link |
it's an instantaneous readout of what's going on.
link |
You can maybe, your subconscious brain
link |
has been generating all the bits for it,
link |
but no, imagination occurs when you,
link |
in your game engine, you remember the past
link |
and you integrate sensory to the present
link |
and you try and work out what you want to do in the future.
link |
And then you go and make that happen.
link |
So the imagination is this, it's like, imagine,
link |
asking what imagination is about, asking what surfing is.
link |
You can see, you can surfboard, surfer, wave coming in.
link |
When you're on that wave and you're surfing,
link |
that's where the imagination is.
link |
I think imagination is just accessing things
link |
that aren't the present moment in the present moment.
link |
So like I'm sitting here and I'm looking at the table
link |
and I can imagine the river and things or whatever it was.
link |
And so it seems to be that it's like,
link |
it's our ability to access things that aren't present.
link |
So conjure up worlds, some of them might be akin
link |
to something that happened to you recently.
link |
Right, but they don't have to be things
link |
that actually happened in your past.
link |
And I think this gets back to assembly theory.
link |
Like the way I would think about imagination
link |
from an assembly theoretic standpoint
link |
is I'm a giant causal graph and I exist in a present moment
link |
as a particular configuration of Sarah.
link |
But there's a lot of, I carry a lot of evolutionary baggage.
link |
I have that whole causal history
link |
and I can access parts of it.
link |
Now, when you talk about getting to something
link |
as complex as us, having as large assembly space as us,
link |
there's ways of, like, there's a lot of things
link |
in that causal graph that have ever actually never existed
link |
in the past history of the universe
link |
because like the universe got big enough
link |
to contain the three of us in this room in time,
link |
but not all the features of each one of us individually
link |
have come into existence as physical objects
link |
we would recognize as individual objects.
link |
This goes back to your point
link |
that we actually have to explain
link |
why things actually even look like objects
link |
and aren't just a smear of mass.
link |
And just on the free will and physics thing,
link |
when you were talking, I was, I just wanna bring this up
link |
because I think it's a really interesting viewpoint
link |
that Nicholas Jisen has that, you know,
link |
like we wanna use the laws of physics
link |
and then say you can't have free will.
link |
And his point is you have to have free will
link |
in order to even choose to set up an experiment
link |
to test the laws of physics.
link |
So in some sense, free will should be more fundamental
link |
than physics is because to even do science,
link |
there's some assumption that the agents have free will.
link |
And I always thought it was really perplexing
link |
that, you know, physics wants to remove agency
link |
because the idea that I could do an experiment here
link |
on this part of earth and then I can move somewhere else
link |
and prepare an identically, you know,
link |
identically prepared experiment, run an experiment again,
link |
seems to imply something about the structure of our universe
link |
that is not encoded in the laws
link |
that we're testing in those experiments.
link |
So this kind of dream of physics
link |
that you can do multiple experiments, different locations
link |
and then validate each other,
link |
you're saying that's an illusion?
link |
No, I'm saying that requires decision making
link |
and free will to be a real thing, I think.
link |
Like I think the fact that we can do science
link |
And I think people, you know,
link |
the standard canon in physics would be,
link |
well, you could trace all of that back
link |
to the initial condition of the universe,
link |
but the whole point of science is
link |
I can imagine doing the experiment and I can do it
link |
and then I can do it again and again and again
link |
all over the planet.
link |
To you, imagination is somehow fundamentally
link |
generative of novelty.
link |
So it's not like the universe could have predicted
link |
the things you imagined.
link |
Imagination, so coming back to novelty,
link |
I think novelty can exist outside of imagination,
link |
but it supercharges it, it's another transition, I think.
link |
I mean, I would say,
link |
I mean, this may be a boring statement,
link |
but I would say the fact that, sorry?
link |
I'm not sure, these are hard questions.
link |
Yeah, I mean, I think the fact that objects exist
link |
is yet another proof that time is fundamental
link |
and novelty exists, right?
link |
Because I think, again, if you ask a physicist
link |
to write down in their infinite Bible of the universe,
link |
let's call it the Bible, the Mac, you know.
link |
Yeah, well, I mean, the mathematical universe,
link |
whether you're Max Tegmark or Sean Carroll
link |
Or Stephen Wolfram, okay?
link |
Yeah, I love it too.
link |
It's lots of pretty pictures.
link |
It's really interesting that they cope with the enormity
link |
of the universe by saying,
link |
well, it's all there, mathematics, it all exists, right?
link |
And I would say that that's why I'm excited
link |
about the future of the universe
link |
because although it is somehow dependent upon the past,
link |
it is not constrained just by the past,
link |
which is kind of mad.
link |
Yeah, that's what free will is.
link |
It's not constrained by the past.
link |
It's dependent on the past.
link |
This moment, it's not just dependent,
link |
this moment is the past,
link |
and yet it has the capacity
link |
to generate a totally unpredictable future.
link |
I mean, the other thing I would say
link |
that's super important for human beings, right?
link |
Human beings have actually very little causal control
link |
I realized this the other week.
link |
Oh, the immediate future immediately.
link |
Yeah, yeah, so what happened,
link |
so this is what I think it is.
link |
The way, by reinterpreting your past,
link |
I mean, talk about from a kind of cognitive,
link |
psychological cognitive point of view,
link |
by reinterpreting your past in your current mind,
link |
you could actually help you shape your future again.
link |
So you have much more freedom to interpret your past,
link |
to act in the present, to change your future,
link |
than you do to change your future.
link |
It may sound weird, so I'm saying to everybody,
link |
imagine your past, think about your past,
link |
reinterpret your past in the nicest way you can,
link |
then imagine what you can do next,
link |
or imagine your past in a more negative way
link |
and what you do next,
link |
and look at those two counterfactuals, they're different.
link |
Yeah, it's fascinating.
link |
I mean, Daniel Kahneman talks about this
link |
that most of our life is lived in our memories.
link |
It's interesting, because you can essentially,
link |
in imagination, choose the life you live.
link |
So maybe free will exists in imagination.
link |
Choices are made in your imagination,
link |
and that results in you basically able to control
link |
how the future unrolls, because you're like,
link |
imagining, like reinterpreting constantly
link |
the things that happen to you.
link |
Exactly, so if you want to increase
link |
your amount of free will, those people that have,
link |
I don't think everyone has equal amounts of agency,
link |
because of our sad constraints,
link |
where they're happenstance, health, economic,
link |
born in a certain place, right?
link |
But those of us that have the ability to go back
link |
and reinterpret our past and use that to change the future
link |
are the ones that exert most agency in the present,
link |
and I want to achieve higher degrees of agency
link |
and enable everyone else to do that as well,
link |
to have more fun in the universe.
link |
Then we'll hit that peak, the maximum fun point.
link |
I don't think there's ever gonna be a maximum,
link |
I think the wonderful thing about the future
link |
is there's always gonna be more fun.
link |
Yeah, you, I think, again, going back to Twitter,
link |
I think you, Lee, tweeted something
link |
about being a life maximalist,
link |
that you want to maximize the number of life,
link |
the amount of life in the universe, so,
link |
and that's the more general version of that goal,
link |
is to maximize the amount of fun in the universe,
link |
because life is a subset of fun, there are all kinds of,
link |
I suppose they're either correlated or exactly equal,
link |
Anyway, speaking of fun,
link |
let me ask you about alien sightings.
link |
So there's been quite a bit of UFO sightings
link |
and all that kind of stuff.
link |
What do you think would be the first time
link |
when humans sight aliens, see aliens,
link |
in a sort of unquestionable way?
link |
This extremely strong and arguable way
link |
we've made contact with aliens.
link |
Sarah, what would it look like?
link |
Obviously, the space of possibilities is huge here,
link |
but if you were to kind of look into the future,
link |
what would that look like?
link |
Would it be inklings of UFOs here and there
link |
that slowly unravel a mystery?
link |
Or would it be like an obvious, overwhelming signal?
link |
So I think we have an obsession
link |
with making contact with events.
link |
So what do I mean by that is,
link |
like people have a UFO sighting, they make contact.
link |
And I always think, you know,
link |
what's interesting to me about the UFO narratives right now
link |
is not that I have a disbelief
link |
about what people are experiencing or feeling,
link |
but like the discussion right now
link |
is sort of at the level of modern mythology.
link |
Aliens are our mythos in modern culture.
link |
And when you treat it like that,
link |
then I wanna think about when do things
link |
that we traditionally only regularize through mythology
link |
actually become things that become standard knowledge?
link |
So, you know, like it used to be, you know,
link |
variations in the climate were described
link |
by some kind of gods or something.
link |
And now it's like, you know,
link |
our technology picks up an anomaly
link |
or someone sees something, we say it's aliens.
link |
And I think the real thing is it's not contact with events,
link |
but like first contact is actually contact
link |
with knowledge of the phenomenon or the explanation.
link |
And so this is very subtle and very abstract,
link |
but when does it become something
link |
that we actually understand what it is
link |
that we're talking about?
link |
That's first contact.
link |
Would you make the myth,
link |
would you give credit to the myth,
link |
the mythology as first contact?
link |
I think it's the rudimentary
link |
that we have some understanding that there's a phenomenon
link |
that we have to understand and regularize.
link |
Right, to understand that there is weather.
link |
You have to construct a pathology around that weather.
link |
It's something that's controllable.
link |
I see mythology basically as like baby knowledge.
link |
It could be that, you know,
link |
although there's lots of alien sight,
link |
so called alien sightings, right?
link |
So there is a number of things you can do.
link |
You could just dismiss them and say they're not true.
link |
They're kind of made up.
link |
Or you say, well, there's something interesting here, right?
link |
We keep seeing a commonality, right?
link |
We see the same phenomenon again and again and again.
link |
Also, there's this interesting thing
link |
about human imagination.
link |
Even if they are, let's not say made up,
link |
but misappropriated kind of other inputs,
link |
the fact that human consciousness
link |
is capable of imagining a contact with aliens.
link |
Does that not tell us about something
link |
about where we are in our position,
link |
in our culture, in our technology?
link |
It tells us about where in time we are.
link |
Could it be that we're making contact with,
link |
let's say that, so let's say,
link |
let's take the most miserable version.
link |
There are no aliens in the universe.
link |
Life is only on Earth.
link |
That then, the interpretation of that
link |
is we're desperate to kind of understand
link |
why we're the only life in the universe, right?
link |
The other one is the other most extreme
link |
is that aliens are visiting all the time.
link |
We're just not able to capture them coherently,
link |
or there's a big conspiracy and there's Area 51
link |
and there are lizards everywhere and there's that.
link |
Or I'm kind of in favor of the idea
link |
that maybe humanity is waking up to the idea
link |
that we aren't alone in the universe
link |
and we're just running the simulation
link |
and we're seeing some evidence.
link |
We don't know what life is yet.
link |
We do have some anomalies out there.
link |
We can't explain everything.
link |
And over time, you know, we will start to unpack that.
link |
One very plausible thing we might do,
link |
which might be boring for the average alien observer
link |
or believes that aliens are,
link |
as in intelligent aliens are visiting Earth,
link |
it could be that we might go to the outer solar system
link |
and find a new type of life
link |
that has completely new chemistry,
link |
bring these cells back to Earth,
link |
where you could say in my hand,
link |
on Earth, here's RNA, DNA, and proteins,
link |
and look, cells self replicate.
link |
From Titan, we got this new set of molecules,
link |
new set of cells, and we feed it stuff and it grows.
link |
That for me, if we were able to do that,
link |
which would be like the most,
link |
that would be my UFO sign.
link |
That's a good test.
link |
So you feed it and it grows.
link |
We've made, so not until you know how to feed the thing,
link |
We can make a comic book, you know,
link |
the tiger that came for tea, the alien that came for tea.
link |
What would you say is between the two of you
link |
is the biggest disagreement about alien life out there?
link |
Is it from the basic framework of thinking about
link |
what is life to maybe what aliens look like
link |
to alien civilizations, to UFO sightings?
link |
What would you think?
link |
So I would say the biggest one is that
link |
the emergence of life does not have to be,
link |
that it can't just happen once on the planet,
link |
that it could be two or more life forms
link |
present on the planet at once.
link |
And I think Sarah doesn't agree with that.
link |
I think that's like logically inconsistent.
link |
That's really polite.
link |
You're saying it's nonsense.
link |
Because you think that, yeah.
link |
So likely is that, so the idea that,
link |
what does it look like?
link |
Let's imagine two alien civilizations
link |
coexisting on a planet.
link |
What's that look like exactly?
link |
So I would say, I think I've got to get around your argument.
link |
Yeah, let's say that on this planet,
link |
there's just like, there's lots of available chemistry
link |
and one life form gets some emerges based on carbon
link |
and interacts and there's an ecosystem based on carbon.
link |
And there's an orthogonal, and so it's planetary phenomena,
link |
which is what you, I think, right?
link |
But there's also one that carries on silicon.
link |
And because there's enough energy and there's enough stuff,
link |
that these life forms might not actually
link |
necessarily compete evolutionarily.
link |
Yeah, but they would have to not interact at all
link |
because they're going to be co constructing
link |
each other's causal chains.
link |
I think that's what you just got me, yeah.
link |
So there's no overlap in terms of their causal chains
link |
or a very limited overlap.
link |
Yeah, so I think the only way I can get away with that
link |
is to say, right, life can emerge on a planet underneath.
link |
The lizard people under the crust of the earth.
link |
I think, I think, I think we, let's go to,
link |
I think, but look, as you can see, we disagree.
link |
So, and I think Sarah actually has convinced me
link |
because of the life is a planetary,
link |
the emergence of life is a planetary phenomena.
link |
And actually, because of the way evolution selection works,
link |
then nothing occurs in isolation.
link |
The causal chains interact.
link |
So there is a common, there's a consensus model
link |
for life on the earth.
link |
But you don't think you can place aliens from elsewhere
link |
onto the, can't you just place multiple alien civilizations
link |
Right, but I think, so you can take two origin life events
link |
that were independent and co mingle them.
link |
But I don't think when you're talking about, when you,
link |
when you look at the interaction of that structure,
link |
it's like the same idea as like an experiment
link |
being an example of life, right?
link |
That's a really abstract and subtle concept.
link |
And I guess what I'm saying is life is information
link |
propagating through matter.
link |
So once you start having things interacting,
link |
they in some sense co mingle and they become part
link |
of the same chain.
link |
So the co mingling starts quickly, proceeds,
link |
we proceed to co mingle quickly.
link |
Right, right, so you could say, so the question is then,
link |
the more interesting question is,
link |
are there two distinct origins events?
link |
And I still think that there's reasons that
link |
on a single planet, you would have one origins event
link |
because of the timescales of cycling of geochemistry
link |
on a planet and also the fact that I don't think
link |
that the origin of life happens in a pool
link |
and like radiates outward through evolutionary processes.
link |
I think it's a multi scale phenomenon happens
link |
at the level of individual molecules interacting,
link |
collections of molecules interacting
link |
and entire planetary scale cycles.
link |
So life as we know it has always been multi scale
link |
and there's brilliant examples of individual mutations
link |
at the genome level changing global climate, right?
link |
So there's a tight coupling between things that happen
link |
at the largest scale, our planetary scale
link |
and the smallest scale that life mediates.
link |
But it still might be difficult within something
link |
you would call as a single alien civilization.
link |
You know, there's species and stuff.
link |
And they might not be able to communicate.
link |
But you're asking about life, not species, right?
link |
What's the difference between one living civilization?
link |
This is almost like a category question.
link |
Versus species, because it can be very different.
link |
Because there's like literally islands
link |
that you can evolve different kinds of turtles and stuff.
link |
So I guess what I'm saying is weird.
link |
If you look at the structure
link |
of two interacting living things, populations,
link |
and you look in their past
link |
and they have independent origins for their causal chain,
link |
then you would say one was alien.
link |
You know, they have different independent origins events.
link |
But if you look at their future
link |
by virtue of the fact they're interacting,
link |
their causal chains have become commingled.
link |
So then in the future, they are not independent.
link |
So that's why you would even define them as alien.
link |
So the structure across time is two examples of life
link |
become one example of life
link |
because life is the entire structure across time.
link |
Right, but there could be a lot of variation within.
link |
Yeah, so the question we're all interested in
link |
is how many independent origins
link |
of a complexifying causal chain are there in the universe?
link |
See, but the idea of origin is easy for you to define?
link |
Because like, when the species split
link |
in the evolutionary process,
link |
and you get like a dolphin versus a human
link |
or a Neanderthal versus Homo sapiens, isn't there?
link |
Let me make a distinction here quickly.
link |
So I think, sorry to interrupt.
link |
What we're saying, I mean, Sarah won that argument
link |
because I think she's right,
link |
that once the causal chains interact and going forward,
link |
so we're talking about a number of things.
link |
Let's go all the way back before origin of life.
link |
Chemistry emerges, so there's all these,
link |
I would say there's probably mechanistically,
link |
the chemistry is desperately trying to find any way
link |
to get replicators.
link |
The ribosome kind of was really rubbish at the beginning
link |
and they just competed, competed, competed,
link |
and you got better and better ribosome.
link |
Suddenly that was a technology.
link |
The ribosome is the technology that, boom,
link |
allowed evolution to start.
link |
So what I was trying to, why I interrupted you,
link |
is say that once evolution has started
link |
using that technology, then you can speciate.
link |
And I was trying to, and I think what Sarah said
link |
was, convinced me of, because I was like, no,
link |
we're gonna have lots of different chemistry,
link |
shadow biosphere on Earth, and she's like, no, no, no.
link |
You have to have this, you have to get to this
link |
minimum evolutionary machine.
link |
And then when that occurs, speciation occurs.
link |
Exactly what it's like, dolphins, humans,
link |
everything on Earth.
link |
But when you're looking at aliens or alien life,
link |
there's not gonna be two different types of chemistry
link |
because they compete and interact and cooperate
link |
because the causal chains overlap.
link |
One might kill the other, one might combine with the other,
link |
and then you go on and then you have this average.
link |
And sure, there might be respeciation.
link |
It might have two types of emerging chemistry.
link |
It almost looks like the origin of life on Earth
link |
required two different prelife forms,
link |
the peptide world and the RNA world.
link |
Somehow they got together, and by combining,
link |
you got the ribosome.
link |
And that was the minimum competent entity for evolution.
link |
And would all alien civilizations
link |
have an evolutionary process on a planet?
link |
So it's almost the definition of life.
link |
To create all those memories, you have to have something.
link |
Things have to change in time.
link |
But there has to be selection.
link |
That's like an efficient, there's no other way to do it.
link |
No, well, never say never, because soon as I say that.
link |
That's the part that depresses me, though,
link |
going back to the earlier discussion on violence
link |
and things, and I don't know where,
link |
somebody was tweeting about this recently,
link |
but how much death had to die.
link |
We're talking about life.
link |
And I guess a lot of murder had to occur.
link |
Right, so selection means things had to be weeded out,
link |
Well, we can celebrate that.
link |
Death makes way for a tulip.
link |
Yeah, I mean, and also one of the most interesting features
link |
of major extinction events in the history of our planet
link |
is how much novelty emerged immediately after, right?
link |
So, and of course, a lot of people make arguments
link |
we wouldn't be here if the dinosaurs didn't go extinct.
link |
So, in some ways, we can attribute our existence
link |
But I guess I was just wondering and sort of like,
link |
if I was gonna build a universe myself
link |
in the most optimistic way, would I retain that feature?
link |
But it does seem to be a universe.
link |
I think you have to.
link |
I mean, I think we're probably being
link |
over anthropomorphizing.
link |
I remember watching the blue,
link |
I think it was the blue planet,
link |
David Attenborough was showing these seals
link |
and because of climate change,
link |
some seals were falling off a cliff.
link |
And how tragic that was.
link |
I was like, I'm saying my son, that's pretty cool.
link |
Look at those ones down there.
link |
They've obviously got some kind of mutations,
link |
some and they're not doing that Darth thing.
link |
And so that poor gene will be weeded out.
link |
Of course, at the individual level, it looks tragic.
link |
And of course, as human beings have the ability
link |
to abstract and we empathize,
link |
we don't wanna cause suffering on other human beings
link |
and we should retain that.
link |
But we shouldn't look back in time
link |
and say, how many butterflies had to die?
link |
I remember making this,
link |
how many, if you think about the caterpillar
link |
become the chrysalis and then the butterfly getting out,
link |
how many, if that suffering, we call it suffering,
link |
if that process of pruning had not occurred,
link |
we have no butterflies.
link |
So none of the butterfly beauty in the world
link |
without all that pruning.
link |
So pruning is required,
link |
but we shouldn't anthropomorphize
link |
and feel sorry for the biological entities,
link |
because that seems to be a backwards way of looking at it.
link |
What we should do is project forward
link |
and maybe think about what values we have across our species
link |
and our ecosystem and our fellow human beings.
link |
You know, now that we know that animals suffer
link |
at some level, think about humane farming.
link |
When we find that plants can, in fact,
link |
are conscious and can think and have pain,
link |
then we'll do humane gardening.
link |
Until that point, we won't do it, right?
link |
Famous chemist endorses the majestic nature of murder.
link |
I didn't say that, but any case.
link |
Well, I just inserted it.
link |
I have a hard time with it, though.
link |
I think the way you put it is kind of...
link |
But it's the reality of, it is beautiful.
link |
You know, there's an Instagram account
link |
called natureismetal,
link |
and I keep following it and unfollowing it
link |
because I can't handle it for prolonged periods of time.
link |
We evolve together, you die alone.
link |
We evolve together, but you die alone.
link |
We live alone, too.
link |
It's the Gatsby thing.
link |
We evolve together.
link |
Where's the together?
link |
The together is the murder and the sex, sex and murder.
link |
My romantic vision of it is to try to make me happy, Sarah,
link |
instead of sad Sarah.
link |
I talk in third person when I think very abstractly.
link |
Sorry, is, you know, like this whole,
link |
like certain things can coexist,
link |
so the universe is trying to maximize existence,
link |
but there's some things
link |
that just aren't the most productive trajectory together,
link |
but it doesn't mean that they don't exist
link |
on another timeline or another chain somewhere else.
link |
Like, and maybe you would call that, like,
link |
then some kind of multiverse or things,
link |
but what am I saying?
link |
I think you can't.
link |
I just, you can't go down a level.
link |
I'm just making stuff up.
link |
It makes me feel better.
link |
I don't understand.
link |
And we need, we need.
link |
No, I know, I know.
link |
Yeah, if you look at bacteria, if you look at virus,
link |
I mean, just the number of organisms that are constantly,
link |
like looking at bacteria, they're just dying nonstop.
link |
It's like a slaughter.
link |
Well, and this goes back to the conversation about God.
link |
I mean, like, there's the whole thing about, like,
link |
why is the universe unable to suffering?
link |
Individuals don't exist, right?
link |
Individuals, so for this, I think,
link |
if you think about life as an entity on Earth, right?
link |
Let's just, let's just go back a second.
link |
I mean, I like to, I'll be ludicrous for a second.
link |
You don't exist, right?
link |
But you, but the actions you do,
link |
the product of evolution exists, right?
link |
The objects you create exist quantitatively
link |
in the real world.
link |
If you then understand life on Earth or alien life
link |
or any life in the universe as this integrated entity
link |
where you need, you need cells in your body to die.
link |
Otherwise you'd just get really big
link |
and you wouldn't be able to walk around, right?
link |
So, you know, you do.
link |
Yeah, yeah, yeah, yeah.
link |
It's the patterns that persist, not the physical things.
link |
And of course, we, you know, we have, we have,
link |
we place immense values on fellow human beings
link |
and I'm, majestic professor does like
link |
other individual human beings.
link |
Now you're talking in third person too.
link |
I know, it happens, right?
link |
So death, would you say, I mean,
link |
because you said evolution is a fundamental part of life.
link |
So death is a fundamental part of life.
link |
It might, right now, it might not be in the future.
link |
We might hack some aspects of death
link |
because, and we'll evolve in different ways.
link |
But isn't there, I think Sarah mentioned
link |
like this life density.
link |
Is it, can't that become a problem?
link |
Like too much, too much bureaucracy,
link |
too much baggage builds up.
link |
Like you need to keep erasing stuff.
link |
I think it's okay that we dissipate.
link |
Like, I don't think of it like, like, I mean.
link |
No, but I mean, like, like we're so fixated
link |
on ourselves as individuals and agents.
link |
And we were talking about this last night,
link |
actually over dinner, but like,
link |
you know, an individual persists
link |
for a certain amount of time,
link |
but what you want to do,
link |
like if you're really concerned with immortality
link |
is not to live indefinitely as an individual,
link |
but maximize your causal impact.
link |
So like, what are the traces of you that are left?
link |
And you're still a real, I always think of Einstein,
link |
like for a period of time, he was a real physical thing
link |
we would identify as a human.
link |
And now we just see echoes of that human
link |
in all of the ways that we talk about his,
link |
you know, causal impact are frankly great
link |
is another great example.
link |
How many Easter eggs could you leave in the future?
link |
It's like, oh, I got you.
link |
So I guess the question is,
link |
how much do you want to control the localization
link |
of certain features of say a prop,
link |
a packet of propagating information we might call person
link |
and keep them localized
link |
to one individual physical structure?
link |
Or do you want to, you know,
link |
is there a time when that just becomes a dissipated feature
link |
of the society that it once existed in?
link |
And I'm okay with the dissipated feature
link |
because I just think that makes more room
link |
for more creativity in the future.
link |
So you mentioned engineering life in the lab.
link |
Let me take you to computer science world.
link |
What about robots?
link |
So is it possible to engineer,
link |
so you're really talking about like engineering life
link |
at the chemistry level,
link |
but do you think it's possible to engineer a life
link |
at the like humanoid level, at the dog level?
link |
Like, or is that, like at which level can we instill
link |
the magic of life into inanimate stuff?
link |
No, I think you could do it at every level.
link |
I just think that we're particularly interested
link |
in chemistry because it's the origin of life transition
link |
that presumably, or at least that's how I feel about it,
link |
it's going to give you the most interesting
link |
or deepest insights into the physics.
link |
But presumably everything that we do and build
link |
is an example of life.
link |
And the question is just how much do you want to take
link |
from things that we have now and put them into,
link |
like examples of life and copy them into machines?
link |
I saw that there was this tweet again.
link |
I think you were at the Mars conference
link |
and you were hanging out with a humanoid robot.
link |
Yes, that was a fun time.
link |
Making lots of new friends at Mars 2020.
link |
Did you guys color match ahead of time with the robot
link |
or did that accidentally happen?
link |
Accidentally, I went up and I wanted to say hi.
link |
Torpoise, would that be the correct name for the color?
link |
We didn't color coordinate our outfits.
link |
Well, you didn't, maybe the robot did.
link |
The robot probably did, much more stylish.
link |
So for people who are just listening,
link |
there's a picture of Sarah standing next
link |
to a humanoid robot.
link |
I guess you like them with a small head and perfect vision.
link |
Actually, no, I just.
link |
What did they perfectly, there's a LIDAR.
link |
No, I mean, I think I was just deeply interested
link |
What was, sorry to interrupt, was it manual control?
link |
Was it actually stabilizing itself?
link |
Oh no, it was walking around.
link |
It was pretty impressive.
link |
I mean, actually there's some videos online
link |
of Jeff Bezos walking with one of those
link |
across the lawn nearby there.
link |
Yeah, but there you go.
link |
That's incredible, isn't it?
link |
See, you look at the walking robot.
link |
Where did the idea for walking come from?
link |
It was invented by evolution, right?
link |
And us as human beings, able to conceptualize and design
link |
and engineer the causal change.
link |
So that robot is evidence of life.
link |
And so I think what's going to happen is there's the,
link |
we want to find where the spark comes from mechanistically.
link |
How can you literally go from sand to cells?
link |
So that's the first transition that I think,
link |
there are a number of problems we want to do.
link |
Make life in the lab.
link |
Then we want to make life in the lab
link |
and want to suddenly start to make intelligent life
link |
or life that can solve, start to solve abstract problems.
link |
And then we want to make life that is conscious.
link |
I think it has to happen in that order.
link |
Getting towards this artificial general intelligence.
link |
I think that artificial general intelligence
link |
can't exist in a vacuum.
link |
It has to have a causal change
link |
all the way back to Luca, right?
link |
And so the question I think,
link |
I really like the question is to say,
link |
what are we, how is,
link |
how is our pursuit of more and more lifelike?
link |
I know you want to, you like your robots.
link |
You want to project into them.
link |
You want to interact with them.
link |
I think you would want,
link |
if you have a robot dog and the robot dog
link |
does everything expected of a normal dog
link |
and you can't tell the difference,
link |
you're not really going to ask the question anymore.
link |
If it's a real dog or not,
link |
or you've got a personality, you're interacting with it.
link |
And so I think what would be interesting
link |
would be to kind of understand
link |
the computational architecture, how that evolves.
link |
Cause you could then, you know,
link |
teleport the personality from one object to the other
link |
and say, right, does it act the same?
link |
And I think that as we go along,
link |
we're going to get better and better
link |
at integrating our consciousness into machines.
link |
Well, let me ask you that question just to link on it.
link |
I would, I would call that a living conscious thing
link |
potentially, I as a human allegedly,
link |
but would you, as a person trying to define life,
link |
if you pass the Turing test, are you a life form?
link |
One of the reasons I walked up to the robot
link |
was because I wanted to meet the robot, right?
link |
So I, it felt like I was,
link |
and I abase a lot of my interaction with reality
link |
on emotion and feeling,
link |
but like how do you feel about an interaction?
link |
And I always love your point about like,
link |
is it enough to have that shared experience
link |
with a robot, right?
link |
So walking up to it,
link |
does it feel like you're interacting with a living thing?
link |
And it did to an extent, but in some degrees,
link |
it feels like you're interacting with a baby living thing.
link |
So I think our relationship with technology
link |
and particularly robots we build is really interesting
link |
because basically they exist as objects in our future
link |
in some sense, like we're a much older evolutionary lineage
link |
than robots are, but we're all part of the same causal chain
link |
and presumably, you know, they're kind of in their infancy.
link |
So it's almost like you're looking at the future of life
link |
when you're looking at them,
link |
but it hasn't really become life in a full manifestation
link |
of whatever it is that they're gonna become.
link |
And, you know, the example of the walking robot
link |
was super interesting, but they also had a dolphin
link |
that they put in the pool at the cocktail party at Mars,
link |
and it looked just like a real dolphin swimming in the pool.
link |
And, you know, it's in this kind of uncanny valley
link |
because, and I was having this conversation
link |
with a gentleman named Mutu who was super perceptive,
link |
but he was basically saying like,
link |
it made him feel really uncomfortable.
link |
Yeah, and I think a lot of people would have that response.
link |
And I guess my point about it is it is kind of interesting
link |
because you're basically trying to make a thing
link |
that you think is nonliving mimic a living thing.
link |
And so the thought experiment I would wanna run in that case
link |
is imagine we replaced every living thing on earth
link |
with a robot equivalent, like all the dolphins and things.
link |
And in some sense, then you're making,
link |
if you think that the robots aren't experiencing reality,
link |
for example, in the way that
link |
a biologically evolved thing would,
link |
you're basically making the philosophical zombie argument
link |
become real and basically building reality into a simulation
link |
because you've made everything quote unquote fake
link |
You've replaced everything with a physical simulation of it.
link |
So as opposed to being excited by the possibility
link |
of creating something new,
link |
you're terrified of humans being replaced.
link |
I was just trying to run like what would be
link |
the absolute thought experiment.
link |
But I don't think that scenario would actually play out.
link |
I guess what I think is weird for why we feel
link |
this kind of uncanny valley interacting
link |
with something like the robot dolphin
link |
is we're looking at an object we know is kind of
link |
in the future in the sense of like
link |
if everything's ordered in time,
link |
but it's borrowing from a structure
link |
that we have common history with,
link |
and it's basically copying in a kind of superficial way
link |
things from one part of the causal chain to another.
link |
Well, that's a video of...
link |
Everybody believed it was real.
link |
They look so real.
link |
And obviously the technology was developed for movies, so.
link |
Well, I think we're confusing our emotional response
link |
and understanding the causal chain
link |
of how we got there, right?
link |
Because the philosophical dot zombie argument
link |
thinks about objects just appearing, right,
link |
that you're facsimile in some way,
link |
whereas there is the causal, the chain of events
link |
that caused the dolphin to be built went for a human being.
link |
Yeah, would a philosophical zombie
link |
still have a high assembly index?
link |
Because it can't be, philosophical zombies can't,
link |
like Boltzmann brains, just can't appear out of nowhere.
link |
Well, I guess my question would be in that scenario
link |
where you built all the robots
link |
and replaced everything on Earth with robots,
link |
would the biosphere be as creative
link |
under that scenario or not?
link |
Yeah, that's a good question.
link |
Are there quantitative differences
link |
you would notice over time?
link |
And it's not obvious either way, right?
link |
It's not obvious right now,
link |
because we don't really, we don't understand,
link |
we haven't built into machines how we work.
link |
So that's, I think, one of the big missing things
link |
that we're both looking for, right?
link |
This is a robot, it's a cute robot.
link |
But the point, Sarah, is that the biosphere
link |
won't be as creative if you did it right now.
link |
No, of course, I think that's why people don't like it.
link |
But in the future, we will be able to solve the problem
link |
of origin of life, intelligence, and consciousness
link |
because they exist in physical substrates.
link |
We just don't understand enough
link |
about the material substrate and the causal chain.
link |
But I'm very confident we will get to an AGI,
link |
but it won't be what people think.
link |
It won't be, solution won't be a, we'll get fooled a lot.
link |
And so GPT3 is getting better at fooling us
link |
and GPT153 might really fool us,
link |
but it won't have the magic we're looking for.
link |
It won't be a creative,
link |
but it will help us understand the differences between.
link |
Really though, because isn't that what love is, being fooled?
link |
Like what, why are you not giving much value
link |
to the emotional connection with objects,
link |
with robots, with humans?
link |
Emotion is a thing which happens
link |
when your expectation function is dashed
link |
and something else happens, right?
link |
I mean, that's what emotion is.
link |
Is that what love is too?
link |
You were expecting one thing and something else happened.
link |
I don't think that's true either.
link |
Well, what is it then?
link |
I think, no, emotion, look, I'm sorry, emotion is that,
link |
but there's no more.
link |
No, I think love is just fulfilling your purpose.
link |
No, but, okay, I mean, look, look.
link |
Like whatever that means.
link |
That's the opposite of what's gonna happen.
link |
I think the happiest is like when you're doing.
link |
All right, all right, all right.
link |
If you want me to define.
link |
Follow your bliss.
link |
Let me define love quickly.
link |
In terms of assembly space, right?
link |
I didn't think I'd be doing this today.
link |
I can't wait till Assembly Theory 101 is taught
link |
and the second lecture is Assembly Theory of Love.
link |
No, no, but look, actually, but look, but.
link |
It's being surprised.
link |
The expectation is being broken.
link |
I'm just, I'm not.
link |
I'm not an emotional being, but I would say,
link |
so let's talk, so we'll talk about emotion a bit,
link |
but love is more complex.
link |
Love is a very complex set of emotions together
link |
and logical stuff, but if you've got this thing,
link |
this person that's on this causal chain
link |
that has this empathy for this other thing,
link |
love is being able to project ahead in your assembly space
link |
and work out what you're,
link |
the person you're in love with has a need for
link |
and to do that for them without selflessly, right?
link |
Cause you can project ahead what they're gonna need
link |
and they are there and maybe you can see
link |
someone who's gonna fall over
link |
and you catch them before they fall over
link |
or maybe you can anticipate that someone's gonna be hungry
link |
and without helping you, you just help them.
link |
That's what love is.
link |
That just sounds like empathy.
link |
But it's more complex than that, right?
link |
It's more complex.
link |
It's more about not just empathy, it's understanding.
link |
It's about kind of sharing that experience.
link |
That's an expression of love though.
link |
That's not what it's like to feel love.
link |
Like feeling love is like,
link |
I think it's like when you're aligned with things
link |
that you feel like are your purpose
link |
or your reason for existing.
link |
So if you have those feelings towards a robot,
link |
why is that robot?
link |
I mean, cause you said like the AGI,
link |
we'll build an AGI,
link |
but there'll be a fundamental difference in AGI.
link |
I don't think we'll build it.
link |
It's gonna emerge from our technology.
link |
I think you guys all argue the same thing.
link |
I just said that GPT,
link |
we do not correctly capture the causal chain that we have.
link |
But don't you think it captures,
link |
because GPT3 is fundamentally trained
link |
on a corpus of knowledge,
link |
like the internet.
link |
Don't you think it gets better and better and better
link |
at capturing the memory of all the...
link |
It will be better at fooling you.
link |
And at some point you won't care.
link |
But when it comes to,
link |
my guess, this is a quick,
link |
this is what I was getting to right before we got,
link |
I got in the love trap.
link |
It was like Lee Cronin in the love trap.
link |
Sounds like a good fan name.
link |
SAD, okay, SAD, assembly space of SAD.
link |
No, is that, so sure.
link |
But I think there are other features
link |
that allow that we pull on innovation
link |
that allow us to do more than what we just see in GPT3.
link |
So if you're being fooled there.
link |
So I think what I mean is human beings have this ability
link |
to be surprising and creative.
link |
Whereas is it Dali, this thing,
link |
or if you take GPT3 is not gonna create a new verb.
link |
Shakespeare created new verbs.
link |
And that required Shakespeare to think outside of language
link |
in a different domain.
link |
So I think having that connections across multiple domains
link |
is what you need for AGI.
link |
Yeah, but I don't know if you need,
link |
I don't know if there's any limitations to GPT
link |
and not being able to be across domain.
link |
The number one problem is,
link |
it's instantiated in a resource limited substrate
link |
and that we don't, in silicon.
link |
The architectures used for training, for learning
link |
is about fooling, it's not about understanding.
link |
And I think that there is some understanding that we have
link |
that is not yet symbolically representable.
link |
Language, learning language and using language
link |
seems to be fundamentally about fooling, not understanding.
link |
Why do you use language exactly?
link |
I might disagree with that quite fundamentally actually,
link |
but I'm not sure I understand
link |
how I make a coherent argument for that.
link |
But my feeling is that there is comprehension in reality
link |
in our consciousness below language.
link |
And we use those for language,
link |
for all sorts of expressions.
link |
And we don't yet understand that there's a gap.
link |
We will get there, but I'm saying,
link |
wouldn't it be interesting, it's a bit like saying,
link |
could I facsimile you or Sarah into a new human being?
link |
And let's just say, I could copy all your atoms
link |
and the positions of all your atoms and electrons
link |
into this other person, they would be you.
link |
And it's quite easy to show using assembly theory
link |
because actually the feature space that you have,
link |
that graph, the only way to copy you
link |
is to create you on that graph.
link |
So everything that's happened to you in your past,
link |
we have to have a faithful record for.
link |
If you want another copy of Lex,
link |
you have to do the exact thing.
link |
Want another copy of Sarah, want another copy of Lee.
link |
The exact past has to be replicated.
link |
Let me push back on that a little bit.
link |
That's maybe from an assembly theory perspective,
link |
but I don't think it's that difficult
link |
to recreate a version of me, like a clone,
link |
that would make everybody exactly equally as happy.
link |
They wouldn't care which one.
link |
And there's two of me and then they get to pick which one
link |
and they'll kill either one, they'll be fine.
link |
As long as they're forced to kill.
link |
But here's what will happen is,
link |
let's say we make artificial Lex
link |
and it was like, wow, so cool.
link |
It looks the same interact.
link |
Then there'll be this battle of like,
link |
right, we're going to tell the difference.
link |
We're going to basically keep nudging Lex
link |
and artificial Lex until we get novelty from one
link |
and we'll kill the other one.
link |
And I think, thank God.
link |
We're not, novelty is a fuzzy concept.
link |
That's the whole problem of novelty.
link |
So I will define novelty, it's not fuzzy.
link |
Novelty is the ability for you to create architectures
link |
that are, or create an architecture.
link |
So let's say you've got a corpus of architectures known.
link |
You can write down, you've got some distance measure.
link |
And then I create a new one and the distance measure
link |
so far away from what you'd expected.
link |
There's no linear algebra going to get there.
link |
It's like, that is creativity.
link |
And we don't know how to do that yet on any level.
link |
Well, I was also thinking about like your argument
link |
about free will, like you wouldn't be able to know
link |
it was, it doesn't work instantaneously.
link |
It's not like a micro level thing,
link |
but more a macro level thing over the scale of trajectories
link |
or longer term decisions.
link |
So if you think that the novelty manifests
link |
over those longer timescales,
link |
it might be the two Lexes diverge quite a bit
link |
over certain timescales of their behavior.
link |
But nobody would notice the difference.
link |
And the universe, the earth won't notice the difference.
link |
The universe won't notice the difference.
link |
The universe would notice the difference.
link |
No, the universe doesn't know about its novelty
link |
that's being generated.
link |
That's the whole point of novelty.
link |
Yeah, but this is what selection is, right?
link |
It's like taking nearly equivalent ones
link |
and then deciding like the universe selects, right?
link |
So whatever selection is,
link |
select some things to persist in time.
link |
Yeah, it's gonna select the artificial one
link |
just because it likes that one better.
link |
Well, you're mixing up two arguments here.
link |
So look, let's go back a second.
link |
What are you basing this argument on, Lex?
link |
I'm just saying that I kind of don't think,
link |
cause at least said that it's not possible.
link |
Like if you copy every single molecule in a person's body,
link |
that's not going to be the same person.
link |
That they won't have the same assembly index.
link |
It won't be the same person.
link |
And I just don't, I think copying, you can compress.
link |
Not only do I disagree with that,
link |
I just, I think you can even compress a person down
link |
to some where you can fool the universe.
link |
I'm saying, let me restate it.
link |
It is not possible to copy somebody on,
link |
because you, unless you copy the causal history.
link |
Also, you can't have two identical.
link |
I mean, actually I really like the idea
link |
that everything in the universe is unique.
link |
So even if like there were two Lex's.
link |
I know you like that idea
link |
cause you're human and you think you're unique.
link |
But also I can make a logical argument for it
link |
that even if we could copy all of your molecules
link |
and all their positions, the other you would be there.
link |
And you have a different position in space.
link |
And the other thing.
link |
You're distinguishable.
link |
Yeah, the other thing I was gonna add.
link |
How unique are you?
link |
Just by the position in space really.
link |
Sure, but then how much does that light translation
link |
of Lex sitting there affect the future?
link |
I see, but wait a minute.
link |
Is part of the definition of something being interesting
link |
is how much it affects the future?
link |
But let me come back.
link |
But let me come back one point quickly
link |
that you were making.
link |
Sure, I think I probably agree, yes.
link |
There's two Lex's, right?
link |
There's a robot Lex that you just basically,
link |
It's just coded to emulate you.
link |
Are you robot Lex?
link |
I wouldn't know, right?
link |
But let's get that.
link |
It's a very important point here
link |
because he's ducking and diving between this eye.
link |
So if I facsimile you into a robot,
link |
then your robot might be,
link |
would be a representation of you now,
link |
but fundamentally be boring
link |
because you go and have other ideas.
link |
If, however, you built an architecture
link |
that itself is capable of generating novelty,
link |
you would diverge in your causal chain
link |
and you'd both be equally interesting to interact with.
link |
We don't know that mechanism.
link |
All I'm trying to say is we don't yet know that mechanism.
link |
We do not know the mechanism that generates novelty.
link |
And at the moment in our AIs, we are emulating,
link |
we are not generating.
link |
You don't think we're sneaking up on that?
link |
No, no, there is no ghost in the machine.
link |
And I want there to be one.
link |
I want the same thing you want.
link |
I know you want that as a human
link |
because everything you just said
link |
makes you feel more special.
link |
I want to be, no, no, no, screw my specialness.
link |
I just want to be surprised.
link |
You don't think a robot can surprise you.
link |
If I, if you can produce an algorithm
link |
instantiated in a robot to surprise me,
link |
I will, I will, I will, I will,
link |
I will have one of those robots, it'll be brilliant.
link |
But they won't, it won't surprise me.
link |
But why, why is it a problem
link |
to think that humans are special?
link |
Maybe it's not the special, you're right.
link |
It's the better than.
link |
Because then you start to not recognize
link |
the magic in other life forms
link |
that you either have created or you have observed.
link |
Because I just think there is magic
link |
in legged robots moving about.
link |
And they are full of surprises.
link |
And functionality.
link |
I know why you like cellular automata, right?
link |
But the specialness in your robot
link |
comes from the roboticist that built it.
link |
It's part of the lineage.
link |
And so that's fine.
link |
I'm happy with that.
link |
That's what I felt like looking at the standing robot
link |
was I was looking at four billion years of evolution.
link |
So I think I'm happy.
link |
I mean, I'm happy we're gonna coexist.
link |
I'm just saying you're gonna get more excitement.
link |
There's something missing
link |
in our understanding of intelligence.
link |
Intelligence isn't just training.
link |
The way the neural network is conceived right now is great
link |
and it's lovely and it'll be better
link |
and we will argue forever.
link |
But you want to know, wouldn't it be great if I said,
link |
look, I know how to invent an architecture
link |
and I can give it a soul.
link |
And what I mean by a soul is some,
link |
I know for real that there is internal reference.
link |
Soon as I not fake internal reference.
link |
And if we could generate that mechanism
link |
for internal reference, that's why our goal direct.
link |
That's why you have to develop a test for goal directness.
link |
Get that goal directness.
link |
You would love that robot more than the one
link |
that's just made to look like it does
link |
because you'll have more fun with it
link |
because you better generate search, other problems,
link |
get to more novelty.
link |
Hell, you'd be able to fall in love with that robot for real,
link |
but not the one that's faking it.
link |
What about fake it till you make it?
link |
Well, I think a lot of people fall in love with fake humans.
link |
It's nice to fall in love with something
link |
that's full of novelty, yes.
link |
I could imagine all kinds of robots
link |
that I would want to have a close relationship with.
link |
And I don't mean like sexual, I mean like intimacy.
link |
But I just don't think that novelty generation
link |
is such a special.
link |
Okay, there's like mathematical novelty
link |
or something like that,
link |
and then there's just humans being surprised.
link |
I think we're easily surprised.
link |
That's fine, but that's that.
link |
But you don't think that's a good definition of novelty?
link |
I'm happy to be surprised,
link |
but not globally surprised because someone else,
link |
but I really want, I was wondering why I'm a scientist.
link |
I really want to be the first to be surprised about something
link |
and the first thing in the universe
link |
to create that novelty
link |
and to know for sure that that novelty
link |
has never occurred anywhere else.
link |
That's a real buzz, right?
link |
Is there a way to really know that?
link |
You have to have a really big lookup table.
link |
Yeah, you're never going to be know for sure, right?
link |
That's one of the hard things about being
link |
a scientist searching for this type of novelty.
link |
Maybe that's why mathematics, mathematicians love discovery,
link |
but actually they are creating.
link |
And then when they create a new mathematical structure
link |
that they can then, you can write code to work out
link |
whether that structure exists before that.
link |
That's almost why I would love to have been a mathematician
link |
from that regard to invent new math
link |
that really I know pretty much for sure
link |
does not exist anywhere else in the universe
link |
because it's so contingent.
link |
Right, but this gets into like you said a few times
link |
that I still really don't understand
link |
how you actually plan to do this,
link |
to build an experiment that detects
link |
how the universe is generating novelty
link |
or that time is the mechanism.
link |
So the problem that we all have,
link |
which I think is what Lex is pushing against
link |
is if I build the experiment,
link |
you don't know what you put into it.
link |
So you don't know what, like if you,
link |
unless you can quantify everything you put in,
link |
all of your agency, all the boundary conditions,
link |
you don't know if you somehow biased it in some way.
link |
So is the novelty actually intrinsic to that experiment
link |
or to that robot, or is it something you gave it,
link |
but you didn't realize you gave it?
link |
It's gonna be, it's gonna asymptote towards that, right?
link |
You're never gonna know for sure,
link |
but you can start to take out,
link |
you can use good Bayesian approaches
link |
and just keep updating and updating and updating
link |
until you point to one sense of purpose.
link |
So you wanna bound on how much novelty generation
link |
it could be, got it.
link |
So the ability to generate novelty
link |
is correlated with high assembly index,
link |
with assembly index?
link |
Cause the space of possibilities is bigger.
link |
So that's the key.
link |
This could be a good, so I have a running joke
link |
of like why Lex is single.
link |
This could be a good part four.
link |
So what you're looking for in a robot partner
link |
is ability to generate novelty.
link |
And that's, I suppose you would say
link |
it's a good definition of intelligence too.
link |
Boy, is novelty a fuzzy concept.
link |
Is creativity better?
link |
Yeah, I mean, that's all pretty fuzzy.
link |
It's kind of the same.
link |
Maybe that's why aliens haven't come yet
link |
is cause we're not creating enough novelty.
link |
Like there's some kind of a hierarchy
link |
of novelty in the universe.
link |
Well, I think novelty is like things surprise you, right?
link |
So it's a very passive thing,
link |
but I guess what I meant by saying creativity
link |
is I think it's much more active.
link |
Like you think there's like a mechanism
link |
of like the things that exist are generating the creativity.
link |
Novelty seems to be there's some spontaneousness
link |
and it's completely decoupled from the things that exist.
link |
I think creativity is the mechanism
link |
and novelty is the observable.
link |
Novelty could just be surprising.
link |
Your model of the world was broken
link |
and not necessarily in a positive way.
link |
So there's three things now.
link |
All right, let's go.
link |
You got surprise, which is basically,
link |
I mean, I'm surprised all the time
link |
cause I don't read very much.
link |
I was like, oh wow, this is,
link |
I often used to invent new scientific, you know, ideas.
link |
And I was really surprised by that.
link |
And then I went and looked in literature properly
link |
So surprise, that's to the extent
link |
that you don't have full information.
link |
Creativity, the act of pushing on that kind of
link |
on the causal structure and novelty,
link |
which is measuring that degree, right?
link |
So, and I think that's pretty well defined in that regard.
link |
So you want your robot, I mean,
link |
and in the end, this is what I'm saying,
link |
and in the end, that's why actually the way the internet
link |
and the printing press share some,
link |
I actually think creativity has dropped a bit
link |
since the internet because everyone's just, you know,
link |
just regurgitating stuff.
link |
But of course, now it's beginning to accelerate again
link |
cause everyone's using this tool to be creative
link |
and boom, it's exploding.
link |
I think that's what happens
link |
when you create these new technologies.
link |
That's really helpful.
link |
There's a difference between novelty and surprise.
link |
Okay, I think I was thinking about surprise.
link |
If you give me a toy that surprises me for a bit,
link |
A robot that surprises me, you know.
link |
An experiment that surprises you.
link |
Yeah, I mean, that's why I love doing experiments
link |
cause I'm, I can't.
link |
It's still exciting.
link |
Yeah. Surprise is exciting.
link |
Even negative surprise,
link |
like some people love drama in relationships.
link |
Like, it's like, why the hell, why'd you do this?
link |
That could be exciting too.
link |
I could imagine companies selling updates
link |
to their companion robots
link |
that just basically generate negative surprise
link |
just to spice things up a bit.
link |
Yeah, it's the push and pull.
link |
That's one of the components of love.
link |
As you said, love is a complicated thing.
link |
I wanted to mention this cause you also tweeted,
link |
I think this was Sarah.
link |
No, it might've been Lee.
link |
But it was a survey published in Nature
link |
showing that scientists find.
link |
That was me, yeah.
link |
Anyway, there's a plot.
link |
This is published in Nature
link |
of what scientists find beautiful in their work
link |
and it separates biologists and physicists.
link |
It'd be nice if you showed the full plot.
link |
And there's simplicity, elegance, hidden order,
link |
inner logic of systems, symmetry, complexity, harmony,
link |
Is there any interesting things that stand out to you?
link |
I think the fact that biologists like complexity
link |
and pleasing colors.
link |
Oh, there's pleasing colors on there?
link |
Pleasing colors or shapes.
link |
And then physicists obviously love simplicity
link |
Simplicity and elegance.
link |
Simplicity, elegance.
link |
They love symmetry.
link |
And then biologists love complexity.
link |
And, well, they just love a little bit less.
link |
They love everything a little bit less,
link |
but complexity a little bit more.
link |
A little bit more.
link |
That's so interesting.
link |
And pleasing colors or shapes.
link |
Do you think it's a useful,
link |
I forget what your tweet was
link |
that this is missing some of the something.
link |
Oh, no, I think it's because I think about
link |
how explanations become causal to our future.
link |
So I have this whole philosophy
link |
that the theories we build
link |
and the way we describe reality
link |
should have the largest breadth of possibilities
link |
for the future of what we can accomplish.
link |
So in some sense, it's not like Occam's razor.
link |
It's not for simplicity.
link |
It's for optimism or the kind of future you can build.
link |
And so I think you have to think this way
link |
when you're thinking about life and alien life,
link |
because ultimately we're trying to build,
link |
I mean, science is just basically
link |
our narratives about reality.
link |
And now you're building a narrative
link |
that is what we are as physical systems.
link |
It seems to me it needs to be as positive as possible
link |
because it's basically gonna shape
link |
the future trajectory where we're going.
link |
And we don't use that as a heuristic in theory building
link |
because we think theories
link |
are about predicting features of the world,
link |
But if you look at the history of all of the development
link |
of human thought, it's caused the things that happen next.
link |
So it's not just about looking at the world
link |
It's about actually that feedback loop that's missing.
link |
And it's not in any of those categories.
link |
What do you think is the most beautiful idea
link |
in the physics of life, in the chemistry of life,
link |
in this, through all your exploration
link |
with assembly theory, what is the thing
link |
that made you step back and say this idea is beautiful
link |
or potentially beautiful?
link |
For me, it's that the universe is a creative place.
link |
I guess I want to think, and whether it's true or not,
link |
is that we are special in some way
link |
and it's not like an arbitrary added on epiphenomenon
link |
or ad hoc feature of the universe that we exist,
link |
but it's something deep and intrinsic
link |
to the structure of reality.
link |
And to me, the most beautiful ideas that come out of that
link |
is that the reason we exist is for the universe
link |
to generate more things and to think about itself
link |
and use that as a mechanism for creating more stuff.
link |
So like the life that this, however common it is,
link |
is an intrinsic part, is a fundamental part
link |
of this universe, at least, that we live in.
link |
I mean, it's always interesting to me
link |
because we have theories of quantum mechanics and gravity
link |
and they're supposed to be like
link |
our most fundamental theories right now.
link |
And they describe things like the interaction
link |
of massive bodies or the way that charges accelerate
link |
or all these kind of features.
link |
And they're these really deep theories
link |
and they tell us a lot about how reality works,
link |
but they're completely agnostic to our existence.
link |
And I can't help but think that whatever describes us
link |
has to be even deeper than that.
link |
And I think incorporating memory, I guess,
link |
causality, whatever the term you want to use
link |
into the physics view of the world might be.
link |
That's the easiest way to do it.
link |
It's the cleanest, so here we go again
link |
with the physicist, I'm a physicist.
link |
The cleanest, I was gonna say the simplest,
link |
most elegant way of resolving all of the kind of ways
link |
that we have these paradoxes associated with life
link |
when you, it's not that life is not,
link |
current physics is not incompatible with life,
link |
but it doesn't explain life.
link |
And then you want to know where are the explanatory gaps
link |
and this idea that we have an assembly
link |
that time is fundamental and objects actually
link |
are extended in time and have physical extent in time
link |
is the cleanest way of resolving
link |
a lot of the explanatory gaps.
link |
So I've been, I struggled with assembly theory
link |
for many years, because I could see this gap.
link |
And I think when I first met Sarah
link |
and we realized we were kind of talking
link |
about the same problem, but we were,
link |
we understood another language.
link |
It was quite hilarious actually,
link |
because it's like, I have no idea what you're talking about,
link |
but I think it sounds right.
link |
So for me, the most beautiful thing about assembly theory
link |
is I realized the assembly theory explains
link |
why the universe, why life is the universe
link |
developing a memory, but not only that poetically,
link |
I could actually go and measure it.
link |
And I was like, holy shit, we would just,
link |
we physically measured this thing,
link |
this abstract thing, and we could measure it.
link |
And not only could we measure it,
link |
but we can then start to quantify the causal consequences.
link |
Because, I mean, I think as a kind of inventing this
link |
together with Sarah and her team, I thought
link |
there was a quite a high chance that,
link |
we're doing science, there's such a high probability
link |
we're wrong on this.
link |
And I remember kind of trying to go to hard physicists,
link |
mathematicians, complexity theorists,
link |
and everyone just kind of giving me
link |
such a hard time about it.
link |
And so, this is kind of, you've just done this,
link |
you've just done that, you've just recapitulated
link |
And I was unable, I lacked the language to really explain,
link |
and I had to, it was a real struggle.
link |
So this realization that life, what life does
link |
that physics cannot understand, or chemistry,
link |
is the universe develops a memory
link |
that's causally actionable, and then we can measure it,
link |
but it isn't just a one thing,
link |
there is this intrinsic property
link |
of all the objects in the universe.
link |
Like I've said before, but me holding up this water bottle,
link |
it isn't any other water bottle,
link |
but it is a sum total of all the water bottles
link |
that have existed, right?
link |
And will likely change the future of water bottles
link |
and for other objects.
link |
So it's this kind of, so for me,
link |
assembly theory explains the soul in stuff.
link |
But it is the monology, it's not like
link |
Sheldrake's morphic resonance,
link |
where we have this kind of wooey thing
link |
permeating the universe, it is the interaction
link |
of objects of other objects.
link |
And some objects have more instantaneous causal power,
link |
that's life, living things, and some objects
link |
are the instantaneous output of that causal power,
link |
dead objects, but they're part of the lineage.
link |
And that for me is fascinating and really beautiful.
link |
And I think that even if we're determined
link |
to be totally wrong, I think it will help us,
link |
help hopefully understand what life is
link |
and go into tech life elsewhere and make life in the lab.
link |
How does that make you feel, by the way?
link |
Does it make you feel less special,
link |
that you're so deeply integrated,
link |
interconnected to the lineage?
link |
I mean, okay, on one level, I just wanted in my life
link |
as a scientist, I wanted to have an interesting idea
link |
just once or an original idea.
link |
I mean, it was like, you know, so I think that was cool
link |
that we had this idea and we were playing with it.
link |
And I think also that I kind of, I mean,
link |
it took me ages to realize that Sarah had also had
link |
the same kind of form, coming towards the same formulation
link |
just from a completely different point because I,
link |
but no, it makes me feel special.
link |
And it also makes me feel connected to the universe.
link |
It also makes me feel not just humble about, you know,
link |
being a living object in the universe,
link |
but the fact that it makes me really optimistic
link |
about what the universe is gonna do in the future
link |
because we're not just isolated phenomena, we are connected.
link |
I will be able to have, you know, one of my small objectives
link |
in life is to change the future of the universe
link |
in some profound way, just by existing.
link |
Yeah, that's not ambitious at all.
link |
I think it's also good because it makes me feel less lonely
link |
because I just realized I'm not like,
link |
I mean, I'm a unique assembly structure,
link |
but I have so much overlap with the other entities
link |
I interact with that we're not completely individual, right?
link |
And yet your existence does have a huge amount of impact
link |
on how this whole thing unrolls on the future of the world.
link |
As individuals, that's, yeah.
link |
But I was gonna say. Local packets of agency.
link |
I think we all have a profound impact on the future,
link |
some more than others, right?
link |
All human beings, all life.
link |
And I mean, that's why I think it's a privilege
link |
in a way for, you know, to say,
link |
to assert some degree of ego and agency,
link |
you know, I'm gonna make a computer
link |
or make an origin life machine or we can do this thing.
link |
But actually it's just like, you know,
link |
my life's probably living, if there is a God
link |
or there's a soul in everything,
link |
it's really laughing at us going,
link |
I fool these guys by giving them ego.
link |
So they strive for this stuff and look what it does
link |
for the assembly space of the universe.
link |
And there's always a possibility
link |
that science can't answer all of it.
link |
So that part's challenging for me.
link |
There might be a limit to this thing.
link |
Let me ask you a bunch of ridiculous questions
link |
and I demand relatively short answers.
link |
Lee, what's the scariest thing you've ever done?
link |
Or what's a scary thing that pops to mind?
link |
Giving seminars in front of other scientists.
link |
That's, yeah, that is terrifying.
link |
I could, if I had more time,
link |
I would ask you about the most embarrassing,
link |
but we'll spare you.
link |
What about you Sarah, scariest thing?
link |
Up there, some of the scary things you've done.
link |
Actually the scariest for me was deciding
link |
I wanted to get divorced
link |
because it was like a totally radical like.
link |
Life transformation.
link |
Yeah, because we had been married for a really long time.
link |
And I think it was just so much like,
link |
I realized like so much of my individual agency
link |
I didn't realize I had before.
link |
And that was just really like scary, like empowering scary,
link |
but like terrifying.
link |
Like you were living in a kind of one way
link |
for your whole life.
link |
And then you realized your life could be a different way.
link |
Yeah, there's a between humans.
link |
I mean, that's the beautiful thing about love
link |
is the connection you have,
link |
but it's also becomes a dependency and breaking that.
link |
Whether it's a mentor, what's your parents,
link |
your close friends.
link |
It's almost like waking up.
link |
Like just there's a different reality.
link |
Yeah, that was scary.
link |
Reinventing yourself.
link |
Okay, if you could, Lee,
link |
maybe I'll actually we'll alternate.
link |
Sarah, if you could be someone else for a day,
link |
someone alive today,
link |
you haven't met yet, haven't met yet.
link |
Or maybe you could do one who you've met.
link |
The woman's brilliant.
link |
I would just like to experience,
link |
like I just, I think she's got such an interesting
link |
and very deep understanding of social reality.
link |
But you also said you have appreciation
link |
of love for fashion.
link |
But that's actually the same.
link |
Like I just think it's really interesting
link |
because we live in a social reality
link |
which is completely artificially constructed.
link |
And some people are really genius about moving through that.
link |
And I think she's particularly good at it.
link |
I wonder if she's good at understanding or if she's just.
link |
I think it's very deeply intrinsic to her.
link |
So I don't know if she has much.
link |
She's like surfing a wave.
link |
How much cognitive awareness she has of it
link |
or how strategic it is.
link |
But I think it's deeply fascinating.
link |
So I guess that's the first one that comes to mind.
link |
What about you, Lee?
link |
If you could be somebody for a day.
link |
Don't say Yoshua Bach.
link |
Don't say Kim Kardashian.
link |
Let's do it off the table.
link |
No, I was gonna say I would like to be a,
link |
does it have to be here today?
link |
I was gonna say I'd like to be the latest arm processor.
link |
I would like to be the latest arm processor.
link |
I'd like to understand.
link |
I would like to know what it feel like to basically.
link |
You like being objects.
link |
I like being objects.
link |
I've always obsessed with being objects
link |
ever since I was a kid.
link |
What's the best part of being an arm processor for a day?
link |
I mean, I'd like to understand how I access my memory,
link |
what it anticipates coming next in clock cycles.
link |
What about how it feels like?
link |
Yeah, I wanna know how it feels like to be.
link |
All right, if, Lee, if everyone on Earth disappeared
link |
and it was just you left, what would your days look like?
link |
What would you do?
link |
Nobody else left to impress.
link |
Nobody, no, probably can't really do any real science
link |
What would you do with your remaining days?
link |
Get every possible tool I could
link |
and put it in my workshop and just make stuff.
link |
As, so try to make stuff.
link |
Just try and make stuff.
link |
I'm pretty much making companions probably, yeah.
link |
So in the physical space.
link |
What about you, Sarah?
link |
What would you, when you're just left alone on Earth,
link |
you're the last person.
link |
Are there animals in this scenario?
link |
I was gonna say, I would just,
link |
I would try to walk the entire planet,
link |
at least all the landmass.
link |
Well, that's true.
link |
So you probably don't know if there's stuff.
link |
You could be searching for plants or other humans
link |
or other animals. And what would I eat?
link |
You just have daily just allotment.
link |
I would just walk all the time, I think.
link |
That's just what came to mind.
link |
You're the explorer.
link |
I would just walk.
link |
And I guess I would make a goal of covering
link |
all of the entire Earth.
link |
Because what else are you gonna do with your time?
link |
What's an item on your bucket list, Sarah,
link |
that you haven't done yet, but you hope to do?
link |
You know what's funny with my bucket list?
link |
I only know it was on my bucket list once I check it off.
link |
Once you check it off.
link |
So your bucket list is like a fog.
link |
It's like a mystery almost by doing it.
link |
Yeah, so it's very subconsciously driven.
link |
So it's in your subconscious in there.
link |
You're bringing it to the surface.
link |
I think most of the steering of our agency
link |
is in our subconscious anyway,
link |
so I just kind of go with the flow.
link |
But I guess, no, seriously.
link |
Yeah, no, I get it.
link |
I guess, but I would like to go on a submarine,
link |
like to the bottom of the ocean.
link |
I think that'd be really cool.
link |
To the bottom of the ocean.
link |
Are you captivated by the mystery of the ocean?
link |
Like how little you know.
link |
Yeah, what about you, Lee?
link |
What item on your bucket list?
link |
I don't have a bucket list, but I've just made one.
link |
I would love to take a computer to the moon or Mars
link |
and make drugs off world.
link |
Be the first chemist to make drugs off world.
link |
The first drug manufacturer in space.
link |
Do they have to be somehow like be able to habitate,
link |
like be able to survive on that particular space?
link |
Or like what's the connection between being on Mars
link |
and doing many things?
link |
I just would like to be there.
link |
I would like to take the ability to have command
link |
and control over chemicals programmatically off earth
link |
to somewhere else in the universe.
link |
That just seems like you like difficulty
link |
engineering problems.
link |
Before I die, if I can do that, that's great.
link |
Would you travel to space if you could?
link |
Yeah, yeah, that's what I'm saying.
link |
I'd love to go into space, but not just to be a tourist.
link |
I wanna take scientific experiment in space
link |
and do a thing in space that's never been done before.
link |
That's a real possibility.
link |
So that's why there's no point in listening
link |
things I can't do, yeah.
link |
All right, what small act of kindness
link |
were you once shown that you will never forget?
link |
Small act of kindness, not big.
link |
Somebody was just kind to you.
link |
Somebody did something sweet.
link |
When I was a PhD student, someone helped me out
link |
with just, basically, I needed a computer.
link |
I needed some power, computation power,
link |
and someone took pity on me and helped me.
link |
I was really touched.
link |
They didn't have to.
link |
And they were actually quite, they were a disabled scientist
link |
and they had other things to do
link |
rather than help some random PhD student.
link |
Gave me access, taught me a lot of stuff.
link |
Yeah, actually, when you're a grad student
link |
or when you're a student, when you're even a student,
link |
the younger it is, the better.
link |
The attention, the support, the love you get
link |
from an older person, a teacher,
link |
something like that is super powerful.
link |
And from the perspective of the teacher,
link |
they might not realize the impact they have,
link |
but that little bit, those few words,
link |
a little bit of help can have a lot of impact.
link |
What about you, Sarah?
link |
Somebody give you a free Starbucks at some point?
link |
I love free Starbucks.
link |
I like it when you're in the line at Starbucks
link |
and somebody buys your coffee in front of you
link |
and then you buy the next one.
link |
I love those, but that's not my example.
link |
It makes me happy.
link |
And now my kids get all excited when we do it,
link |
when we go in, we're the first ones in line doing it.
link |
But I guess I can use a similar example
link |
about just being a student.
link |
So Paul Davies is a very well known theoretical physicist
link |
and he was generous enough with his time
link |
to take me on as a postdoc.
link |
But before I became his postdoc,
link |
he invited me to a workshop
link |
at Arizona State University in the Beyond Center
link |
and took a walk with me around campus
link |
just to talk about ideas after.
link |
And I think there were two things
link |
that were completely generous about that.
link |
One is Paul's philosophy is always interacting
link |
with young people.
link |
It's like you interact with a mind in the room.
link |
It doesn't matter how well known or whatever.
link |
It's like you evaluate the person for the person.
link |
But he also gave me a book,
link |
The Eerie Silence that he had written and he wrote in it,
link |
This is How EE Gets to ET,
link |
which was an antimeric excess,
link |
which I worked on as a PhD student
link |
was the origin of homochirality
link |
all the way up to what the book was about,
link |
which was are we alone in the universe
link |
and is there intelligent life out there?
link |
And it was just so much about the questions I wanted to ask
link |
because it was just everything about,
link |
just it was just really, really kind.
link |
Like that it's okay to ask these questions
link |
and you can actually have strong enough to answer them.
link |
A lot of my career is mostly his encouragement
link |
to ask deep questions.
link |
Like he gave me the space to do it
link |
in ways that a lot of previous mentors had.
link |
I mean, I've had a good experience with mentors,
link |
but it was like go off the deep end,
link |
ask the hardest questions.
link |
And I think that's the best gift you can give somebody.
link |
because you're both fascinating minds
link |
and not, I would say, nonstandard in the best possible way.
link |
Is there advice you can give to young folks
link |
how to be nonstandard, how to stand out,
link |
novelty, how to generate novelty?
link |
That's what I want on my tombstone.
link |
He generated novelty.
link |
I just love doing science.
link |
And so when I was younger, I was just,
link |
just wanted to, I mean, I'm still not sure
link |
I'm a real scientist, right?
link |
So my advice for the young people
link |
is just, if you just, if you love asking questions,
link |
then don't be afraid to ask the question,
link |
even if it pisses people off,
link |
because if you piss people off,
link |
you're probably asking the right question.
link |
What I would say though, is don't do what I did,
link |
which is just piss everyone off.
link |
Try and work out how to, you know, I think,
link |
if other people are challenged by your questions,
link |
you will get not only respect,
link |
but people will give you, create space for you,
link |
because you're doing something really new.
link |
I really try to create space in my academic career,
link |
with my team, really try and praise them
link |
and push them to do new things.
link |
So my advice is, try to do new things,
link |
get feedback, and the universe will help you.
link |
Because the universe likes novelty.
link |
I think so, right?
link |
This one will keep them around.
link |
What about you, Sarah?
link |
You too like to ask the really out there big questions.
link |
Yeah, because I have a strong passion for them.
link |
So I think it goes back to the love.
link |
Like if you're doing the thing you're supposed to be doing,
link |
you should really love it.
link |
So I always tell people that they should do
link |
the thing they're most passionate about.
link |
But I think a flip side of that is,
link |
that's when you become in some way,
link |
like not to sound cheesy,
link |
but like your best version of yourself.
link |
So I guess for me, as I become more successful in my career,
link |
I feel like I can be more myself as an individual.
link |
And so there's this, I've always been following
link |
the questions I'm most interested in,
link |
which very early on I was discouraged
link |
from doing by many people
link |
because they thought they were unanswerable questions.
link |
And I always just thought,
link |
well, if no one's even trying to answer them,
link |
of course they're gonna be unanswerable.
link |
And then that was kind of an odd viewpoint.
link |
But the more I found my way in that space,
link |
the more I also made a space for myself as a person,
link |
because you're basically generating the niche
link |
that you wanna exist in.
link |
And so I think that's part of it,
link |
is not just to follow your passion,
link |
but also think about like,
link |
who do you wanna be and create that?
link |
Yeah, who am I, who do you wanna be?
link |
I mean, yeah, play temporally with it.
link |
Yeah, who am I now?
link |
Who do I wanna be now?
link |
But who do I wanna be in the future?
link |
They're not decoupled.
link |
Yeah, I always wonder if that's like,
link |
if I become something, am I finding myself
link |
or am I creating myself?
link |
And I think those are somehow the same kind of thing.
link |
I do feel often like I was always meant
link |
to be this kind of thing.
link |
But is that created or discovered?
link |
But basically go towards that direction.
link |
If you were abducted by aliens, Sarah.
link |
Excellent, I'm waiting, they can come find me.
link |
They're on a spaceship.
link |
And then they somehow figured out the language you speak
link |
and ask you, what are you?
link |
What is, explain yourself.
link |
Not you, Sarah, but the species.
link |
Life on Earth, like we don't have time.
link |
We're busy grad students from another planet.
link |
What's interesting about human civilization?
link |
What's interesting about you?
link |
You specifically, too.
link |
They could be very kind of personal, kind of pushy.
link |
And yeah, how would you begin to describe?
link |
Because, you know, obviously I self identify
link |
as a scientist and a physicist,
link |
but intrinsically I feel more like an artist.
link |
But it's almost like you're an artist
link |
that you don't know what you're painting yet.
link |
And I guess I feel like that's humanity,
link |
like in some sense, we're creating something
link |
I think is profound and potentially very beautiful
link |
in existence of the universe.
link |
But we're just so, not night, we're just early.
link |
We're early, we're young.
link |
We don't know what we're doing yet.
link |
Yeah, what's with the nuclear weapons?
link |
That'd be a question, too.
link |
Like what are you guys, why?
link |
What are we doing with them?
link |
This creativity that you talk, it sounds very nice,
link |
but it's, you're seem to be.
link |
We're making things that are.
link |
Like very destructive and like the rockets,
link |
this seems very aggressive.
link |
This is my blinders on.
link |
I mean, it goes back to the whole conversation
link |
I have a hard time regularizing certain aspects
link |
of reality into what I wanna envision.
link |
And that's obviously problematic.
link |
But nuclear power has also given us a lot of good things.
link |
So both, that's human nature.
link |
Both human beings and the technology we create
link |
has the capacity for evil and the capacity for good.
link |
Yeah, and we can't all be good all the time.
link |
I mean, there's like this huge misnomer
link |
that you need to be liked by everyone universally.
link |
And obviously that's like an ideal,
link |
but it's physically impossible.
link |
Like you can't get a group of people in a room
link |
and have everyone like each other all the time.
link |
So I think that kind of tension is actually really important
link |
that we have different aesthetics, different goals.
link |
And sometimes conflict comes out of that.
link |
Speaking of which, do Yuli and Yoshua Bach
link |
ever say anything nice to each other?
link |
Or is it always conflict?
link |
We never have conflict.
link |
We argue, but I don't think arguments are bad.
link |
I mean, I think the problem I have,
link |
not problem, I think.
link |
And he's not here to defend himself.
link |
No, I just, I don't necessarily understand the,
link |
I mean, he's just talking at such a high level.
link |
You know, I'm a dimwit, so I'm like, I spend some,
link |
so I think a lot of our conflict is not conflict.
link |
We actually, we actually have a, I think,
link |
I mean, I can't speak for Yash,
link |
I have a deep appreciation for him, he's brilliant.
link |
But I think I'm kind of frustrated and I'm trying to,
link |
he thinks the universe is a computer
link |
and I want to turn the universe into a computer.
link |
Yeah, that's a small disagreement.
link |
So what would you, how would you defend your life
link |
to an alien when you're being abducted?
link |
Would you focus on the specifics of your life?
link |
No, no, no, I would be, I would try to be
link |
as random as possible and try and confuse them.
link |
That might be the wiser choice.
link |
The Easter eggs in reality.
link |
No, I mean, if aliens abducted me.
link |
Would you play dumb?
link |
No, no, no, I would try and be as random as possible.
link |
I would try and do something that would surprise
link |
the hell out of them, which I thought,
link |
I mean, I'd probably risk it, they might kill me,
link |
but I think that'd probably be funny.
link |
That might, yeah, they might want to study you
link |
for prolonged periods of time.
link |
My reasoning is, if I wanted to stay alive, okay,
link |
so if the thing is, if I wasn't going back to Earth
link |
and the job was to stay alive, if I could be
link |
as surprising as possible, they'd keep me around
link |
like a pet, right?
link |
Petly, on the aliens page book.
link |
So you'd be okay being a pet?
link |
Well, no, but I mean.
link |
The last human that survives would just be
link |
a pet to the aliens.
link |
I don't know, but I mean, I think that might be fun
link |
because then I might get some feedback
link |
from their curiosity, but yeah.
link |
Let me ask you this question.
link |
Given our conversation has a very different meaning,
link |
not a more profound meaning, perhaps,
link |
but would you rather lose all of your old memories
link |
or never be able to make new ones?
link |
I would have to lose all my old memories.
link |
Again, it's the novelty.
link |
What about you, Sarah?
link |
I'm the same because I don't think,
link |
like, it's about the future experience, right?
link |
And in some sense, like you were saying earlier,
link |
most of our lived experience is actually in our memories.
link |
So if you can't generate new memories,
link |
it's like you're not alive anymore.
link |
What comforts you on bad days?
link |
When you look at human civilization,
link |
when you look at your own life,
link |
what gives you hope?
link |
What makes you feel good about what we're doing
link |
about life at the small scale of you as a human
link |
and at the big scale of us as a human civilization,
link |
maybe the big scale of the universe?
link |
Children, my kids.
link |
But I also mean that in like a grand sense of like,
link |
not a grand, but like future minds in some sense.
link |
So for me, like the most bleak movie ever,
link |
people worry about apocalyptic things
link |
like AI existential risk and climate change,
link |
which children of men.
link |
The whole premise of the movie was
link |
there can be no children born on the entire planet.
link |
And the youngest person on the planet
link |
is like 18 years old or something.
link |
Like, can you imagine a world without children?
link |
It's just, it's harrowing.
link |
That's the scariest thing.
link |
So I think what gives me hope is always youth
link |
and the hope of children
link |
and the possibilities of the future they see.
link |
And they grow up in a completely different reality
link |
And I think we have a hard time seeing
link |
what their reality actually looks like.
link |
But I think most of the time it's super interesting.
link |
Yeah, they have dreams, they have imagination,
link |
they have this kind of excitement.
link |
It's so cool, it's so fun to watch.
link |
And yeah, you feel like you're almost getting in the way
link |
of all that imagination.
link |
What about you, Lee?
link |
What gives you hope?
link |
So when I go back to my eight year old self,
link |
the thing that I dreamed of as my eight year old self
link |
was this world in which technology became programmable
link |
and there was the internet and I'd get information
link |
and I would expand my consciousness by just,
link |
you know, getting access to everything that was going on.
link |
And this happened in my lifetime.
link |
I mean, we really do have that.
link |
I mean, okay, there's some bad things.
link |
You know, there's TikTok, everyone just, whatever,
link |
all the bad things about social media.
link |
I mean, I can't quite believe my luck being born now.
link |
Being able to program reality in some way.
link |
Yeah, and the thing that I really find fascinating
link |
about human beings is just how ingenious they are.
link |
I'm, you know, whether it's from my kids,
link |
my research group, my peers, other companies,
link |
just how ingenious everyone is.
link |
And I'm pretty sure humanity has,
link |
or our causal chain in which humanity is a vital part
link |
in the future is gonna have a lot of fun.
link |
And I'm just, yeah, it's just mind blowing just to watch.
link |
And, you know, so humans are ingenious
link |
and I hope to help them be more ingenious if I can.
link |
Well, what gives me hope,
link |
what makes me feel good on bad days
link |
is the existence of wild minds like yours,
link |
novelty generators, assembly structures
link |
that generate novelty and do so beautifully
link |
and then tweet about it.
link |
Sarah, I really, really enjoy talking to you.
link |
I enjoy following you.
link |
Sarah Lee, I hope to talk to you many times
link |
in the future, maybe with Yoshua Bach.
link |
You're just incredible people.
link |
Thank you for everything you do.
link |
Thank you for talking today.
link |
Really, really appreciate it.
link |
Yeah, brilliant to be here.
link |
Thanks for listening to this conversation
link |
with Sarah Walker and Lee Cronin.
link |
To support this podcast,
link |
please check out our sponsors in the description.
link |
Now, let me leave you with some words from Arthur C. Clarke.
link |
Two possibilities exist.
link |
Either we are alone in the universe or we are not.
link |
Both are equally terrifying.
link |
And let me, if I may, add to that
link |
by saying that both possibilities, at least to me,
link |
are both terrifying and exciting.
link |
And keeping these two feelings in my heart
link |
is a fun way to explore, to wonder, to think, and to live,
link |
always a little bit on the edge of madness.
link |
Thank you for listening.
link |
I hope to see you next time.