back to indexJoscha Bach: Nature of Reality, Dreams, and Consciousness | Lex Fridman Podcast #212
link |
The following is a conversation with Yosha Bach, his second time in the podcast.
link |
Yosha is one of the most fascinating minds in the world, exploring the nature of intelligence,
link |
cognition, computation, and consciousness.
link |
To support this podcast, please check out our sponsors, Coinbase, Codecademy, Linode,
link |
Netsuite, and ExpressVPN.
link |
Their links are in the description.
link |
This is the Lex Friedman podcast, and here is my conversation with Yosha Bach.
link |
Thank you for once again coming on to this particular Russian program and sticking to
link |
the theme of a Russian program.
link |
Let's start with the darkest of topics.
link |
So this is inspired by one of your tweets.
link |
You wrote that, quote, when life feels unbearable, I remind myself that I'm not a person.
link |
I am a piece of software running on the brain of a random ape for a few decades.
link |
It's not the worst brain to run on.
link |
Have you experienced low points in your life?
link |
Have you experienced depression?
link |
Of course, we all experience low points in our life, and we get appalled by the things,
link |
by the ugliness of stuff around us.
link |
We might get desperate about our lack of self regulation, and sometimes life is hard.
link |
And I suspect you don't get to your life, nobody does, to get to their life without
link |
low points and without moments where they are despairing.
link |
And I thought that let's capture this state and how to deal with that state.
link |
And I found that very often you realize that when you stop taking things personally, when
link |
you realize that this notion of a person is a fiction, similar as it is in Westworld where
link |
the robots realize that their memories and desires are just stuff that keeps them in
link |
the loop, and they don't have to act on those memories and desires, that our memories and
link |
expectations is what make us unhappy.
link |
And the present rarely does.
link |
The day in which we are, for the most part, it's okay, right?
link |
When we are sitting here, right here, right now, we can choose how we feel.
link |
And the thing that affects us is the expectation that something is going to be different from
link |
what we wanted to be or the memory that something was different from what we wanted it to be.
link |
And once we basically zoom out from all this, what's left is not a person, what's left
link |
is this state of being conscious, which is a software state.
link |
And software doesn't have an identity, it's a physical law.
link |
And it's a law that acts in all of us, and it's embedded in a suitable substrate.
link |
And we didn't pick that substrate, right?
link |
We are mostly randomly instantiated on it.
link |
And they're all these individuals, and everybody has to be one of them.
link |
And eventually, you're stuck on one of them and have to deal with that.
link |
So you're like a leaf floating down the river.
link |
You just have to accept that there's a river and you just float wherever there takes you.
link |
You don't have to do this.
link |
The thing is that the illusion that you are an agent is a construct.
link |
And what part of that is actually under your control.
link |
And I think that our consciousness is largely a control model for our own attention.
link |
So we notice where we are looking and we can influence what we are looking, how we are
link |
disambiguating things, how we put things together in our mind.
link |
And the whole system that runs us is this big cybernetic motivational system.
link |
So we're basically like a little monkey sitting on top of an elephant.
link |
And we can prod this elephant here and there to go this way or that way.
link |
And we might have the illusion that we are the elephant or that we are telling it what
link |
And sometimes we notice that it walks into a completely different direction.
link |
And we didn't set this thing up.
link |
It just is the situation that we find ourselves in.
link |
How much prodding can we actually do of the elephant?
link |
But I think that our consciousness cannot create the motive force.
link |
Is the elephant consciousness in this metaphor?
link |
No, the monkey is the consciousness.
link |
The monkey is the attentional system that is observing things.
link |
There's a large perceptual system combined with a motivational system that is actually
link |
providing the interface to everything and our own consciousness, I think, is the tool
link |
that directs the attention of that system, which means it singles out features and performs
link |
conditional operations for which it needs an index memory.
link |
But this index memory is what we perceive as our stream of consciousness.
link |
But the consciousness is not in charge.
link |
That's an illusion.
link |
So everything outside of that consciousness is the elephant.
link |
So it's the physics of the universe, but it's also society that's outside of your...
link |
I would say the elephant is the agent.
link |
So there is an environment through which the agent is stomping, and you are influencing
link |
a little part of that agent.
link |
So is the agent a single human being?
link |
What which object has agency?
link |
That's an interesting question.
link |
I think a way to think about an agent is that it's a controller with a set point generator.
link |
The notion of a controller comes from cybernetics and control theory.
link |
Control system consists out of a system that is regulating some value and the deviation
link |
of that value from a set point.
link |
And it has a sensor that measures the system's deviation from that set point and an effector
link |
that can be parameterized by the controller.
link |
So the controller tells the effector to do a certain thing.
link |
And the goal is to reduce the distance between the set point and the current value of the
link |
And there's an environment which disturbs the regulated system, which brings it away
link |
from that set point.
link |
So the simplest case is the thermostat.
link |
The thermostat is really simple because it doesn't have a model.
link |
The thermostat is only trying to minimize the set point deviation in the next moment.
link |
And if you want to minimize the set point deviation over a longer time span, you need to integrate
link |
it, you need to model what is going to happen.
link |
So for instance, when you think about that your set point is to be comfortable in life,
link |
maybe you need to make yourself uncomfortable first.
link |
So you need to make a model of what's going to happen when.
link |
And this is the task of the controller is to use its sensors to measure the state of
link |
the environment and the system that is being regulated and figure out what to do.
link |
And if the task is complex enough, the set points are complicated enough.
link |
And if the controller has enough capacity and enough sensor feedback, then the task
link |
of the controller is to make a model of the entire universe that it's in, the conditions
link |
under which it exists and of itself.
link |
And this is a very complex agent and we are in that category.
link |
And an agent is not necessarily a thing in the universe.
link |
It's a class of models that we use to interpret aspects of the universe.
link |
And when we notice the environment around us, a lot of things only make sense at the level
link |
that should we're entangled with them, we interpret them as control systems that make
link |
models of the world and try to minimize their own set points.
link |
So the models are the agents?
link |
The agent is a class of model.
link |
And we notice that we are an agent ourselves.
link |
We are the agent that is using our own control model to perform actions.
link |
We notice we would use a change in the model and things in the world change.
link |
And this is how we discover the idea that we have a body, that we are a situated environment
link |
and that we do have a first person perspective.
link |
Still don't understand what's the best way to think of which object has agency with respect
link |
Is it the contents of the brain that has agency?
link |
Like what's the actuators that you're referring to?
link |
What is the controller and where does it reside?
link |
Or is it these impossible things?
link |
Because I keep trying to ground it to space time, the three dimension of space and the
link |
one dimension of time.
link |
What's the agent in that for humans?
link |
There is not just one.
link |
It depends on the way in which you're looking at the thing in which you're framing it.
link |
Imagine that you are, say, Angela Merkel and you are acting on behalf of Germany.
link |
And you could say that Germany is the agent.
link |
And in the mind of Angela Merkel, she is Germany to some extent because in the way in which
link |
she acts, the destiny of Germany changes.
link |
There are things that she can change that basically affect the behavior of that nation
link |
So it's hierarchies to go to another one of your tweets with, I think you're playfully
link |
mocking Jeff Hawkins with saying his brain's all the way down.
link |
So it's like, it's agents all the way down.
link |
It's agents made up of agents, made up of agents.
link |
Like if Angela Merkel is Germany and Germany is made up a bunch of people and the people
link |
are themselves agents in some kind of context.
link |
And then people are made up of cells, each individual.
link |
So is it agents all the way down?
link |
I suspect that has to be like this in a world where things are self organizing.
link |
Most of the complexity that we are looking at, everything in life is about self organization.
link |
So I think up from the level of life, you have agents.
link |
And below life, you rarely have agents because sometimes you have control systems that emerge
link |
randomly in nature and try to achieve a set point, but they're not that interesting agents
link |
And because to make an interesting model of the world, you typically need a system that
link |
is during complete.
link |
Can I ask you a personal question?
link |
What's the line between life and non life?
link |
It's personal because you're a life form.
link |
So what do you think in this emerging complexity, at which point does the things that are being
link |
living and have agency?
link |
Personally, I think that the simplest answer that is that life is cells.
link |
So it's a particular kind of principle that we have discovered to exist in nature.
link |
It's modular stuff that consists out of basically this DNA tape with a redried head on top
link |
of it that is able to perform arbitrary computations and state transitions within the cell.
link |
And it's combined with a membrane that insulates the cell from its environment.
link |
And there are chemical reactions inside of the cell that are in disequilibrium.
link |
And the cell is running in such a way that this disequilibrium doesn't disappear.
link |
And if the cell goes into an equilibrium state, it dies.
link |
And it requires some things like a neck entropy extractor to maintain this disequilibrium.
link |
So it's able to harvest neck entropy from its environment and keep itself running.
link |
So there's information and there's a wall to maintain this disequilibrium.
link |
But isn't this very earth centric like what you're referring to as...
link |
I'm not making a normative notion.
link |
You could say that there are probably other things in the universe that are cell like
link |
And you could also call them life.
link |
But eventually it's just a willingness to find an agreement of how to use the terms.
link |
I like cells because it's completely co extension of this the way that we use the word even
link |
before we knew about cells.
link |
So people were pointing at some stuff and saying, this is somehow animate.
link |
And this is very different from the non animate stuff.
link |
And what's the difference between the living and the dead stuff?
link |
And it's mostly whether the cells are working or not.
link |
And also this boundary of life where we say that, for instance, the virus is basically
link |
an information packet that is subverting the cell and not life by itself.
link |
That makes sense to me.
link |
And it's somewhat arbitrary.
link |
You could of course say that systems that permanently maintain a disequilibrium and
link |
can self replicate are always life.
link |
And maybe that's a useful definition too.
link |
But this is eventually just how you want to use the word.
link |
Is it so useful for conversation?
link |
But is it somehow fundamental to the universe?
link |
Do you think there's a actual line to eventually be drawn between life and non life?
link |
Or is it all kind of continuum?
link |
I don't think it's a continuum, but there's nothing magical that is happening.
link |
Living systems are a certain type of machine.
link |
What about non living systems?
link |
Is it also a machine?
link |
There are non living machines, but the question is at which point is the system able to perform
link |
arbitrary state transitions to make representations?
link |
And living things can do this.
link |
And of course, we can also build non living things that can do this, but we don't know
link |
anything in nature that is not a cell and is not created by Stalala Life that is able
link |
Not only do we not know.
link |
I don't think we have the tools to see otherwise.
link |
I always worry that we look at the world too narrowly.
link |
There could be life of a very different kind right under our noses that we're just not
link |
seeing because we're not either limitations of our cognitive capacity or we're just not
link |
open minded enough, either with the tools of science or just the tools of our mind.
link |
Yeah, that's possible.
link |
I find this thought very fascinating and I suspect that many of us ask ourselves in childhood,
link |
what are the things that we are missing?
link |
What kind of systems and interconnections exist that are outside of our gaze?
link |
But we are looking for it and physics doesn't have much room at the moment for opening up
link |
something that would not violate the conservation of information as we know it.
link |
Yeah, but I wonder about time scale and spatial scale, whether we just need to open up our
link |
idea of how life presents itself.
link |
It could be operating in a much slower time scale, a much faster time scale.
link |
It's almost sad to think that there's all this life around us that we're not seeing
link |
because we're just not thinking in terms of the right scale, both time and space.
link |
What is your definition of life?
link |
What do you understand as life?
link |
Entities of sufficiently high complexity, they're full of surprises.
link |
I don't have a free will, so that just came out of my mouth, I'm not sure that even makes
link |
There are certain characteristics, so complexity seems to be a necessary property of life and
link |
I almost want to say it has ability to do something unexpected.
link |
It seems to me that life is the main source of complexity on earth.
link |
And complexity is basically a bridgehead that order builds into chaos by modeling, by processing
link |
information in such a way that you can perform reactions that would not be possible for dump
link |
And this means that you can harvest like entropy, that dump systems cannot harvest, and this
link |
is what complexity is mostly about.
link |
In some sense, the purpose of life is to create complexity.
link |
I mean, there seems to be some kind of universal drive towards increasing pockets of complexity.
link |
I don't know what that is.
link |
That seems to be like a fundamental, I don't know if it's a property of the universe or
link |
it's just a consequence of the way the universe works, but there seems to be this small pockets
link |
of emergent complexity that builds on top of each other and starts having like greater
link |
and greater complexity by having like a hierarchy of complexity.
link |
Little organisms building up a little society that then operates almost as an individual
link |
organism itself and all of a sudden you have Germany and Merkel.
link |
That's not obvious to me.
link |
Everything that goes up has to come down at some point, right?
link |
So if you see this big exponential curve somewhere, it's usually the beginning of an S curve or
link |
something eventually reaches saturation and the S curve is the beginning of some kind
link |
of bump that goes down again.
link |
And there is just this thing that when you are in sight of an evolution of life, you
link |
are on top of a puddle of negentropy that is being sucked dry by life.
link |
And during that happening, you see an increase in complexity because life forms are competing
link |
with each other to get more and more and finer and finer corner of that negentropy extraction.
link |
But I feel like that's a gradual, beautiful process like that almost follows a process
link |
akin to evolution.
link |
And the way it comes down is not the same way it came up.
link |
The way it comes down is usually harshly and quickly.
link |
So usually there's some kind of catastrophic event.
link |
The Roman Empire took a long time.
link |
But would that be, would you classify that as a decrease in complexity though?
link |
I think that this size of the cities that could be fed has decreased dramatically.
link |
And you could see that the quality of the art decreased and it did so gradually.
link |
And maybe future generations, when they look at the history of the United States in the
link |
21st century, will also talk about the gradual decline, not something that suddenly happens.
link |
Do you have a sense of where we are?
link |
Are we on the exponential rise?
link |
Are we at the peak?
link |
Or are we at the downslope of the United States Empire?
link |
It's very hard to say from a single human perspective, but it seems to me that we are
link |
probably at the peak.
link |
I think that's probably the definition of optimism and cynicism.
link |
So my nature of optimism is, I think we're on the rise, but I think this is just all
link |
the matter of perspective that nobody knows.
link |
But I do think that erring on the side of optimism, like you need a sufficient number,
link |
you need a minimum number of optimists in order to make that up thing actually work.
link |
And so I tend to be on the side of the optimists.
link |
I think that we are basically species of grasshoppers that have turned into locusts.
link |
And when you are in that locust mode, you see an amazing rise of population numbers
link |
and of the complexity of the interactions between the individuals.
link |
But ultimately, the question is, is it sustainable?
link |
See, I think we're a bunch of lions and tigers that have become domesticated cats to use
link |
a different metaphor.
link |
And so I'm not exactly sure we're so destructive or just softer and nicer and lazier.
link |
I think we have monkeys and not the cats.
link |
And if you look at the monkeys, they are very busy.
link |
The ones that have a lot of sex, those monkeys?
link |
Not just the bonobos.
link |
I think that all the monkeys are basically a discontent species that always needs to
link |
Well, the gorillas seem to have a little bit more of a structure, but it's a different
link |
Okay, you mentioned the elephant and the monkey riding the elephant, and consciousness is
link |
And there's some prodding that the monkey gets to do.
link |
And sometimes the elephant listens.
link |
I heard you got into some content.
link |
Maybe you can correct me, but I heard you got into some contentious free will discussions.
link |
Is this with Sam Harris or something like that?
link |
Not that I know of.
link |
People in clubhouse told me you made a bunch of big debate points about free will.
link |
Well, let me just then ask you, where in terms of the monkey and the elephant, do you think
link |
we land in terms of the illusion of free will?
link |
How much control does the monkey have?
link |
We have to think about what the free will is in the first place.
link |
We are not the machine.
link |
We are not the thing that is making the decisions.
link |
We are a model of that decision making process.
link |
And there is a difference between making your own decisions and predicting your own decisions.
link |
And that difference is the first person perspective.
link |
And what basically makes decision making and the conditions of free will distinct from
link |
just automatically doing the best thing is that we often don't know what the best thing
link |
We make decisions under uncertainty.
link |
We make informed bets using a betting algorithm that we don't yet understand because we haven't
link |
reverse engineered our own minds efficiently.
link |
We don't know the expected rewards.
link |
We don't know the mechanism by which we estimate the rewards and so on.
link |
But there is an algorithm.
link |
We only observe ourselves performing where we see that we weight facts and factors and
link |
the future and then some kind of possibility, some motive gets raised to an intention.
link |
And that's informed bet that the system is making.
link |
And that making of the informed bet, the representation of that is what we call free
link |
And it seems to be paradoxical because we think that's the crucial thing is about it
link |
that it's somehow indeterministic.
link |
And yet if it was indeterministic, it would be random.
link |
And it cannot be random because if it was random, if just dice were being thrown in
link |
the universe randomly forces you to do things, it would be meaningless.
link |
So the important part of the decisions is always the deterministic stuff.
link |
But it appears to be indeterministic to you because it's unpredictable.
link |
Because if it was predictable, you wouldn't experience it as a free will decision.
link |
You would experience it as just doing the necessary right thing.
link |
And you see this continuum between the free will and the execution of automatic behavior
link |
when you're observing other people.
link |
So for instance, when you are observing your own children, if you don't understand them,
link |
you will use this agent model where you have an agent with a set point generator.
link |
And the agent is doing the best it can to minimize the difference to the set point.
link |
And it might be confused and sometimes impulsive or whatever, but it's acting on its own free
link |
And when you understand what happens in the mind of the child, you see that it's automatic.
link |
And you can outmodel the child, you can build things around the child that will lead the
link |
child to making exactly the decision that you are predicting.
link |
And under these circumstances, like when you were a stage misvision or somebody who is
link |
dealing with people that you sell a car to, and you completely understand the psychology
link |
and the impulses and the space of thoughts that this individual can have at that moment.
link |
Under these circumstances, it makes no sense to attribute free will.
link |
Because it's no longer decision making under uncertainty.
link |
You are already certain.
link |
For them, there's uncertainty, but you already know what they're doing.
link |
But what about for you?
link |
So is this akin to like systems like cellular automata, where it's deterministic, but when
link |
you squint your eyes a little bit, it starts to look like there's agents making decisions
link |
at the higher sort of when you zoom out and look at the entities that are composed by the
link |
You know, there's underlying simple rules that make the system evolve in deterministic
link |
It looks like there's organisms making decisions.
link |
Is that where the illusion of free will emerges?
link |
That jump in scale?
link |
It's a particular type of model, but this jump in scale is crucial.
link |
The jump in scale happens whenever you have too many parts to count, and you cannot make
link |
a model at that level, and you try to find some higher level regularity.
link |
And the higher level regularity is a pattern that you project into the world to make sense
link |
And agency is one of these patterns, right?
link |
You have all these cells that interact with each other, and the cells in our body are
link |
set up in such a way that they benefit if their behavior is coherent, which means that
link |
they act as if they were serving a common goal, and which that means that they will
link |
evolve regulation mechanisms that act as if they were serving a common goal.
link |
And now you can make sense of all these cells by projecting the common goal into them.
link |
So for you, then, free will is an illusion?
link |
No, it's a model, and it's a construct.
link |
It's basically a model that the system is making of its own behavior, and it's the
link |
best model that it can come up with under the circumstances, and it can get replaced
link |
by a different model, which is automatic behavior, when you fully understand the mechanism under
link |
which you are acting.
link |
Yeah, but another word for model is what, story.
link |
So the story you're telling, I mean, do you actually have control?
link |
Is there such a thing as a you, and is there such a thing as you having control?
link |
So like, are you manifesting your evolution as an entity?
link |
In some sense, the you is the model of the system that is in control.
link |
It's a story that the system tells itself about somebody who is in control.
link |
And the contents of that model are being used to inform the behavior of the system.
link |
So the system is completely mechanical, and the system creates that story like a loom,
link |
and then it uses the contents of that story to inform its actions and writes the results
link |
of that action into the story.
link |
So how's that non illusion?
link |
The story is written then, or a rather, we're not the writers of the story.
link |
Yes, but we always knew that.
link |
No, we don't know that.
link |
When did we know that?
link |
I think that's mostly a confusion about concepts.
link |
The conceptual illusion in our culture comes from the idea that we live in physical reality,
link |
and that we experience physical reality, and that we have ideas about it.
link |
And then you have this dualist interpretation, where you have two substances, res extensor,
link |
the world that you can touch, and that is made of extended things, and res cogitans,
link |
which is the world of ideas, and in fact, both of them are mental representations.
link |
One is the representations of the world as a game engine that your mind generates to
link |
make sense of the perceptual data, and the other one, yes, that's what we perceive as
link |
the physical world.
link |
But we already know that the physical world is nothing like that, right?
link |
But the mechanics is very different from what you and me perceive as the world.
link |
The world that you and me perceive is a game engine, and there are no colors and sounds
link |
in the physical world that only exist in the game engine generated by your brain.
link |
And then you have ideas that cannot be mapped onto extended regions, right?
link |
So the objects that have a spatial extension in the game engine, res extensor, and the
link |
objects that don't have a physical extension in the game engine are ideas.
link |
And they both interact in our mind to produce models of the world.
link |
But when you play video games, I understand that what's actually happening is zeros and
link |
ones inside of a computer, inside of a CPU and a GPU, but you're still seeing the rendering
link |
of that, and you're still making decisions whether to shoot, to turn left or to turn
link |
right if you're playing a shooter, or every time you start thinking about Skyrim and Elder
link |
Scrolls and walking around in a beautiful nature and swinging a sword.
link |
But it feels like you're making decisions inside that video game, so even though you
link |
don't have direct access in terms of perception to the bits, to the zeros and ones, it still
link |
feels like you're making decisions, and your decisions actually feels like they're being
link |
applied all the way down to the zeros and ones.
link |
So it feels like you have control, even though you don't direct access to reality.
link |
So there is basically a special character in the video game that is being created by
link |
the video game engine, and this character is serving the aesthetics of the video game,
link |
Yes, but I feel like I have control inside the video game, like all those 12 year olds
link |
that kick my ass on the internet.
link |
So when you play the video game, it doesn't really matter that there's zeros and ones,
link |
you don't care about the bits of the bus, you don't care about the nature of the CPU
link |
What you care about are the properties of the game that you're playing, and you hope
link |
that the CPU is good enough.
link |
And a similar thing happens when we interact with physics.
link |
The world that you and me are in is not the physical world.
link |
The world that you and me are in is a dream world.
link |
How close is it to the real world, though?
link |
We know that it's not very close, but we know that the dynamics of the dream world
link |
match the dynamics of the physical world to a certain degree of resolution, but the causal
link |
structure of the dream world is different.
link |
So you see, for instance, waves crashing on your feet, right, but there are no waves
link |
There's only water molecules that have tangents between the molecules that are the result
link |
of electrons in the molecules interacting with each other.
link |
But aren't they very consistent where you're seeing a very crude approximation?
link |
Aren't our dream world very consistent, like to the point of being mapped directly one
link |
to one to the actual physical world as opposed to us being completely tricked?
link |
This is like where you have like Donald Hock.
link |
It's not an illusion.
link |
It's a form of data compression.
link |
It's an attempt to deal with the dynamics of too many parts to count at the level at
link |
which we are entangled with the best model that you can find.
link |
So we can act in that dream world.
link |
And our actions have impact in the real world, in the physical world.
link |
To which we don't have access.
link |
But it's basically like accepting the fact that the software that we live in, the dream
link |
that we live in is generated by something outside of this world that you and me are in.
link |
So is the software deterministic and do we not have any control?
link |
So free will is having a conscious being.
link |
Free will is the monkey being able to steer the elephant.
link |
It's slightly different.
link |
Basically, in the same way as you are modeling the water molecules in the ocean that engulf
link |
your feet when you are walking on the beach as waves and there are no waves, but only
link |
the atoms on more complicated stuff underneath the atoms and so on.
link |
You know that, right?
link |
You would accept, yes, there is a certain abstraction that happens here.
link |
It's a simplification of what happens.
link |
And a simplification that is designed in such a way that your brain can deal with it temporarily
link |
and spatially in terms of resources and tuned for the predictive value.
link |
So you can predict with some accuracy whether your feet are going to get wet or not.
link |
But it's a really good, it's a really good interface and approximation.
link |
It's like E equals m t squared is a good, equations are good approximations for what
link |
they're much better approximation.
link |
So to me waves is a really nice approximation of what's all the complexity that's happening
link |
Basically, it's a machine learning model that is constantly tuned to minimize surprises.
link |
So it basically tries to predict as well as it can what you're going to perceive next.
link |
Are we talking about which is the machine learning, our perception system or the dream
link |
The dream world is the result of the machine learning process of the perception system.
link |
That's doing the compression.
link |
So the model of you as an agent is not a different type of model or it's a different
link |
type but not different as in its model like nature from the model of the ocean.
link |
Some things are oceans, some things are agents.
link |
And one of these agents is using your own control model, the output of your model, the
link |
things that you perceive yourself as doing.
link |
What about the fact that when you're standing with the water on your feet and you're looking
link |
out into the vast open water of the ocean and then there's a beautiful sunset and the
link |
fact that it's beautiful and then maybe you have friends or loved one with you and you
link |
As the dream world?
link |
It's all happening inside of the dream.
link |
You see, the word dream makes it seem like it's not real.
link |
Of course, it's not real.
link |
The physical universe is real but the physical universe is incomprehensible and it doesn't
link |
have any feeling of realness.
link |
The feeling of realness that you experience gets attached to certain representations where
link |
your brain assesses this is the best model of reality that I have.
link |
So the only thing that's real to you is the thing that's happening at the very base of
link |
reality like the…
link |
For something to be real, it needs to be implemented.
link |
So the model that you have of reality is real in as far as it is a model.
link |
It's an appropriate description of the world to say that there are models that are being
link |
experienced but the world that you experience is not necessarily implemented.
link |
There is a difference between a reality, a simulation and a simulacrum.
link |
The reality that we're talking about is something that fully emerges over a causally
link |
closed lowest layer.
link |
The idea of physicalism is that we are in that layer, that basically our world emerges
link |
Every alternative to physicalism is a simulation theory which basically says that we are in
link |
some kind of simulation universe and the real world needs to be an apparent universe of
link |
that where the actual causal structure is.
link |
And when you look at the ocean in your own mind, you are looking at a simulation that
link |
explains what you're going to see next.
link |
So we are living in a simulation.
link |
Yes, but a simulation generated by our own brains.
link |
And this simulation is different from the physical reality because the causal structure
link |
that is being produced, what you are seeing is different from the causal structure of
link |
If not, then you are going to end up in some kind of institution where people will take
link |
care of you because your behavior will be inconsistent.
link |
Your behavior needs to work in such a way that it's interacting with an accurately
link |
predictive model of reality and if your brain is unable to make your model of reality predictive,
link |
you will need help.
link |
So what do you think about Donald Hoffman's argument that it doesn't have to be consistent
link |
the dream world to what he calls like the interface to the actual physical reality
link |
where there could be, I think he makes an evolutionary argument, which is like it could
link |
be an evolutionary advantage to have the dream world drift away from physical reality.
link |
I think that only works if you have tenure.
link |
As long as you're still interacting with the ground tools, your model needs to be somewhat
link |
Well, in some sense, humans have achieved a kind of tenure in the animal kingdom.
link |
At some point, we became too big to fail, so we became postmodernist.
link |
It all makes sense now, the version of reality that we like.
link |
Yeah, but basically, you can do magic.
link |
You can change your assessment of reality, but eventually, reality is going to come bite
link |
you in the ass if it's not predictive.
link |
Do you have a sense of what is that base layer of physical reality?
link |
You have these attempts at the theories of everything, the very, very small of like string
link |
theory or what Stephen Wolfram talks about with the hypergrass, these are these tiny,
link |
tiny, tiny, tiny objects, and then there is more like quantum mechanics that's talking
link |
about objects that are much larger, but still very, very, very tiny.
link |
Do you have a sense of where the tiniest thing is that is like at the lowest level, the turtle
link |
at the very bottom?
link |
Do you have a sense?
link |
I don't think that you can talk about where it is because space is emergent over the activity
link |
So coordinates only exist in relation to the things, other things.
link |
And so you could, in some sense, abstract it into locations that can hold information
link |
and trajectories that the information can take between the different locations.
link |
And this is how we construct our notion of space.
link |
And physicists usually have a notion of space that is continuous.
link |
And this is a point where I tend to agree with people like Stephen Wolfram, who are very
link |
skeptical of the geometric notions.
link |
I think that geometry is the dynamics of too many parts to count.
link |
And when there are no infinities, if there were two infinities, you would be able to
link |
be running into contradictions, which is, in some sense, what Gödel and Turing discovered
link |
in response to Hilbert's call.
link |
So there are no infinities.
link |
There are no infinities.
link |
There is unboundedness.
link |
But if you have a language that talks about infinity, at some point, the language is going
link |
to contradict itself, which means it's no longer valid.
link |
In order to deal with infinities and mathematics, you have to postulate the existence initially.
link |
You cannot construct the infinities.
link |
And that's an issue, right?
link |
You cannot build up an infinity from zero.
link |
But in practice, you never do this, right?
link |
When you perform calculations, you only look at the dynamics of too many parts to count.
link |
And usually, these numbers are not that large.
link |
They're not Googles or something.
link |
The infinities that we are dealing with in our universe are, mathematically speaking,
link |
relatively small integers.
link |
And still, what we're looking at is dynamics where a trillion things behave similar to
link |
a hundred trillion things or something that is very, very large, because they're converging.
link |
And these converging dynamics, these operators, this is what we deal with when we are doing
link |
The geometry is stuff where we can pretend that it's continuous, because if we subdivide
link |
the space sufficiently fine grained, these things approach a certain dynamic.
link |
And this approached dynamic, that is what we mean by it.
link |
But I don't think that infinity would work, so to speak, that you would know the last
link |
digit of pi and that you have a physical process that rests on knowing the last digit of pi.
link |
Yeah, that could be just a peculiar quirk of human cognition that we like discrete.
link |
Discrete makes sense to us.
link |
So in terms of our intuitions.
link |
No, the issue is that everything that we think about needs to be expressed in some
link |
kind of mental language, not necessarily a natural language, but some kind of mathematical
link |
language that your neurons can speak, that refers to something in the world.
link |
And what we have discovered is that we cannot construct a notion of infinity without running
link |
into contradictions, which means that such a language is no longer valid.
link |
And I suspect this is what made Pythagoras so unhappy when somebody came up with the
link |
notion of irrational numbers before it was time, where there's this myth that he had
link |
this person killed when he blabbed out the secret that not everything can be expressed
link |
as a ratio between two numbers, but that there are numbers between the ratios.
link |
The world was not ready for this.
link |
And I think he was right that has confused mathematicians very seriously because these
link |
numbers are not values, they are functions.
link |
And so you can calculate these functions to a certain degree of approximation, but you
link |
cannot pretend that pi has actually a value.
link |
Pi is a function that would approach this value to some degree.
link |
But nothing in the world rests on knowing pi.
link |
How important is this distinction between discrete and continuous for you to get to
link |
Because there's a, I mean, in discussion of your favorite flavor of the theory of everything,
link |
there's a few on the table.
link |
So there's string theory, there's a particular, there's a loop quantum gravity, which focus
link |
on one particular unification.
link |
There's just a bunch of favorite flavors of different people trying to propose a theory
link |
Eric Weinstein and a bunch of people throughout history.
link |
And then of course, Stephen Wolfram, who I think is one of the only people doing a discrete
link |
There's a bunch of physicists who do this right now.
link |
And like Tofoli and Tomasello and digital physics is something that is, I think, growing
link |
But the main reason why this is interesting is because it is important sometimes to settle
link |
I don't think that you need infinities at all and you never needed them.
link |
You can always deal with very large numbers and you can deal with limits, right?
link |
We are fine with doing that.
link |
You don't need any kind of affinity.
link |
You can build your computer algebra systems just as well without believing in infinity
link |
in the first place.
link |
So you're okay with limits?
link |
So basically a limit means that something is behaving pretty much the same if you make
link |
the number larger, right?
link |
Because it's converging to a certain value and at some point the difference becomes
link |
negligible and you can no longer measure it.
link |
And in this sense, you have things that if you have an n gone, which has enough corners,
link |
then it's going to behave like a circle at some point, right?
link |
And it's only going to be in some kind of esoteric thing that cannot exist in a physical
link |
universe that you would be talking about this perfect circle.
link |
And now it turns out that it also wouldn't work in mathematics because you cannot construct
link |
mathematics that has infinite resolution without running into contradictions.
link |
So that is itself not that important because we never did that, right?
link |
It's just a thing that some people thought we could.
link |
And this leads to confusion.
link |
So for instance, Roger Penrose uses this as an argument to say that there are certain
link |
things that mathematicians can do dealing with infinities.
link |
And by extension, our mind can do that computers cannot do.
link |
He talks about that there's the human mind can do certain mathematical things that the
link |
computer has defined by the universal towing machine cannot.
link |
So that it has to do with infinity.
link |
It's one of the things.
link |
So he is basically pointing at the fact that there are things that are possible in the
link |
mathematical mind and in pure mathematics that are not possible in machines that can
link |
be constructed in the physical universe.
link |
And because he's an honest guy, he thinks this means that present physics cannot explain
link |
operations that happen in our mind.
link |
Do you think he's right?
link |
So let's leave his discussion of consciousness aside for the moment.
link |
Do you think he's right about just what he's basically referring to as intelligence?
link |
So is the human mind fundamentally more capable as a thinking machine than a universal towing
link |
But so he's suggesting that, right?
link |
So our mind is actually less than a towing machine.
link |
There can be no towing machine because it's defined as having an infinite tape.
link |
And we always only have a finite tape.
link |
But he's saying it's better.
link |
Our minds can only perform finitely many operations.
link |
But he's saying it can do the kind of computation that the towing machine cannot.
link |
And that's because he thinks that our minds can do operations that have infinite resolution
link |
And I don't think that's the case.
link |
Our minds are just able to discover these limit operators over too many parts to count.
link |
What about his idea that consciousness is more than a computation?
link |
So it's more than something that a towing machine can do.
link |
So again, saying that there's something special about our mind that cannot be replicated in
link |
The issue is that I don't even know how to construct a language to express this statement
link |
Well, the basic statement is there's a human experience that includes intelligence, that
link |
includes self awareness, that includes the hard problem of consciousness.
link |
And the question is, can that be fully simulated in the computer, in the mathematical model
link |
of the computer, as we understand it today?
link |
As your parenthesis, no.
link |
So the universal towing machine cannot simulate the universe.
link |
So the interesting question is, and you have to ask him this, is why not?
link |
What is the specific thing that cannot be modeled?
link |
And when I looked at his writings, and I haven't read all of it, but when I read, for instance,
link |
the section that he writes in the introduction and wrote to infinity, the thing that he specifically
link |
refers to is the way in which human minds deal with infinities.
link |
And that itself can, I think, easily be deconstructed.
link |
A lot of people feel that our experience cannot be explained in a mechanical way.
link |
And therefore, it needs to be different.
link |
And I concur, our experience is not mechanical.
link |
Our experience is simulated.
link |
It exists only in a simulation.
link |
The only simulation can be conscious.
link |
Physical systems cannot be conscious because they're only mechanical.
link |
Cells cannot be conscious.
link |
Neurons cannot be conscious.
link |
Brains cannot be conscious.
link |
People cannot be conscious as far as you, if you understand them as physical systems.
link |
What can be conscious is the story of a system in the world where you write all these things
link |
You have experiences for the same reason that a character and novel have experiences, because
link |
it's written into the story.
link |
And now the system is acting on that story.
link |
And it's not a story that is written in a natural language.
link |
It's written in a perceptual language, in this multimedia language of the game engine.
link |
And in there, you write in what kind of experience you have and what this means for the behavior
link |
of the system, for your behavior tendencies, for your focus, for your attention, for your
link |
experience of valence, and so on.
link |
And this is being used to inform the behavior of the system in the next step.
link |
And then the story updates with the reactions of the system and the changes in the world,
link |
And you live inside of that model.
link |
You don't live inside of the physical reality.
link |
And I mean, just to linger on it, like, you see, okay, it's in the perceptual language,
link |
the multimodal perceptual language, that's the experience.
link |
That's what consciousness is within that model, within that story.
link |
But do you have agency?
link |
When you play a video game, you can turn left and you can turn right in that story.
link |
So in that dream world, how much control do you have?
link |
Is there such a thing as you in that story?
link |
Like is it right to say the main character, you know, everybody's NPCs, and then there's
link |
the main character, and you're controlling the main character?
link |
Or is that an illusion?
link |
Is there a main character that you're controlling?
link |
I'm getting to the point of, like, the free will point.
link |
Imagine that you are building a robot that plays soccer.
link |
And you've been to MIT computer science, you basically know how to do that.
link |
And so you would say the robot is an agent that solves a control problem, how to get
link |
the ball into the goal.
link |
And it needs to perceive the world and the world is disturbing him in trying to do this.
link |
So he has to control many variables to make that happen and to project itself and the
link |
ball into the future and understand its position on the field relative to the ball and so on
link |
in the position of its limbs or in the space around it and so on.
link |
So it needs to have an adequate model that abstracting reality in a useful way.
link |
And you could say that this robot does have agency over what it's doing in some sense.
link |
And the model is going to be a control model.
link |
And inside of that control model, you can possibly get to a point where this thing is
link |
sufficiently abstract to discover its own agency.
link |
Our current robots don't do that.
link |
They don't have a unified model of the universe.
link |
But there's not a reason why we shouldn't be getting there at some point in the not
link |
too distant future.
link |
And once that happens, you will notice that the robot tells a story about a robot playing
link |
So the robot will experience itself playing soccer in a simulation of the world that it
link |
uses to construct a model of the locations of its legs and limbs in space on the field
link |
with relationship to the ball.
link |
And it's not going to be at the level of the molecules.
link |
It will be an abstraction that is exactly at the level that is most suitable for past
link |
planning of the movements of the robot.
link |
It's going to be a high level abstraction, but a very useful one that is as predictive
link |
as we can make it.
link |
And in that side of that story, there is a model of the agency of that system.
link |
So the model can accurately predict that the contents of the model are going to be driving
link |
the behavior of the robot in the immediate future.
link |
But there's the hard problem of consciousness, which I would also, there's a subjective experience
link |
of free will as well, that I'm not sure where the robot gets that, where that little leap
link |
Because for me right now, everything I imagine with that robot as it gets more and more and
link |
more sophisticated, the agency comes from the programmer of the robot still, of what
link |
was programmed in.
link |
You would probably do an end to end learning system.
link |
You maybe need to give it a few priors, so you nudge the architecture in the right direction
link |
that it converges more quickly.
link |
But ultimately, discovering the suitable hyperparameters of the architecture is also only a search
link |
And as the search process was evolution that has informed our brain architecture so we
link |
can converge in a single lifetime on useful interaction with the world and the formation
link |
The problem is if we define hyperparameters broadly, so it's not just the parameters that
link |
control this end to end learning system, but the entirety of the design of the robot.
link |
You have to remove the human completely from the picture.
link |
And then in order to build the robot, you have to create an entire universe because you can't
link |
just shortcut evolution.
link |
You have to go from the very beginning.
link |
In order for it to have, because I feel like there's always a human pulling the strings
link |
and that makes it seem like the robot is cheating.
link |
It's getting a shortcut to consciousness.
link |
And you are looking at the current Boston Dynamics robots.
link |
It doesn't look as if there is somebody pulling the strings.
link |
It doesn't look like cheating anymore.
link |
Okay, so let's go there because I got to talk to you about this.
link |
So obviously with the case of Boston Dynamics, as you may or may not know, it's always either
link |
hard coded or remote controlled.
link |
There's no intelligence.
link |
So I don't know how the current generation of Boston Dynamics robots works, but what
link |
I've been told about the previous ones was that it's basically all cybernetic control,
link |
which means you still have feedback mechanisms and so on.
link |
But it's not deep learning for the most part as it's currently done.
link |
For the most part, just identifying a control hierarchy that is congruent to the limbs that
link |
exist and the parameters that need to be optimized for the movement of these limbs.
link |
And then there is a convergence progress.
link |
So it's basically just regression that you would need to control this.
link |
But again, I don't know whether that's true.
link |
That's just what I've been told about how they work.
link |
We have to separate several levels of discussions here.
link |
So the only thing they do is pretty sophisticated control with no machine learning in order
link |
to maintain balance or to write itself.
link |
It's a control problem in terms of using the actuators to when it's pushed or when it steps
link |
on a thing that's uneven, how to always maintain balance.
link |
And there's a tricky heuristics around that, but that's the only goal.
link |
Everything you see Boston Dynamics doing in terms of that to us humans is compelling,
link |
which is any kind of higher order movement like turning, wiggling its butt, jumping back
link |
on its two feet, dancing, dancing is even worse because dancing is hard coded in.
link |
It's it's choreographed by humans.
link |
There's choreography software, like there is no of all that high level movement.
link |
There's no anything that you can call certainly can't call AI, but there's no even like basic
link |
It's all hard coded in.
link |
And yet we humans immediately project agency onto them, which is this is fast.
link |
So the gap here is it doesn't necessarily have agency.
link |
What it has is cybernetic control.
link |
And the cybernetic control means you have a hierarchy of feedback loops that keep the
link |
behavior in certain boundaries so the robot doesn't fall over and it's able to perform
link |
the movements and the choreography cannot really happen with motion capture because the robot
link |
would fall over because the physics of the robot, the weight distribution and so on is
link |
different from the weight distribution in the human body.
link |
So if you were using the directly motion captured movements of a human body to project
link |
it into this robot, it wouldn't work.
link |
You can do this with a computer animation that will look a little bit off, but who cares?
link |
But if you want to correct for the physics, you need to basically tell the robot where
link |
it should move its limbs.
link |
And then the control algorithm is going to approximate a solution that makes it possible
link |
visit the physics of the robot.
link |
And you have to find the basic solution for making that happen.
link |
And there's probably going to be some regression necessary to get the control architecture to
link |
make these movements.
link |
But those two layers are separate.
link |
The thing, the higher level instruction of how you should move and where you should
link |
move is a high level.
link |
I expect that the control level of these robots at some level is dumb.
link |
Because it's just the physical control movement, the motor architecture.
link |
But it's a relatively smart motor architecture.
link |
It's just that there is no high level deliberation about what decisions to make necessarily.
link |
But it doesn't feel like free will or consciousness.
link |
That was not where I was trying to get to.
link |
I think that in our own body, we have that too.
link |
So we have a certain thing that is basically just a cybernetic control architecture that
link |
is moving our limbs.
link |
And deep learning can help in discovering such an architecture if you don't have it
link |
in the first place.
link |
If you already know your hardware, you can maybe handcraft it.
link |
But if you don't know your hardware, you can search for such an architecture.
link |
And this work already existed in the 80s and 90s.
link |
People were starting to search for control architectures by motor babbling and so on
link |
and just use reinforcement learning architectures to discover such a thing.
link |
And now imagine that you have the cybernetic control architecture already inside of you.
link |
And you extend this a little bit.
link |
So you are seeking output, for instance, or rest or and so on.
link |
And you get to have a baby at some point.
link |
And now you add more and more control layers to this.
link |
And the system is reverse engineering its own control architecture and builds a high
link |
level model to synchronize the pursuit of very different conflicting goals.
link |
And this is how I think you get to purposes.
link |
Purposes are models of your goals.
link |
The goals might be intrinsic as the result of the different set point violations that
link |
you have, hunger and thirst for very different things and rest and pain avoidance and so
link |
And you put all these things together and eventually you need to come up with a strategy
link |
to synchronize them all.
link |
And you don't need just to do this alone by yourself because you are state building organisms.
link |
We cannot function as an isolation the way that homo sapiens is set up.
link |
So our own behavior only makes sense when you zoom out very far into a society or even
link |
into ecosystemic intelligence on the planet and our place in it.
link |
So the individual behavior only makes sense in these larger contexts.
link |
And we have a number of priors built into us.
link |
So we are behaving as if we were acting on these high level goals pretty much right from
link |
And eventually in the course of our life, we can reverse engineer the goals that we
link |
are acting on what actually are our higher level purposes.
link |
And the more we understand that, the more our behavior makes sense.
link |
But this is all at this point, complex stories within stories that are driving our behavior.
link |
Yeah, I just don't know how big of a leap it is to start creating a system that's able
link |
to tell stories within stories, like how big of a leap that is from where currently Boston
link |
Dynamics is or any robot that's operating in the physical space.
link |
And that leap might be big if it requires to solve the hard problem of consciousness,
link |
which is telling a hell of a good story.
link |
I suspect that consciousness itself is relatively simple.
link |
What's hard is perception and the interface between perception and reasoning.
link |
There's, for instance, the idea of the consciousness prior that would be built into such a system
link |
by Yoshua Bengio, and what he describes, and I think that's accurate, is that our own model
link |
of the world can be described through something like an energy function.
link |
The energy function is modeling the contradictions that exist within the model at any given point.
link |
And you try to minimize these contradictions, the tangents in the model.
link |
And to do this, you need to sometimes test things.
link |
You need to conditionally disambiguate figure and ground.
link |
You need to distinguish whether this is true or that is true.
link |
Eventually, you get to an interpretation, but you will need to manually depress a few
link |
points in your model to let it snap into a state that makes sense.
link |
And this function that tries to get the biggest dip in the energy function in your model according
link |
to Yoshua Bengio is related to consciousness.
link |
It's a low dimensional discrete function that tries to maximize this dip in the energy
link |
Yeah, I think I would need to dig into details because I think the way he uses the word consciousness
link |
is more akin to self awareness, like modeling yourself within the world as opposed to the
link |
subjective experience, the hard problem.
link |
No, it's not even the self within the world.
link |
The self is the agent and you don't need to be aware of yourself in order to be conscious.
link |
The self is just a particular content that you can have, but you don't have to have.
link |
But you can be conscious in, for instance, a dream at night or during a meditation state
link |
where you don't have a self, where you're just aware of the fact that you are aware.
link |
And what we mean by consciousness and colloquial sense is largely this reflexive self awareness
link |
that we become aware of the fact that you're paying attention, that we are the thing that
link |
We are the thing that pays attention.
link |
I don't see the awareness that we're aware.
link |
The hard problem doesn't feel like it's solved.
link |
It's called a hard problem for a reason because it seems like there needs to be a major leap.
link |
Yeah, I think the major leap is to understand how it is possible that a machine can dream,
link |
that a physical system is able to create a representation that the physical system is
link |
acting on and that is spun force and so on.
link |
But once you accept the fact that you are not in physics, but that you exist inside
link |
of the story, I think the mystery disappears.
link |
Everything is possible in a story.
link |
Exist inside the story.
link |
Your consciousness is being written into the story.
link |
The fact that you experience things is written to the story.
link |
You ask yourself, is this real what I'm seeing?
link |
And your brain writes into the story, yes, it's real.
link |
So what about the perception of consciousness?
link |
So to me, you look conscious.
link |
So the illusion of consciousness, the demonstration of consciousness, I ask for the legged robot,
link |
how do we make this legged robot conscious?
link |
So there's two things that maybe you can tell me if they're neighboring ideas.
link |
One is actually making conscious and the other is make it appear conscious to others.
link |
Are those related?
link |
Let's ask it from the other direction.
link |
What would it take to make you not conscious?
link |
So when you are thinking about how you perceive the world, can you decide to switch from looking
link |
at qualia to looking at representational states?
link |
And it turns out you can.
link |
There is a particular way in which you can look at the world and recognize its machine
link |
nature, including your own.
link |
And in that state, you don't have that conscious experience in this way anymore.
link |
It becomes apparent as a representation.
link |
Everything becomes opaque.
link |
And I think this thing that you recognize everything as a representation, this is typically
link |
what we mean with enlightenment states.
link |
And it can happen on the motivational level, but you can also do this on the experiential
link |
level and the perceptual level.
link |
See, but then I can come back to a conscious state.
link |
Okay, I particularly, I'm referring to the social aspect that the demonstration of consciousness
link |
is a really nice thing at a party when you're trying to meet a new person.
link |
It's a nice thing to know that they're conscious and they can, I don't know how fundamental
link |
consciousness is in human interaction, but it seems like to be at least an important
link |
And I asked that in the same kind of way for robots, you know, in order to create a rich,
link |
compelling human robot interaction, it feels like there needs to be elements of consciousness
link |
within that interaction.
link |
My cat is obviously conscious.
link |
And so my cat can do this party trick.
link |
She also knows that I am conscious.
link |
We're able to have feedback about the fact that we are both acting on models of our own
link |
And the question is, how hard is it for the robot, artificially created robot to achieve
link |
cat level and party tricks?
link |
So the issue for me is currently not so much on how to build a system that creates a story
link |
about a robot that lives in the world, but to make an adequate representation of the
link |
And the model that you and me have is a unified one.
link |
It's one where you basically make sense of everything that you can perceive every feature
link |
in the world that enters your perception can be relationally mapped to a unified model
link |
And we don't have an AI that is able to construct such a unified model yet.
link |
So you need that unified model to do the party trick?
link |
I think that it doesn't make sense if this thing is conscious, but not in the same universe
link |
as you, because you could not relate to each other.
link |
So what's the process, would you say, of engineering consciousness in the machine?
link |
What are the ideas here?
link |
So you probably want to have some kind of perceptual system.
link |
This perceptual system is a processing agent that is able to track sensory data and predict
link |
the next frame in the sensory data from the previous frames of the sensory data and the
link |
current state of the system.
link |
So the current state of the system is in perception instrumental to predicting what happens next.
link |
And this means you build lots and lots of functions that take all the blips that you
link |
feel on your skin and that you see on your retina or that you hear and puts them into
link |
a set of relationships that allows you to predict what kind of sensory data, what kind
link |
of sensor of blips, your vector of blips you're going to perceive in the next frame.
link |
This is tuned and it's constantly tuned until it gets as accurate as it can.
link |
You build a very accurate prediction mechanism that is step one of the perception.
link |
So first you predict, then you perceive and see the error in your prediction.
link |
And you have to do two things to make that happen.
link |
One is you have to build a network of relationships that are constraints, that take all the variants
link |
in the world to put each of the variances into a variable that is connected with relationships
link |
to other variables.
link |
And these relationships are computable functions that constrain each other.
link |
So when you see a nose that points a certain direction in space, you have a constraint
link |
that says there should be a phase nearby that has the same direction.
link |
And if that is not the case, you have some kind of contradiction that you need to resolve
link |
because it's probably not a nose, what you're looking at, it just looks like one.
link |
So you have to reinterpret the data until you get to a point where your model converges.
link |
And this process of making the sensory data fit into your model structure is what PRG
link |
calls the assimilation.
link |
And accommodation is the change of the models, where you change your model such a way that
link |
you can assimilate everything.
link |
So you're talking about building a hell of an awesome perception system that's able
link |
to do prediction and perception and correct and improving.
link |
Wait, there's more.
link |
Yes, there's more.
link |
So the first thing that we wanted to do is minimize the contradictions in the model.
link |
And of course, it's very easy to make a model in which you minimize the contradictions just
link |
by allowing that it can be in many, many possible states.
link |
So if you increase degrees of freedom, you will have fewer contradictions.
link |
But you also want to reduce the degrees of freedom because degrees of freedom mean uncertainty.
link |
You want your model to reduce uncertainty as much as possible.
link |
But reducing uncertainty is expensive.
link |
So you have to have a tradeoff between minimizing contradictions and reducing uncertainty.
link |
And you have only a finite amount of compute and experimental time and effort available
link |
to reduce uncertainty in the world.
link |
So you need to assign value to what you observe.
link |
So you need some kind of motivational system that is estimating what you should be looking
link |
at and what you should be thinking about it, how you should be applying your resources
link |
to model what that is.
link |
So you need to have something like convergence links that tell you how to get from the present
link |
state of the model to the next one.
link |
You need to have these compatibility links that tell you which constraints exist and
link |
which constraint violations exist.
link |
And you need to have some kind of motivational system that tells you what to pay attention
link |
So now we have a second agent next to the perceptual agent.
link |
We have a motivational agent.
link |
This is a cybernetic system that is modeling what the system needs, what's important for
link |
the system, and that interacts with the perceptual system to maximize the expected reward.
link |
And you're saying a motivational system is some kind of, like, what is it, a high level
link |
narrative over some lower level?
link |
No, it's just your brainstem stuff, the limbic system stuff that tells you, OK, now you
link |
should get something to eat because I've just measured your geo blood sugar.
link |
So you mean the motivational system, like the lower level stuff, like hungry?
link |
So there's basically physiological needs and some cognitive needs and some social needs
link |
and they all interact.
link |
And they all implemented different parts in your nervous system as the motivational system.
link |
But they're basically cybernetic feedback loops.
link |
It's not that complicated, it's just a lot of code.
link |
And so you now have a motivational agent that makes your robot go for the ball or that makes
link |
your worm go to eat food and so on.
link |
And you have the perceptual system that lets it predicts the environment so it's able
link |
to solve that control problem to some degree.
link |
And now what we learned is that it's very hard to build a machine learning system that
link |
looks at all the data simultaneously to see what kind of relationships could exist between
link |
So you need to selectively model the world.
link |
You need to figure out, where can I make the biggest difference if I would put the following
link |
Sometimes you find a gradient for that, right?
link |
When you have a gradient, you don't need to remember where you came from.
link |
You just follow the gradient until it doesn't get any better.
link |
But if you have a world where the problems are discontinuous and research spaces are
link |
discontinuous, you need to retain memory of what you explored.
link |
You need to construct a plan of what to explore next.
link |
And this thing that means that you have next to this perceptual construction system and
link |
the motivational cybernetics, an agent that is paying attention to what it should select
link |
at any given moment to maximize reward.
link |
And this scanning system, this attention agent is required for consciousness.
link |
And consciousness is its control model.
link |
So it's the indexed memories that this thing retains when it manipulates the perceptual
link |
representations to maximize the value and minimize the conflicts and to increase coherence.
link |
So the purpose of consciousness is to create coherence in your perceptual representations,
link |
remove conflicts, predict the future, construct counterfactual representations so you can coordinate
link |
your actions and so on.
link |
And in order to do this, it needs to form memories.
link |
These memories are partial binding states of the working memory contents that are being
link |
revisited later on to backtrack, to undo certain states, to look for alternatives.
link |
And these indexed memories that you can recall, that is what you perceive as your stream of
link |
And being able to recall these memories, this is what makes you conscious.
link |
If you could not remember what you paid attention to, you wouldn't be conscious.
link |
So consciousness is the index in the memory database, okay.
link |
But let me sneak up to the questions of consciousness a little further.
link |
So we usually relate suffering to consciousness.
link |
So the capacity to suffer, I think to me that's a really strong sign of consciousness, is
link |
a thing that can suffer.
link |
How is that useful?
link |
And, like, in your model, what you just described, which is indexing of memories, and what is
link |
the coherence with the perception, with this predictive thing that's going on in the perception,
link |
how does suffering relate to any of that, you know, the higher level of suffering that
link |
Basically, pain is a reinforcement signal.
link |
And pain is a signal that one part of your brain sends to another part of your brain,
link |
or an abstract sense, part of your mind sends to another part of the mind to regulate its
link |
behavior, to tell it the behavior that you're currently exhibiting should be improved.
link |
And this is a signal that I tell you to move away from what you're currently doing and
link |
push into a different direction.
link |
So pain gives you a part of you an impulse to do something differently.
link |
But sometimes this doesn't work, because the training part of your brain is talking to
link |
the wrong region, or because it has the wrong model of the relationships in the world.
link |
Maybe you're mismodeling yourself, or you're mismodeling the relationship of yourself to
link |
the world, or you're mismodeling the dynamics of the world.
link |
So you're trying to improve something that cannot be improved by generating more pain.
link |
But the system doesn't have any alternative.
link |
So it doesn't get better.
link |
What do you do if something doesn't get better, and you want it to get better?
link |
You increase the strength of the signal, and when the signal becomes chronic, when it becomes
link |
permanent, without a change inside, this is what we call suffering.
link |
And the purpose of consciousness is to deal with contradictions, with things that cannot
link |
The purpose of consciousness, I think, is similar to a conductor in an orchestra.
link |
When everything works well, the orchestra doesn't need much of a conductor, as long
link |
But when there is a lack of coherence, or something is consistently producing disharmony
link |
and mismatches, then the conductor becomes alert and interacts with it.
link |
So suffering attracts the activity of our consciousness.
link |
And the purpose of that is ideally that we bring new layers online, new layers of modeling
link |
that are able to create a model of the dysregulation so we can deal with it.
link |
And this means that we typically get higher level consciousness, so to speak.
link |
We get some consciousness above our pay grade, maybe, if we have some suffering early in
link |
There are really interesting people at trauma early on in their childhood.
link |
And trauma means that you are suffering an injury for which the system is not prepared,
link |
which it cannot deal with, which it cannot insulate itself from, so something breaks.
link |
And this means that the behavior of the system is permanently disturbed in a way that some
link |
mismatch exists now in a regulation, that just by following your impulses, by following
link |
the pain in the direction which it hurts, the situation doesn't improve, but get worse.
link |
And so what needs to happen is that you grow up, and that part that is grown up is able
link |
to deal with the part that is stuck in this earlier phase.
link |
Yeah, so it leads to growth, adding extra layers to your cognition.
link |
Let me ask you then, because I got a stick on suffering, the ethics of the whole thing.
link |
So not our consciousness, but the consciousness of others, you've tweeted, one of my biggest
link |
fears is that insects could be conscious.
link |
The amount of suffering on earth would be unthinkable.
link |
So when we think of other conscious beings, is suffering a property of consciousness that
link |
we're most concerned about?
link |
So I'm still thinking about robots, how to make sense of other nonhuman things that appear
link |
to have the depth of experience that humans have.
link |
And to me, that means consciousness and the darkest side of that, which is suffering,
link |
the capacity to suffer.
link |
So I start thinking, how much responsibility do we have for those other conscious beings?
link |
That's where the definition of consciousness becomes most urgent.
link |
Like having to come up with the definition of consciousness becomes most urgent, is who
link |
should we, and should we not be torturing?
link |
There's no general answer to this.
link |
Was jingles kind of doing anything wrong?
link |
It depends on how you look at it.
link |
Well, he drew a line somewhere where this is us and that's them.
link |
It's the circle of empathy.
link |
It's like, you don't have to use the word consciousness, but these are the things that
link |
matter to me if they suffer or not, and these are the things that don't matter to me.
link |
But when one of his commanders failed him, he broke his spine and let him die in a horrible
link |
And so in some sense, I think he was indifferent to suffering, or he was not indifferent in
link |
the sense that he didn't see it as useful if he inflicted suffering.
link |
But he did not see it as something that had to be avoided.
link |
That was not the goal.
link |
The question was, how can I use suffering and the infliction of suffering to reach my goals
link |
from his perspective?
link |
I see so like different societies throughout history put different value on the different
link |
individuals, different psychies, but also even the objective of avoiding suffering.
link |
Some societies probably, I mean, this is where religious belief really helps, that afterlife,
link |
that doesn't matter that you suffer or die, what matters is you suffer honorably so that
link |
you enter the afterlife.
link |
It seems to be superstitious to me, basically beliefs that assert things for which no evidence
link |
exist are incompatible with sound epistemology.
link |
And I don't think that religion has to be superstitious, otherwise it should be condemned
link |
Okay, actually, you're somebody who's saying, we live in a dream world, we have zero evidence
link |
It's not the case.
link |
They are limits to what languages can be constructed, mathematics breaks solid evidence
link |
for its own structure.
link |
And once we have some idea of what languages exist and how a system can learn and what
link |
learning itself is in the first place, and so on, we can begin to realize that our intuitions
link |
that we are able to learn about the regularities of the world and minimize surprise and understand
link |
the nature of our own agency to some degree of abstraction.
link |
That's not an illusion.
link |
So just because we live in a dream world doesn't mean mathematics can't give us
link |
a consistent glimpse of physical, of objective reality.
link |
We can basically distinguish useful encodings from useless encodings.
link |
And when we apply our truth seeking to the world, we know we usually cannot find out
link |
whether a certain thing is true.
link |
What we typically do is we take the state vector of the universe separated into separate
link |
objects that interact with each other through interfaces.
link |
And this distinction that we are making is not completely arbitrary.
link |
It's done to optimize the compression that we can apply to our models of the universe.
link |
So we can predict what's happening with our limited resources.
link |
In this sense, it's not arbitrary.
link |
But the separation of the world into objects that are somehow discrete and interacting
link |
with each other is not the true reality.
link |
The boundaries between the objects are projected into the world, not arbitrarily projected,
link |
but still it's only an approximation of what's actually the case.
link |
And we sometimes notice that we run into contradictions when we try to understand high level things
link |
like economic aspects of the world and so on or political aspects or psychological aspects
link |
where we make simplifications and the objects that we are using to separate the world are
link |
just one of many possible projections of what's going on.
link |
And so it's not in this postmodernist sense completely arbitrary and you're free to pick
link |
what you want or dismiss what you don't like because it's all stories.
link |
No, that's not true.
link |
You have to show for every model of how well it predicts the world.
link |
So the confidence that you should have in the entities of your models should correspond
link |
to the evidence that you have.
link |
Can I ask you in a small tangent to talk about your favorite set of ideas and people, which
link |
What is postmodernism?
link |
How would you define it?
link |
And why, to you, is it not a useful framework of thought?
link |
Postmodernism is something that I'm really not an expert on.
link |
And postmodernism is a set of philosophical ideas that is difficult to lump together,
link |
that is characterized by some useful thinkers, some of them poststructuralists and so on.
link |
And I'm mostly not interested in it because I think that it's not leading me anywhere
link |
that I find particularly useful.
link |
It's mostly, I think, born out of the insight that the ontologies that we impose on the
link |
world are not literally true and that we can often get to a different interpretation by
link |
the world by using a different ontology that is different separation of the world into
link |
interacting objects.
link |
But the idea that this makes the world a set of stories that are arbitrary, I think, is
link |
wrong, and the people that are engaging in this type of philosophy are working in an
link |
area that I largely don't find productive.
link |
There is nothing useful coming out of this.
link |
So this idea that truth is relative is not something that has, in some sense, informed
link |
physics or theory of relativity, and there is no feedback between those.
link |
There is no meaningful influence of this type of philosophy on the sciences or on engineering
link |
But there is a very strong information of this on ideology because it basically has become
link |
an ideology that is justifying itself by the notion that truth is a relative concept.
link |
And it's not being used in such a way that the philosophers or sociologists that take
link |
up these ideas say, oh, I should doubt my own ideas because maybe my separation of the
link |
world into objects is not completely valid and I should maybe use the different one
link |
and be open to pluralism of ideas.
link |
But it mostly exists to dismiss the ideas of other people.
link |
It becomes a political weapon of sorts to achieve power.
link |
Basically there's nothing wrong, I think, with developing a philosophy around this.
link |
But to develop norms around the idea that truth is something that is completely negotiable
link |
is incompatible with the scientific project.
link |
And I think if the academia has no defense against the ideological parts of the postmodernist
link |
movement, it's doomed.
link |
You have to acknowledge the ideological part of any movement, actually, including postmodernism.
link |
Well, the question is what an ideology is.
link |
And to me, an ideology is basically a viral memeplex that is changing your mind in such
link |
a way that reality gets warped.
link |
It gets warped in such a way that you're being cut off from the rest of human thought space.
link |
And you cannot consider things outside of the range of ideas of your own ideology as
link |
So I mean, there's certain properties to an ideology that make it harmful.
link |
One of them is that dogmatism of just certainty, dogged certainty in that you're right.
link |
You have the truth and nobody else does.
link |
But what is creating this certainty?
link |
It's very interesting to look at the type of model that is being produced.
link |
Is it basically just a strong prior?
link |
And you tell people, oh, this idea that you consider to be very true, the evidence for
link |
this is actually just much weaker than you thought.
link |
And look here at some studies, no, this is not how it works.
link |
It's usually normative, which means some thoughts are unthinkable because they would
link |
change your identity into something that is no longer acceptable.
link |
And this cuts you off from considering an alternative.
link |
And many de facto religions use this trick to lock people into a certain mode of thought.
link |
And this removes agency over your own thoughts.
link |
And it's very ugly to me.
link |
It's basically not just a process of domestication, but it's actually an intellectual castration
link |
It's an inability to think creatively and to bring forth new thoughts.
link |
And ask you about substances, chemical substances that affect the video game, the dream world.
link |
So psychedelics, that increasing have been getting a lot of research done on them.
link |
So in general, psychedelics, psilocybin, MDMA, but also a really interesting one, the big
link |
one, which is DMT.
link |
What and where are the places that these substances take the mind that is operating in the dream
link |
Do you have an interesting sense how this throws a wrinkle into the prediction model?
link |
Is it just some weird little quirk?
link |
Or is there some fundamental expansion of the mind going on?
link |
I suspect that a way to look at psychedelics is that they induce particular types of lucid
link |
So it's a state in which certain connections are being severed in your mind and no longer
link |
Your mind basically gets free to move in a certain direction because some inhibition,
link |
some particular inhibition doesn't work anymore.
link |
And as a result, you might stop having yourself or you might stop perceiving the world as three
link |
And you can explore that state.
link |
And I suppose that for every state that can be induced with psychedelics, there are people
link |
that are naturally in that state.
link |
So sometimes psychedelics that shift you through a range of possible mental states.
link |
And they can also shift you out of the range of permissible mental states that is where
link |
you can make predictive models of reality.
link |
And what I observe in people that use psychedelics a lot is that they tend to be overfitting.
link |
Fitting means that you are using more bits for modeling the dynamics of a function than
link |
And so you can fit your curve to extremely detailed things in the past, but this model
link |
is no longer predictive for the future.
link |
What is it about psychedelics that forces that?
link |
I thought it would be the opposite.
link |
I thought it's a good mechanism for generalization, for regularization.
link |
So it feels like psychedelics expansion of the mind, like forcing your model to be nonpredictive
link |
Meaning like it's almost like, okay, what I would say is psychedelics are akin to traveling
link |
to a totally different environment, like going, if you've never been to like India or something
link |
like that from the United States, very different set of people, different culture, different
link |
food, different roads and values and all those kinds of things.
link |
So psychedelics can, for instance, teleport people into a universe that is hyperbolic,
link |
which means that if you imagine a room that you are in, you can turn around 360 degrees
link |
and you didn't go full circle.
link |
You need to go 20 degrees to full circle.
link |
So the things that people learn in that state cannot be easily transferred in this universe
link |
It could be that if they're able to abstract and understand what happened to them, that
link |
they understand that some part of their spatial cognition has been desynchronized and has
link |
found a different synchronization.
link |
And this different synchronization happens to be a hyperbolic one, right?
link |
So you learn something interesting about your brain.
link |
It's difficult to understand what exactly happened, but we get pretty good idea once
link |
we understand how the brain is representing geometry.
link |
Yeah, but it doesn't give you a fresh perspective on the physical reality.
link |
Who's making that sound?
link |
Is it inside my head or is it external?
link |
Well, there is no sound outside of your mind, but it's making sense of phenomenon physics.
link |
Yeah, in the physical reality, there's sound waves traveling through air.
link |
So that's our model of what happened.
link |
That's our model of what happened.
link |
Don't psychedelics give you a fresh perspective on this physical reality, not this physical
link |
reality, but this more...
link |
What do you call the dream world that's mapped directly to...
link |
The purpose of dreaming at night, I think, is data augmentation.
link |
So you basically...
link |
That's very similar to psychedelics.
link |
...change parameters about the things that you have learned.
link |
And for instance, when you are young, you have seen things from certain perspectives,
link |
but not from others.
link |
So your brain is generating new perspectives of objects that you already know, which means
link |
they can learn to recognize them later from different perspectives, and I suspect that's
link |
the reason why many of us remember to have flying dreams as children, because it's just
link |
different perspectives of the world that we already know, and that it starts to generate
link |
these different perspective changes, and then it fluidly turns this into a flying dream
link |
to make sense of what's happening.
link |
So you fill in the gaps, and suddenly you see yourself flying.
link |
And similar things can happen with semantic relationships.
link |
So it's not just spatial relationships, but it can also be the relationships between
link |
ideas that are being changed.
link |
And it seems that the mechanisms that make that happen during dreaming are interacting
link |
with these same receptors that are being simulated by psychedelics.
link |
So I suspect that there is a thing that I haven't read really about, the way in which
link |
dreams are induced in the brain is not just that the activity of the brain gets tuned
link |
down because your eyes are closed and you no longer get enough data from your eyes, but
link |
there is a particular type of neurotransmitter that is saturating your brain during these
link |
phases, during the RM phases, and you produce controlled hallucinations.
link |
And psychedelics are linking into these mechanisms, I suspect.
link |
So isn't that another trickier form of data augmentation?
link |
But it's also data augmentation that can happen outside of the specification that your brain
link |
So basically people are overclocking their brains, and that produces states that are
link |
subjectively extremely interesting.
link |
But from the outside, very suspicious.
link |
I think I'm overapplying the metaphor of a neural network in my own mind, which I just
link |
think that doesn't lead to overfitting.
link |
But you were just sort of anecdotally saying my experiences with people that have no psychedelics
link |
are that kind of quality.
link |
I think it typically happens.
link |
So if you look at people like Timothy Leary, and he has written beautiful manifestos about
link |
the effect of LSD on people, he genuinely believed, he writes in these manifestos that
link |
in the future, science and art will only be done on psychedelics because it's so much
link |
more efficient and so much better.
link |
And he gave LSD to children in this community of a few thousand people that he had near San
link |
And basically he was losing touch with reality.
link |
He did not understand the effects that the things that he was doing would have on the
link |
reception of psychedelics by society, because he was unable to think critically about what
link |
What happened was that he got in a euphoric state, that euphoric state happened because
link |
he was overfitting.
link |
He was taking this sense of euphoria and translating it into a model of actual success in the
link |
He was feeling better, limitations had disappeared, that he experienced to be existing, but he
link |
didn't get superpowers.
link |
I understand what you mean by overfitting now.
link |
There's a lot of interpretation to the term overfitting in this case.
link |
So he was getting positive rewards from a lot of actions that he shouldn't have put
link |
Yeah, but not just this.
link |
So if you take, for instance, John Lilly, who was studying dolphin languages and aliens
link |
and so on, a lot of people that used psychedelics became very loopy.
link |
And the typical thing that you notice when people are on psychedelics is that they are
link |
in a state where they feel that everything can be explained now.
link |
Everything is clear.
link |
Everything is obvious.
link |
And sometimes they have indeed discovered a useful connection, but not always.
link |
Very often these connections are over interpretations.
link |
I wonder, there's a question of correlation versus causation.
link |
And also I wonder if it's the psychedelics or if it's more the social, like being the
link |
outsider and having a strong community of outsiders and having a leadership position
link |
in an outsider cult like community, that could have a much stronger effect of overfitting
link |
than do psychedelics themselves, the actual substances.
link |
Because it's a counterculture thing.
link |
So it could be that as opposed to the actual substance.
link |
If you're a boring person who wears a suit and tie and works at a bank and takes psychedelics,
link |
that could be a very different effect of psychedelics on your mind.
link |
I'm just sort of raising the point that the people you referenced are already weirdos.
link |
I'm not sure exactly.
link |
No, not necessarily.
link |
A lot of the people that tell me that they used psychedelics in a useful way started
link |
out as squares and were liberating themselves because they were stuck.
link |
They were basically stuck in local optimum of their own self model, of their relationship
link |
to the world, and suddenly they had data augmentation.
link |
They basically saw an experience, a space of possibilities.
link |
They experienced what it would be like to be another person.
link |
And they took important lessons from that experience back home.
link |
I mean, I love the metaphor of data augmentation because that's been the primary driver of
link |
self supervised learning in the computer vision domain is data augmentation.
link |
So it's funny to think of like chemically induced data augmentation in the human mind.
link |
There's also a very interesting effect that I noticed.
link |
I know several people who are speared to me that LSD has cured their migraines.
link |
So severe cluster headaches or migraines that didn't respond to standard medication that
link |
disappeared after a single dose.
link |
And I don't recommend anybody doing this, especially not in the US where it's illegal.
link |
And there are no studies on this for that reason.
link |
But it seems that anecdotally that it basically can reset the serotonergic system.
link |
So it's basically pushing them outside of their normal boundaries.
link |
And as a result, it needs to find a new equilibrium and some people that equilibrium is better.
link |
But it also follows that in other people, it might be worse.
link |
So if you have a brain that is already teetering on the boundary to psychosis, it can be permanently
link |
pushed over that boundary.
link |
Well, that's why you have to do good science, which they're starting to do on all these different
link |
substances of how well it actually works for the different conditions like MDMA seems
link |
to help with PTSD.
link |
Same with psilocybin that, you know, you need to do good science, meaning large studies
link |
Based on the existing studies with MDMA, it seems that if you look at Rick Doblin's
link |
work and what he has published about this and talks about MDMA seems to be a psychologically
link |
relatively safe drug, but it's physiologically not very safe.
link |
That is, there is neurotoxicity if you would use a too large dose.
link |
And if you combine this with alcohol, which a lot of kids do in party settings during
link |
raves and so on, it's very hippotoxic, so basically you can kill your liver.
link |
And this means that it's probably something that is best and most productively used in
link |
clinical setting by people who really know what they're doing.
link |
And I suspect that's also true for the other psychedelics that is, while the other psychedelics
link |
are probably not as toxic as say alcohol, the effects on the psyche can be much more
link |
profound and lasting.
link |
Well, as far as I know, psilocybin, so mushrooms, magic mushrooms, as far as I know in terms
link |
of the studies they're running, I think have no, like they're allowed to do what they're
link |
calling heroic doses.
link |
So that one does not have a toxicity, so they could do like huge doses in a clinical setting
link |
when they're doing study on psilocybin, which is kind of fun.
link |
It seems that most of the psychedelics work in extremely small doses, which means that
link |
the effect on the rest of the body is relatively low, and MDMA is probably the exception.
link |
Maybe ketamine can be dangerous in larger doses because it can depress breathing and
link |
But the LSD and psilocybin work in very, very small doses, at least the active part of them
link |
of psilocybin LSD is only the active part.
link |
But the effect that it can have on your mental wiring can be very dangerous, I think.
link |
Let's talk about AI a little bit.
link |
What are your thoughts about GPT3 and language models trained with cell supervised learning?
link |
It came out quite a bit ago, but I wanted to get your thoughts on it.
link |
In the 90s, I was in New Zealand and I had an amazing professor, Ian Witten, who realized
link |
I was bored in class and put me in his lab, and he gave me the task to discover grammatical
link |
structure in an unknown language.
link |
The unknown language that I picked was English because it was the easiest one to find a corpus
link |
four construct one.
link |
He gave me the largest computer at the whole university.
link |
It had two gigabytes of RAM, which was amazing.
link |
I wrote everything in C with some in memory compression to do statistics over the language.
link |
Next would create a dictionary of all the words, which basically tokenizes everything
link |
and compresses things so they don't need to store the whole word, but just a code for
link |
Then I was taking this all apart in sentences and I was trying to find all the relationships
link |
between all the words and the sentences and do statistics over them.
link |
That proved to be impossible because the complexity is just too large.
link |
If you want to discover the relationship between an article and a noun, and there are three
link |
adjectives in between, you cannot do ngram statistics and look at all the possibilities
link |
that can exist, at least not with the resources that we had back then.
link |
I realized I need to make some statistics over what I need to make statistics over.
link |
I wrote something that was pretty much a hack that did this for at least first order relationships.
link |
I came up with some kind of mutual information graph that was indeed discovering something
link |
that looks exactly like the grammatical structure of the sentence.
link |
Just by trying to encode the sentence in such a way that the words would be written in the
link |
optimal order inside of the model.
link |
What I also found is that if we would be able to increase the resolution of that and not
link |
just use this model to reproduce grammatically correct sentences, we would also be able to
link |
correct stylistically correct sentences by just having more bits in these relationships.
link |
If we wanted to have meaning, we would have to go much higher order.
link |
I didn't know how to make higher order models back then without spending way more years
link |
in research on how to make the statistics over what we need to make statistics over.
link |
This thing that we cannot look at the relationships between all the bits in your input is being
link |
solved in different domains in different ways.
link |
In computer graphics, the computer vision standard method for many years now is convolutional
link |
convolutional neural networks are hierarchies of filters that exploit the fact that neighboring
link |
pixels in images are usually semantically related and distance pixels in images are usually
link |
not semantically related.
link |
You can just by grouping the pixels that are next to each other hierarchically together
link |
reconstruct the shape of objects.
link |
This is an important prior that we build into these models so they can converge quickly.
link |
This doesn't work in language for the reason that adjacent words are often but not always
link |
related and distant words are sometimes related while the words in between are not.
link |
How can you learn the topology of language?
link |
I think for this reason that this difficulty existed, the transformer was invented in natural
link |
language processing not in vision.
link |
What the transformer is doing, it's a hierarchy of layers where every layer learns what to
link |
pay attention to in the given context in the previous layer.
link |
What to make the statistics over.
link |
And the context is significantly larger than the adjacent word.
link |
The context that GP3 has been using, the transformer itself is from 2017 and it wasn't using that
link |
large of a context.
link |
OpenAI has basically scaled up this idea as far as they could at the time.
link |
The context is about 2048 symbols, tokens in the language.
link |
These symbols are not characters, but they take the words and project them into a vector
link |
space where words that are statistically cooccurring a lot are neighbors already.
link |
So it's already a simplification of the problem a little bit.
link |
Every word is basically a set of coordinates in a high dimensional space.
link |
And then they use some kind of trick to also encode the order of the words in a sentence
link |
or in the not just sentence, but 2048 tokens is about couple pages of text or two and a
link |
half pages of text.
link |
And so they managed to do pretty exhaustive statistics over the potential relationships
link |
between two pages of text, which is tremendous, right?
link |
I was just using a single sentence back then and I was only looking for first order relationships
link |
and they were really looking for much, much higher level relationships.
link |
And what they discover after they've fed this enormous amount of trim and they're pretty
link |
much the written internet or the subset of it that had some quality, but substantial portion
link |
of the common flaw that they're not only able to reproduce style, but they're also able
link |
to reproduce some pretty detailed semantics like being able to add three digit numbers
link |
and multiply two digit numbers or to translate between pro and languages and things like
link |
So the results that GBT3 got, I think were amazing.
link |
By the way, I actually didn't check carefully.
link |
It's funny you just mentioned how you coupled semantics to the multiplication.
link |
Is it able to do some basic math on two digit numbers?
link |
Okay, interesting.
link |
I thought, I thought there's a lot of failure cases.
link |
Yeah, it basically fails if you take larger digit numbers.
link |
So four digit numbers and so on makes carrying mistakes and so on.
link |
And if you take larger numbers, you don't get useful results at all.
link |
And this could be an issue of the training set where there are not many examples of successful
link |
long form addition and standard human written text.
link |
Humans aren't very good at doing three digit numbers either.
link |
Yeah, they're not, you're not writing a lot about it.
link |
And the other thing is that the loss function that is being used is only minimizing surprises.
link |
So it's predicting what comes next in a typical text.
link |
It's not trying to go for causal closure first as we do.
link |
But the fact that that kind of prediction works to generate text that's semantically
link |
rich and consistent is interesting.
link |
So yeah, so it's amazing that it's able to generate semantically consistent text.
link |
It's not consistent.
link |
So the problem is that it loses coherence at some point.
link |
But it's also, I think, not correct to say that GP2C is unable to deal with semantics
link |
at all because you ask it to perform certain transformations in text and it performs these
link |
transformation in text.
link |
And the kind of additions that is able to perform are transformations in text, right?
link |
And there are proper semantics involved.
link |
You can also do more.
link |
There was a paper that was generating lots and lots of mathematically correct text and
link |
was feeding this into a transformer.
link |
And as a result, it was able to learn how to do differentiation, integration in race
link |
that according to the authors, Mathematica could not, to which some of the people in
link |
Mathematica responded that they were not using Mathematica in the right way and so on.
link |
I have not really followed the resolution of this conflict.
link |
This part, as a small tangent, I really don't like in machine learning papers, which they
link |
often do anecdotal evidence.
link |
They'll find one example in some kind of specific use of Mathematica and demonstrate, look,
link |
they'll show successes and failures, but they won't have a very clear representation of
link |
how many cases this actually represents.
link |
Yes, but I think as a first paper, this is a pretty good start.
link |
And so the take home message, I think, is that the authors could get better results
link |
from this and their experiments than they could get from the vein, which they were using
link |
computer algebra systems, which means that was not nothing.
link |
And it's able to perform substantially better than GPTSV can, based on a much larger amount
link |
of training data, using the same underlying algorithm.
link |
Let me ask, again, so I'm using your tweets as if this is like Play Doh, right?
link |
As if this is well thought out novels that you've written.
link |
You tweeted, GPT4 is listening to us now.
link |
This is one way of asking, what are the limitations of GPT3 when it scales?
link |
So what do you think will be the capabilities of GPT4, GPT5, and so on?
link |
What are the limits of this approach?
link |
So obviously, when we are writing things right now, everything that we are writing now is
link |
going to be training data for the next generation of machine learning models.
link |
So yes, of course, GPT4 is listening to us.
link |
And I think the tweet is already a little bit older.
link |
And we now have Wudao and we have a number of other systems that basically are placeholders
link |
I don't know what open AIS plans are in this card.
link |
I read that tweet in several ways.
link |
So one is obviously everything you put on the internet has used this training data.
link |
But in a second way, we talked about agency.
link |
I read it as almost like GPT4 is intelligent enough to be choosing to listen.
link |
So not only did a programmer tell it to collect this data and use it for training, I almost
link |
saw the humorous angle, which is like it has achieved AGI kind of thing.
link |
Well, the thing is, could we be already believing in GPT5?
link |
So GPT4 is listening, and GPT5 actually constructed the entirety of the reality where you can
link |
Of course, in some sense, what everybody is trying to do right now in AI is to extend
link |
the transformer to be able to deal with video.
link |
And there are very promising extensions, right?
link |
There's a work by Google that is called Perceiver.
link |
And that is overcoming some of the limitations of the transformer by letting it learn the
link |
topology of the different modalities separately and by training it to find better input features.
link |
So the basic feature abstractions that are being used by this successor to GPT3 are chosen
link |
such a way that it's able to deal with video input.
link |
And there is more to be done.
link |
So one of the limitations of GPT3 is that it's amnesiac.
link |
So it forgets everything beyond the two pages that it currently reads also during generation,
link |
not just during learning.
link |
Do you think that's fixable within the space of deep learning?
link |
Can you just make a bigger, bigger, bigger input?
link |
No, I don't think that our own working memory is infinitely large.
link |
It's probably also just a few thousand bits.
link |
But what you can do is you can structure this working memory.
link |
So instead of just force feeding this thing, a certain thing that it has to focus on and
link |
it's not allowed to focus on anything else because it's network, you allow it to construct
link |
its own working memory as we do.
link |
When we are reading a book, it's not that we are focusing our attention in such a way
link |
that we can only remember the current page.
link |
We will also try to remember other pages and try to undo what we learned from them or modify
link |
what we learned from them.
link |
We might get up and take another book from the shelf.
link |
We might go out and ask somebody that we can edit our working memory in any way that is
link |
useful to put a context together that allows us to draw the right inferences and to learn
link |
So this ability to perform experiments on the world based on an attempt to become fully
link |
coherent and to achieve causal closure, to achieve a certain aesthetic of your modeling,
link |
that is something that eventually needs to be done.
link |
And at the moment, we are skirting this in some sense by building systems that are larger
link |
and faster so they can use dramatically larger resources and human beings can do much more
link |
training data to get two models that in some sense are already very superhuman and in other
link |
ways are laughingly incoherent.
link |
So do you think making the systems, what would you say, multi resolutionals, some of the
link |
language models are focused on two pages.
link |
Some are focused on two books.
link |
Some are focused on two years of reading.
link |
Some are focused on a lifetime, so it's like stacks of GPT3s all the way down.
link |
You want to have gaps in between them.
link |
So it's not necessarily two years, there's no gaps.
link |
It stinks out of two years or out of 20 years or 2,000 years or 2 billion years where you
link |
are just selecting those bits that are predicted to be the most useful ones to understand what
link |
you're currently doing.
link |
And this prediction itself requires a very complicated model and that's the actual model
link |
that you need to be making.
link |
It's not just that you are trying to understand the relationships between things but what
link |
you need to make relationships or discover relationships over.
link |
I wonder what that thing looks like, what the architecture for the thing that's able
link |
to have that kind of model.
link |
I think it needs more degrees of freedom than the current models have.
link |
So it starts out with the fact that you possibly don't just want to have a feed forward model,
link |
but you want it to be fully recurrent.
link |
And to make it fully recurrent, you probably need to loop it back into itself and allow
link |
it to skip connections.
link |
Once you do this, you're predicting the next frame and your internal next frame in every
link |
moment and you are able to skip connection.
link |
It means that signals can travel from the output of the network into the middle of the
link |
network faster than the inputs do.
link |
Do you think it can still be differentiable?
link |
Do you think it still can be a neural network?
link |
Sometimes it can and sometimes it cannot.
link |
So it can still be a neural network but not the fully differentiable one.
link |
And when you want to deal with non differentiable ones, you need to have an attention system
link |
that is discrete and non dimensional and can perform grammatical operations.
link |
You need to be able to perform program synthesis.
link |
You need to be able to backtrack in this operations that you perform on this thing.
link |
This thing needs a model of what it's currently doing.
link |
And I think this is exactly the purpose of our own consciousness.
link |
Yeah, the program things that trickle on your networks.
link |
So let me ask you, it's not quite program synthesis, but the application of these language
link |
models to generation to program synthesis, but generation of programs.
link |
So if you look at GitHub OpenPilot, which is based on OpenAI's codecs, I don't know
link |
if you got a chance to look at it, but it's the system that's able to generate code once
link |
you prompt it with, what is it, like the header of a function with some comments, and it seems
link |
to do an incredibly good job or not a perfect job, which is very important, but an incredibly
link |
good job of generating functions.
link |
What do you make of that?
link |
Are you, is this exciting?
link |
Or is this just a party trick, a demo, or is this revolutionary?
link |
I haven't worked with it yet, so it's difficult for me to judge it, but I would not be surprised
link |
if it turns out to be a revolutionary.
link |
That's because the majority of programming tasks that are being done in the industry right
link |
now are not creative.
link |
People are writing code that other people have written, or they're putting things together
link |
from code fragments that others have had.
link |
And a lot of the work that programmers do in practice is to figure out how to overcome
link |
the gaps in their current knowledge and the things that people have already done.
link |
How to copy and paste from Stack Overflow, that's right.
link |
And so, of course, we can automate that to make it much faster to copy and paste from
link |
Yes, but it's not just copying and pasting.
link |
It's also basically learning which parts you need to modify to make them fit together.
link |
Like literally, sometimes as simple as just changing the variable names, so it fits into
link |
the rest of your code.
link |
Yes, but this requires that you understand the semantics of what you're doing to some
link |
And you can automate some of those things.
link |
The thing that makes people nervous, of course, is that a little bit wrong in a program can
link |
have a dramatic effect on the actual final operation of that program.
link |
So that's one little error, which in the space of language, it doesn't really matter, but
link |
in the space of programs, it can matter a lot.
link |
Yes, but this is already what is happening when humans program code.
link |
So we have a technology to deal with this.
link |
Somehow it becomes scarier when you know that a program generated code that's running a
link |
nuclear power plant, it becomes scarier.
link |
You know humans have errors too.
link |
But it's scarier when a program is doing it because why?
link |
I mean, there's a fear that a program may not be as good as humans to know when stuff
link |
is important to not mess up.
link |
There's a misalignment of priorities, of values, that's potential, that maybe that's the source
link |
I mean, okay, if I give you code generated by GitHub OpenPilot and code generated by
link |
a human and say, here, use one of these, how do you select today and in the next 10 years,
link |
which code do you use?
link |
Wouldn't you still be comfortable with the human?
link |
At the moment, when you go to Stanford to get an MRI, they will write a bill to the
link |
insurance over $20,000.
link |
And of this, maybe half of that gets paid by the insurance and the quarter gets paid
link |
by you and the MRI cost them $600 to make, maybe, probably less.
link |
And what are the values of the person that writes the software and deploys this process?
link |
It's very difficult for me to say whether I trust people.
link |
I think that what happens there is a mixture of proper Anglo Saxon Protestant values where
link |
somebody is trying to serve an abstract radar hole and organize crime.
link |
Well, that's a very harsh, I think that's a harsh view of humanity.
link |
There's a lot of bad people, whether incompetent or just malevolent in this world, yes.
link |
But it feels like the more malevolent, so the more damage you do to the world, the more
link |
resistance you have in your own human heart.
link |
But I don't explain with malevolence or stupidity what can be explained by just people acting
link |
on their incentives.
link |
So what happens in Stanford is not that somebody is evil, it's just that they do what they're
link |
No, it's not evil.
link |
So no, I see that as malevolence.
link |
I see, even like being a good German, as I told you offline, it's not absolute malevolence,
link |
but it's a small amount.
link |
I mean, when you see there's something wrong with the world, it's either incompetence,
link |
you're not able to see it, or it's cowardice that you're not able to stand up.
link |
Not necessarily in a big way, but in a small way.
link |
So I do think that is in a bit of malevolence.
link |
I'm not sure the example you're describing is a good example.
link |
So the question is, what is it that you are aiming for?
link |
And if you don't believe in the future, if you, for instance, think that the dollar is
link |
going to crash, why would you try to save dollars?
link |
If you don't think that humanity will be around in 100 years from now, because global
link |
warming will wipe out civilization, why would you need to act as if it were?
link |
So the question is, is there an overarching aesthetics that is projecting you and the
link |
world into the future, which I think is the basic idea of religion, that you understand
link |
the interactions that we have with each other as some kind of civilization level agent that
link |
is projecting itself into the future?
link |
If you don't have that shared purpose, what is there to be ethical for?
link |
So I think when we talk about Essex and AI, we need to go beyond the insane bias discussions
link |
and so on, where people are just measuring the distance between a statistic to their
link |
preferred current world model.
link |
The optimism, I was a little confused by the previous thing, just to clarify, there is
link |
a kind of underlying morality to having an optimism that human civilization will persist
link |
for longer than 100 years.
link |
I think a lot of people believe that it's a good thing for us to keep living and thriving.
link |
This morality itself is not an end to itself.
link |
It's instrumental to people living in 100 years from now, or 500 years from now.
link |
So it's only justifiable if you actually think that it will lead to people or increase the
link |
probability of people being arrived in that time frame.
link |
And a lot of people don't actually believe that, at least not actively.
link |
But I believe what exactly?
link |
Most people don't believe that they can afford to act on such a model.
link |
Basically what happens in the US is I think that the health care system is for a lot of
link |
people no longer sustainable, which means that if they need the help of the health care
link |
system, they're often not able to afford it.
link |
And when they cannot help it, they are often going bankrupt.
link |
I think the leading cause of personal bankruptcy in the US is the health care system.
link |
And that would not be necessary.
link |
It's not because people are consuming more and more medical services and are achieving
link |
a much, much longer life as a result.
link |
That's not actually the story that is happening, because you can compare it to other countries.
link |
And life expectancy in the US is currently not increasing, and it's not as high as in
link |
all the other industrialized countries.
link |
So some industrialized countries are doing better with a much cheaper health care system.
link |
And what you can see is, for instance, at administrative bloat, the health care system
link |
has maybe to some degree deliberately set up as a job placement program to allow people
link |
to continue living in middle class existence, despite not having a useful use case in productivity.
link |
So they are being paid to push paper around.
link |
And the number of administrators in the health care system has been increasing much faster
link |
than the number of practitioners.
link |
And this is something that you have to pay for.
link |
And also, the revenues that are being generated in the health care system are relatively large,
link |
and somebody has to pay for them.
link |
And the result why they are so large is because market mechanisms are not working.
link |
The FDA is largely not protecting people from malpractice of health care providers.
link |
The FDA is protecting health care providers from competition.
link |
So this is a thing that it has to do with values.
link |
And this is not because people are malicious on all levels, it's because they are not incentivized
link |
to act on a greater whole on this idea that you treat somebody who comes to you as a patient
link |
like you would treat a family member.
link |
Yeah, but we're trying, I mean, you're highlighting a lot of the flaws of the different institutions
link |
the systems were operating under, but I think there's a continued throughout history mechanism
link |
design of trying to design incentives in such a way that these systems behave better and
link |
better and better.
link |
I mean, it's a very difficult thing to operate a society of hundreds of millions of people
link |
So do we live in a society that is ever correcting?
link |
Do we observe that our models of what we are doing are predictive of the future and
link |
when they are not, we improve them?
link |
Our laws adjudicated with clauses that you put into every law, what is meant to be achieved
link |
by that law and the law will be automatically repealed if it's not achieving that, right?
link |
If you are optimizing your own laws, if you're writing your own source code, you probably
link |
make an estimate of what is the thing that's currently wrong in my life?
link |
What is it that I should change about my own policies?
link |
What is the expected outcome?
link |
And if that outcome doesn't manifest, I will change the policy back, right?
link |
Or I would change it to do something different.
link |
Are we doing this on a societal level?
link |
I think it's easy to sort of highlight the, I think we're doing it in the way that, like
link |
I operate my current life, I didn't sleep much last night.
link |
You would say that Lex, the way you need to operate your life is you need to always get
link |
The fact that you didn't sleep last night is totally the wrong way to operate in your
link |
Like you should have gotten all your shit done in time and gotten to sleep because sleep
link |
is very important for health, and you're highlighting, look, this person is not sleeping.
link |
Look, the medical, the healthcare system is operating, but the point is, it seems like
link |
this is the way, especially in the capitalist society, we operate, we keep running into
link |
And last minute, we tried to get our way out through innovation, and it seems to work.
link |
You have a lot of people that ultimately are trying to build a better world and get urgency
link |
about them when the problem becomes more and more imminent.
link |
And that's the way this operates.
link |
But if you look at the history, the long arc of history, it seems like that operating on
link |
deadlines produces progress and builds better and better systems.
link |
You probably agree with me that the US should have engaged in mass production in January
link |
2020, and that we should have shut down the airports early on, and that we should have
link |
made it mandatory that the people that work in nursery homes on campus, rather than living
link |
at home and then coming in and infecting people in the nursing homes that had no immune response
link |
And that is something that was, I think, visible back then, the correct decisions haven't
link |
We would have the same situation again.
link |
How do we know that these wrong decisions are not being made again?
link |
Have the people that made the decisions to not protect the nursing homes been punished?
link |
Have the people that made the wrong decisions with respect to testing that prevented the
link |
development of testing by startup companies and the importing of tests from countries
link |
that already had them?
link |
Have these people been held responsible?
link |
First of all, so what do you want to put before the firing squad?
link |
I think they are… No, just make sure that this doesn't happen
link |
No, but it's not that, yes, they're being held responsible by many voices, by people
link |
There's new leaders being born now, they're going to see rise to the top in 10 years.
link |
This moves slower than, there's obviously a lot of older incompetence and bureaucracy
link |
and these systems move slowly.
link |
They move like science one death at a time.
link |
So, yes, I think the pain that's been felt in the previous year is reverberating throughout
link |
Maybe I'm getting old.
link |
I suspect that every generation in the US after the war has lost the plot even more.
link |
I don't see this development.
link |
The war of World War II?
link |
Yes, so basically there was a time when we were modernist and in this modernist time,
link |
the US felt actively threatened by the things that happened in the world.
link |
The US was worried about possibility of failure and this imminence of possible failure led
link |
There was a time when the government would listen to physicists about how to do things.
link |
And the physicists were actually concerned about what the government should be doing.
link |
So, they would be writing letters to the government and so, for instance, the decision for the
link |
Manhattan Project was something that was driven in a conversation between physicists
link |
and the government.
link |
I don't think that the discussion would take place today.
link |
I think if the virus was much deadlier, we would see a very different response.
link |
I think the virus was not sufficiently deadly and instead, because it wasn't very deadly,
link |
what happened is the current system started to politicize it.
link |
This is what I realized with masks early on.
link |
They were not very quickly became not as a solution, but they became a thing that politicians
link |
used to divide the country.
link |
So, the same things happened with vaccines, same thing.
link |
So, people weren't talking about solutions to this problem because I don't think the
link |
problem was bad enough.
link |
When you talk about the war, I think our lives are too comfortable.
link |
I think in the developed world, things are too good and we have not faced severe dangers.
link |
When the dangers, the severe dangers existential threats are faced, that's when we step up.
link |
On a small scale and a large scale.
link |
Now, that's sort of my argument here, but I did think the virus, I was hoping that it
link |
was actually sufficiently dangerous for us to step up because especially in the early
link |
days, it was unclear.
link |
It still is unclear because of mutations, how bad it might be.
link |
So, I thought we would step up.
link |
The mask's point is a tricky one because to me, the manufacture of masks isn't even the
link |
I'm still to this day, and I was involved with a bunch of this work, have not seen good
link |
signs done on whether masks work or not.
link |
There still has not been a large scale study.
link |
To me, there should be large scale studies on every possible solution, aggressive.
link |
In the same way that the vaccine development was aggressive, there should be masks which
link |
test what kind of tests work really well.
link |
Even the question of how the virus spreads, there should be aggressive studies on that.
link |
To understand, I'm still, as far as I know, there's still a lot of uncertainty about that.
link |
Nobody wants to see this as an engineering problem that needs to be solved.
link |
It's that I was surprised about.
link |
We find that our views are largely convergent, but not completely.
link |
So, I agree with the thing that because our society in some sense perceives itself as too
link |
big to fail, and the virus did not alert people to the fact that we are facing possible failure
link |
that basically put us into the postmodernist mode.
link |
And I don't mean in a philosophical sense, but in a societal sense.
link |
The difference between a postmodern society and the modern society is that the modernist
link |
society has to deal with the ground tools, and the postmodernist society has to deal
link |
Politics becomes a performance, and the performance is done for an audience, and the organized
link |
audience is the media, and the media evaluates itself via other media.
link |
So you have an audience of critics that evaluate themselves.
link |
And I don't think it's so much the failure of the politicians because to get in power
link |
and to stay in power, you need to be able to deal with the published opinion.
link |
Well, I think it goes in cycles because what's going to happen is all of the small business
link |
owners, all the people who truly are suffering and will suffer more because of the effects
link |
of the closure of the economy and the lack of solutions to the virus, they're going
link |
And hopefully, I mean, this is where charismatic leaders can get the world in trouble, but
link |
hopefully will elect great leaders that will break through this postmodernist idea of the
link |
media and the perception and the drama on Twitter and all that kind of stuff.
link |
But you know this can go either way.
link |
When the Weimar Republic was unable to deal with the economic crisis that Germany was
link |
facing, there was an option to go back.
link |
But there were people which thought, let's get back to a constitutional monarchy and
link |
let's get this to work because democracy doesn't work.
link |
And eventually, there was no way back.
link |
When people decided there was no way back, they needed to go forward and the only options
link |
for going forward was to become Stalinist communists, basically an option to completely
link |
expropriate the factories and so on and nationalize them and to reorganize Germany in communist
link |
terms and ally itself with Stalin and fascism.
link |
And both options were obviously very bad.
link |
And the one that the Germans picked led to a catastrophe that devastated Europe.
link |
And I'm not sure if the US has an immune response against that.
link |
I think that the far right is currently very weak in the US, but this can easily change.
link |
Do you think from a historical perspective, Hitler could have been stopped from within
link |
Germany or from outside?
link |
Or this, well, depends on who you want to focus, whether you want to focus on Stalin
link |
But it feels like Hitler was the one as a political movement that could have been stopped.
link |
I think that the point was that a lot of people wanted Hitler, so he got support from a lot
link |
There was a number of industrialists who supported him because they thought that the democracy
link |
is obviously not working in unstable and you need a strongman.
link |
And he was willing to play that part.
link |
There were also people in the US who thought that Hitler would stop Stalin and would act
link |
as a bulwark against Bolshevism, which he probably would have done.
link |
But at which cost?
link |
And then many of the things that he was going to do, like the Holocaust, was something where
link |
people thought this is rhetoric.
link |
He's not actually going to do this, especially many of the Jews themselves, which were humanists.
link |
And for them, this was outside of the scope that was thinkable.
link |
I wonder if Hitler is uniquely, I want to carefully use this term, but uniquely evil.
link |
So if Hitler was never born, if somebody else would come in his place.
link |
So just thinking about the progress of history, how important are those singular figures that
link |
lead to mass destruction and cruelty?
link |
Because my sense is Hitler was unique.
link |
It wasn't just about the environment and the context that gave him.
link |
Another person would not come in his place to do as destructive of the things that he
link |
There was a combination of charisma, of madness, of psychopathy, of just ego, all those things
link |
which are very unlikely to come together in one person in the right time.
link |
It also depends on the context of the country that you're operating in.
link |
If you tell the Germans that they have a historical destiny in this romantic country, the effect
link |
is probably different than it is in other countries.
link |
But Stalin has killed a few more people than Hitler did.
link |
And if you look at the probability that you're survived under Stalin, Hitler killed people
link |
if he thought they were not worth living, or if they were harmful to his racist project.
link |
He basically felt that the Jews would be too cosmopolitan and would not be willing to participate
link |
in the racist redefinition of society and the value of society in an ethno state in
link |
this way, as he wanted it to have it.
link |
So he saw them as a harmful danger, especially since they played such an important role in
link |
the economy and culture of Germany.
link |
And so he had basically had some radical, but rational reason to murder them.
link |
And Stalin just killed everyone.
link |
He basically, Stalin's purges were such a random thing where he said that there's a
link |
certain possibility that this particular part of the population has a number of German collaborators
link |
or something, and we just killed them all.
link |
Right? Or if you look at what Mao did, the number of people that were killed in absolute
link |
numbers were much higher than the Mao that there were under Stalin.
link |
So it's super hard to say.
link |
The other thing is that you look at Genghis Khan and so on, how many people he killed.
link |
When you see there are a number of things that happen in human history that actually
link |
put a substantial dent in the existing population or Napoleon, and it's very difficult to eventually
link |
measure it because what's happening is basically evolution on a human scale where one monkey
link |
figures out a way to become viral and is using this viral technology to change the patterns
link |
of society at a very, very large scale.
link |
And what we find so abhorrent about these changes is the complexity that is being destroyed
link |
That it's basically like a big fire that burns out a lot of the existing culture and structure
link |
that existed before.
link |
And it all just starts with one monkey.
link |
One charismatic ape, and there's a bunch of them throughout history.
link |
But it's in a given environment.
link |
It's basically similar to wildfires in California.
link |
The temperature is rising.
link |
There is less rain falling.
link |
And then suddenly a single spark can have an effect that in other times would be contained.
link |
Speaking of which, I love how we went to Hitler and Stalin from 20, 30 minutes ago, GPT3 generating
link |
doing programs that this is.
link |
The argument was about morality of AI versus human.
link |
So, and specifically in the context of writing programs, specifically in the context of programs
link |
that can be destructive.
link |
So running nuclear power plants or autonomous weapons systems, for example.
link |
And I think your inclination was to say that it's not so obvious that AI would be less
link |
moral than humans or less effective at making a world that would make humans happy.
link |
So I'm not talking about self directed systems that are making their own goals at the global
link |
If you just talk about the deployment of technological systems that are able to see
link |
order and patterns and use this as control models to act on the goals that we give them.
link |
Then if we have the correct incentives to set the correct incentives for these systems,
link |
I'm quite optimistic.
link |
So humans versus AI.
link |
Let me give you an example.
link |
Autonomous weapon system.
link |
Let's say there's a city somewhere in the Middle East that has a number of terrorists.
link |
And the question is what's currently done with drone technologies, you have information
link |
about the location of a particular terrorist and you have a targeted attack, you have a
link |
bombing of that particular building.
link |
And that's all directed by humans at the high level strategy and also the deployment of
link |
individual bombs and missiles like the actual, everything is done by human except the final
link |
targeting and the, like the country, it's like the spot similar thing, like control
link |
What if you give AI control and saying write a program that says here's the best information
link |
available about the location of these five terrorists, here's the city.
link |
Make sure it's all the bombing you do is constrained to the city.
link |
Make sure it's precision based, but you take care of it.
link |
So you do one level of abstraction out and saying take care of the terrorist in the city.
link |
Which are you more comfortable with the humans or the JavaScript GPT three generated code
link |
that's doing the deployment?
link |
I mean, that's, this is the kind of question I'm asking is the kind of bugs that we see
link |
in human nature, are they better or worse than the kind of bugs we see in AI?
link |
They're different bugs.
link |
There is an issue that if people are creating an imperfect automation of a process that normally
link |
requires a mobile judgment, and this mobile judgment is the reason why it cannot be automated
link |
often is not because the computation is too expensive, but because the model that you give
link |
the AI is not an adequate model of the dynamics of the world, because the AI does not understand
link |
the context that it's operating in the right way.
link |
And this is something that already happens with Excel, where you don't need to have an
link |
AI system to do this, if you have an automated process in place, where humans decide using
link |
automated criteria whom to kill when, and whom to target when, which already happens.
link |
And you have no way to get off the kill list once that happens, once you have been targeted
link |
according to some automatic criterion by people, right, in a bureaucracy.
link |
That is the issue.
link |
The issue is not the AI, it's the automation.
link |
So there's something about, right, it's automation, but there's something about the, there's a
link |
certain level of abstraction where you give control to AI to do the automation.
link |
There's a scale that could be achieved that it feels like the scale of bug and scale mistake
link |
and scale of destruction that could be achieved of the kind that humans cannot achieve.
link |
So AI is much more able to destroy an entire country accidentally versus humans.
link |
It feels like the more civilians die as a react or suffer as a consequence of your decisions,
link |
the more weight there is on the human mind to make that decision.
link |
And so like it becomes more and more unlikely to make that decision for humans.
link |
For AI, it feels like it's harder to encode that kind of weight.
link |
In a way, the AI that we're currently building is automating statistics, right, intelligence
link |
is the ability to make models so you can act on them.
link |
And AI is a tool to make better models.
link |
So in principle, if you're using AI wisely, you're able to prevent more harm.
link |
And I think that the main issue is not on the side of the AI, it's on the side of the
link |
human command hierarchy that is using technology irresponsibly.
link |
So the question is how hard is it to encode, to properly encode the right incentives into
link |
And for instance, there's this idea of what happens if we let our airplanes being flown
link |
with AI systems and the neural network is a black box and so on.
link |
And it turns out our neural networks are actually not black boxes anymore.
link |
There are function approximators losing linear algebra, and there are performing things that
link |
we can understand.
link |
But we can also, instead of letting the neural network fly the airplane, use the neural network
link |
to generate a proof of the correct program.
link |
It's a degree of accuracy of the proof that a human could not achieve.
link |
So we can use our AI by combining different technologies to build systems that are much
link |
more reliable than the systems that a human being could create.
link |
And so in this sense, I would say that if you use an early stage of technology to save
link |
labor and don't employ competent people, but just to hack something together because
link |
you can, that is very dangerous.
link |
And if people are acting under these incentives that they get away with delivering shoddy work
link |
more cheaply using AI is less human oversight than before.
link |
That's very dangerous.
link |
The thing is though, AI is still going to be unreliable, perhaps less willing humans,
link |
but it'll be unreliable in novel ways.
link |
And yeah, but this is an empirical question.
link |
And it's something that we can figure out and work with.
link |
So the issue is, do we trust the systems, the social systems that we have in place and
link |
the social systems that we can build and maintain, that they're able to use AI responsibly?
link |
If they can, then AI is good news.
link |
If they cannot, then it's going to make the existing problems worse.
link |
Well, and also who creates the AI, who controls it, who makes money from it, because it's
link |
ultimately humans.
link |
And then you start talking about how much you trust the humans.
link |
So the question is, what does who mean?
link |
I don't think that we have identity per se.
link |
I think that the story of a human being is somewhat random.
link |
What happens is more or less that everybody is acting on their local incentives, what
link |
they perceive to be their incentives.
link |
And the question is, what are the incentives that the one that is pressing the button is
link |
It's nice for those incentives to be transparent.
link |
So for example, I'll give you example, there seems to be a significant distrust of a tech,
link |
like entrepreneurs in the tech space or people that run, for example, social media companies
link |
like Mark Zuckerberg.
link |
There's not a complete transparency of incentives under which that particular human being operates.
link |
We can listen to the words he says or what the marketing team says for a company, but
link |
And that becomes a problem when the algorithms and the systems created by him and other people
link |
in that company start having more and more impact on society.
link |
If the incentives were somehow the definition and the explainability of the incentives
link |
was decentralized such that nobody can manipulate it, no propaganda type manipulation of how
link |
these systems actually operate could be done, then yes, I think AI could achieve much fairer,
link |
much more effective solutions to difficult ethical problems.
link |
But when there's like humans in the loop manipulating the dissemination, the communication
link |
of how the system actually works, that feels like you can run into a lot of trouble.
link |
And that's why there's currently a lot of distrust for people at the heads of companies
link |
that have increasingly powerful AI systems.
link |
I suspect what happened traditionally in the US was that since our decision making is much
link |
more decentralized than in an authoritarian state, people are making decisions autonomously
link |
at many, many levels in the society.
link |
What happened was we created coherence and cohesion in society by controlling what people
link |
thought and what information they had.
link |
The media synchronized public opinion.
link |
And social media have disrupted this.
link |
It's not, I think, so much Russian influence or something.
link |
Everybody's influence.
link |
But a random person can come up with a conspiracy theory and disrupt what people think.
link |
And if that conspiracy theory is more compelling or more attractive than the standardized public
link |
conspiracy theory that we give people as a default, then it might get more traction.
link |
You suddenly have the situation that a single individual somewhere on a farm in Texas has
link |
more listeners than CNN.
link |
Which particular farmer you're referring to in Texas?
link |
Yes, I had dinner with him a couple of times.
link |
This is an interesting situation because you cannot get to be an anchor in CNN if you
link |
don't go through a complicated gatekeeping process.
link |
And suddenly you have random people without that gatekeeping process just optimizing for
link |
And not necessarily with a lot of responsibility for the long term effects of projecting these
link |
theories into the public.
link |
And now there is a push of making social media more like traditional media, which means that
link |
the opinion that is being projected in social media is more limited to an acceptable range
link |
with the goal of getting society into safe waters and increase the stability and cohesion
link |
of society again, which I think is a laudable goal.
link |
And of course, it also is an opportunity to cease the means of indoctrination.
link |
And the incentives that people are under when they do this in such a way that the AI ethics
link |
that we would need becomes very often something like AI politics, which is basically partisan
link |
And this means that whatever one side says, another side is going to be disagreeing with.
link |
In the same way as when you turn masks or the vaccine into a political issue, if you
link |
say that it is politically virtuous to get vaccinated, it will mean that the people that
link |
don't like you will not want to get vaccinated.
link |
And as soon as you have this partisan discourse, it's going to be very hard to make the right
link |
decisions because the incentives get to be the wrong ones.
link |
AI ethics needs to be super boring.
link |
It needs to be done by people who do statistics all the time and have extremely boring, long
link |
winded discussions that most people cannot follow because they are too complicated but
link |
that are dead serious.
link |
These people need to be able to be better at statistics than the leading machine learning
link |
And at the moment, the AI ethics debate is the one where you don't have any barrier to
link |
Everybody who has a strong opinion and is able to signal that opinion in the right way
link |
Strong words to judge it back.
link |
To me, that is a very frustrating thing because the field is so crucially important to our
link |
It's so crucially important.
link |
But the only qualification you currently need is to be outraged by the injustice in
link |
It's more complicated, right?
link |
Everybody seems to be outraged.
link |
So let's just say that the incentives are not always the right ones.
link |
So basically, I suspect that a lot of people that enter this debate don't have a vision
link |
for what society should be looking like in a way that is nonviolent, where we preserve
link |
liberal democracy, where we make sure that we all get along and we are around in a few
link |
hundred years from now, preferably with the comfortable technological civilization around
link |
I generally have a very foggy view of that world, but I tend to try to follow and I think
link |
society should in some degree follow the gradient of love, increasing the amount of love in
link |
And whenever I see different policies or algorithms or ideas that are not doing so, obviously,
link |
that's the ones that kind of resist.
link |
So the thing that terrifies me about this notion is I think that German fascism was driven
link |
It was just a very selective love.
link |
It was a love that basically…
link |
Well, now you're just manipulating.
link |
I mean, that's, you have to be very careful.
link |
You're talking to the wrong person in this way about love.
link |
So let's talk about what love is.
link |
And I think that love is the discovery of shared purpose.
link |
It's the recognition of the sacred and the other.
link |
And this enables nontransactual interactions.
link |
But the size of the other that you include needs to be maximized.
link |
So it's basically appreciation, like deep appreciation of the world around you fully,
link |
including the people that are very different than you, people that disagree with you completely,
link |
including people, including living creatures outside of just people, including ideas and
link |
appreciation of the full mess of it and also has to do with empathy, which is coupled with
link |
a lack of confidence and certainty of your own rightness.
link |
It's like a radical open mindedness to the way forward.
link |
I agree with every part of what you said.
link |
And now, if you scale it up, what you recognize is that love is, in some sense, the service
link |
to a next level agency, to the highest level agency that you can recognize.
link |
It could be, for instance, life on earth or beyond that, where you could say intelligent
link |
complexity in the universe that you try to maximize in a certain way.
link |
But when you think it's true, it basically means a certain aesthetic.
link |
And there is not one possible aesthetic, there are many possible aesthetics.
link |
And once you project an aesthetic into the future, you can see that there are some which
link |
affect from it, which are in conflict with it, that are corrupt, that are evil.
link |
You and me would probably agree that Hitler was evil, because the aesthetic of the world
link |
that he wanted is in conflict with the aesthetic of the world that you and me have in mind.
link |
And so the thing that he destroyed, we want to keep them in the world.
link |
There's a kind of ways to deal, I mean, Hitler's an easier case, but perhaps he wasn't so easy
link |
in the 30s, to understand who's Hitler and who's not before he was there.
link |
No, it's just that there was no consensus that the aesthetics that he had in mind were
link |
I mean, it's difficult.
link |
Love is complicated, because you can't just be so open minded that you let evil walk into
link |
the door, but you can't be so self assured that you can always identify evil perfectly,
link |
because that's what leads to Nazi Germany, having a certainty of what is and wasn't evil,
link |
like always drawing lines of good versus evil.
link |
There has to be a dance between hard stances extending up against what is wrong, and at
link |
the same time, empathy and open mindedness towards not knowing what is right and wrong,
link |
and a dance between those.
link |
I found that when I watched the Miyazaki movies that there is nobody who captures my spirituality
link |
as well as he does.
link |
It's very interesting, and just wishes.
link |
There is something going on in his movies that is very interesting.
link |
For instance, Mononoka is discussing not only an answer to Disney's simplistic notion
link |
of Mowgli, the jungle boy who was raised by wolves, and as soon as he sees people realizes
link |
that he's one of them, and the way in which the moral life and nature is simplified and
link |
romanticized and turned into kitsch is disgusting in the Disney movie, and he answers to this.
link |
He's replaced by Mononoka, this wolf girl who was raised by wolves and who was fierce
link |
and dangerous, and who cannot be socialized because she cannot be tamed, cannot be part
link |
You see, human society, it's something that is very, very complicated.
link |
You see people extracting resources and destroying nature, but the purpose is not to be evil,
link |
but to be able to have a life that is free from, for instance, oppression and violence
link |
and to curb death and disease.
link |
You basically see this conflict which cannot be resolved in a certain way.
link |
You see this moment when nature is turned into a garden, and it loses most of what it
link |
actually is, and humans no longer submitting to life and death and nature.
link |
To these questions, there is no easy answer, so it just turns it into something that is
link |
being observed as a journey that happens and that happens with a certain degree of inevitability,
link |
and the nice thing about all his movies is there is a certain main character, and it's
link |
the same in all movies.
link |
It's this little girl that is basically Heidi, and I suspect that happened because when he
link |
did field work for working on the Heidi movies back then, the Heidi animations, before he
link |
did his own movies, he traveled to Switzerland, and South Eastern Europe, and the Adriatic
link |
and so on, and got an idea about a certain aesthetic and a certain way of life that informed
link |
his future thinking.
link |
Heidi has a very interesting relationship to herself and to the world.
link |
There is nothing that she takes for herself.
link |
She is in a way fearless because she is committed to a service, to a greater whole.
link |
Basically she is completely committed to serving God.
link |
It's not an institutionalized God.
link |
It has nothing to do with the Roman Catholic Church or something like this.
link |
But in some sense, Heidi is an embodiment of the spirit of Protestantism.
link |
This idea of a being that is completely perfect and pure, and it's not a feminist vision
link |
because she is not a girl boss or something like this.
link |
She is the justification for the man in the audience to protect her, to build the civilization
link |
around her that makes her possible.
link |
She is not just the sacrifice of Jesus who is innocent and therefore, nailed to the cross.
link |
She is not being sacrificed.
link |
She is being protected by everybody around her who recognizes that she is sacred and
link |
there are enough around her to see that.
link |
This is a very interesting perspective.
link |
There is a certain notion of innocence, and this notion of innocence is not universal.
link |
It's not in all cultures.
link |
Hitler wasn't innocent.
link |
His idea of Germany was not that there is an innocence that is being protected.
link |
There was a predator that was going to triumph.
link |
And it's also something that is not at the core of every religion.
link |
There are many religions which don't care about innocence.
link |
They might care about increasing the status of something.
link |
And that's a very interesting notion that is quite unique and not claiming it's the
link |
It has a particular kind of aesthetic, which I think makes Miyazaki into the most relevant
link |
Protestant philosopher today.
link |
And you're saying in terms of all the ways that society can operate, perhaps the preservation
link |
of innocence might be one of the best?
link |
No, it's just my aesthetic.
link |
It's a particular way in which I feel that I relate to the world, that is natural to
link |
my own socialization, and maybe it's not an accident that I have cultural roots in Europe,
link |
in a particular world.
link |
So maybe it's a natural convergence point, and it's not something that you will find
link |
in all other times in history.
link |
So I'd like to ask you about Solzhenitsyn and our individual role as ants in this very
link |
So he says that some version of the line between good and evil runs to the heart of
link |
Do you think all of us are capable of good and evil?
link |
What's our role in this play in this game we're all playing?
link |
Is all of us capable to play any role?
link |
Is there an ultimate responsibility to, you mentioned, maintaining innocence or whatever
link |
the highest ideal for society you want, are all of us capable of living up to that?
link |
And that's our responsibility, or is there significant limitations to what we're able
link |
to do in terms of good and evil?
link |
So there is a certain way, if you are not terrible, if you are committed to some kind
link |
of civilizational agency, a next level agent that you are serving, some kind of transcendent
link |
In the eyes of that transcendental principle, you are able to discern good from evil, otherwise
link |
you cannot, otherwise you have just individual aesthetics.
link |
The cat that is torturing a mouse is not evil, because the cat does not envision or
link |
not part of the cat is envisioning a world where there is no violence and nobody is suffering.
link |
If you have an aesthetic where you want to protect innocence, then torturing somebody
link |
needlessly is evil, but only then.
link |
No, but within, I guess the question is within the aesthetic, within your sense of what is
link |
good and evil, we're still, it seems like we're still able to commit evil.
link |
So basically, if you are committing to this next level agent, you are not necessarily
link |
are this next level agent, right?
link |
You are part of it.
link |
You have a relationship to it, like a cell does to its organism, its hyperorganism.
link |
And it only exists to the degree that it's being implemented by you and others.
link |
And that means that you're not completely fully serving it.
link |
You have freedom in what you decide, whether you are acting on your impulses and local
link |
incentives and your verbal impulses, so to speak, or whether you're committing to it.
link |
And what you perceive then is a tension between what you would be doing with respect to the
link |
thing that you recognize as the sacred, if you do, and what you're actually doing.
link |
And this is the line between good and evil, right, where you see, oh, I'm here acting
link |
on my local incentives or impulses, and here I'm acting on what I consider to be sacred
link |
and there's a tension between those.
link |
And this is the line between good and evil that might run through your heart.
link |
And if you don't have that, if you don't have this relationship to a transcendental
link |
agent, you could call this relationship to the next level agent soul, right?
link |
It's not an immortal thing that is intrinsically valuable.
link |
It's a certain kind of relationship that you project to understand what's happening.
link |
Somebody is serving this transcendental sacredness or they're not.
link |
If you don't have a soul, you cannot be evil.
link |
You're addressed a complex natural phenomenon.
link |
So if you look at life, like starting today or starting tomorrow, when we leave here today,
link |
there's a bunch of trajectories that you can take through life, maybe countless.
link |
Do you think some of these trajectories, in your own conception of yourself, some of those
link |
trajectories are the ideal life?
link |
A life that if you were to be the hero of your life story, you would want to be.
link |
Look, is there some Joshua Bach that you're striving to be?
link |
Like this is the question I ask myself as an individual trying to make a better world
link |
in the best way that I could conceive of.
link |
What is my responsibility there?
link |
And how much am I responsible for the failure to do so?
link |
Because I'm lazy and incompetent too often in my own perception.
link |
In my own world view, I'm not very important.
link |
So I don't have place for me as a hero in my own world.
link |
I'm trying to do the best that I can, which is often not very good.
link |
And so it's not important for me to have status or to be seen in a particular way.
link |
It's helpful if others can see me, if a few people can see me, that can be my friends.
link |
I want to clarify.
link |
The hero didn't mean status or perception or some kind of marketing thing, but more
link |
in private, in the quiet of your own mind.
link |
Is there the kind of man you want to be and would consider it a failure if you don't become
link |
That's what I meant by hero.
link |
I don't perceive myself as having such an identity.
link |
And it's also sometimes frustrating.
link |
But it's basically a lack of having this notion of father that I need to be emulating.
link |
I mean, it's the leaf floating down the river.
link |
I worry that sometimes it's more like being the river.
link |
I'm just a fat frog sitting on a leaf on a dirty, muddy lake waiting for a princess to
link |
I forgot which way it goes.
link |
Maybe he kisses somebody.
link |
Can I ask you, I don't know if you know who Michael Malis is, but in terms of constructing
link |
systems of incentives, it's interesting to ask.
link |
I don't think I've talked to you about this before.
link |
Malis espouses anarchism.
link |
So he sees all government as fundamentally getting in the way or even being destructive
link |
to collaborations between human beings thriving.
link |
What do you think?
link |
What's the role of government in a society that thrives?
link |
Is anarchism at all compelling to you as a system?
link |
So not just small government, but no government at all.
link |
Yeah, I don't see how this would work.
link |
The government is an agent that imposes an offset on your reward function, on your payout
link |
metrics, so your behavior becomes compatible with the common good.
link |
So the argument there is that you can have collectives, like governing organizations,
link |
but not government, like where you're born in a particular set of land, and therefore
link |
you must follow this rule or else.
link |
You're forced by what they call violence because there's an implied violence here.
link |
So the key aspect of government is it protects you from the rest of the world with an army
link |
So it has a monopoly on violence.
link |
It's the only one that's able to do violence.
link |
So there are many forms of government.
link |
Not all governments do that, but we find that in successful countries, the government has
link |
a monopoly on violence.
link |
And that means that you cannot get ahead by starting your own army because the government
link |
will come down on you and destroy you if you try to do that.
link |
And in countries where you can build your own army and get away with it, some people
link |
And these countries is what we call failed countries in a way.
link |
And if you don't want to have violence, the point is not to appeal to the moral intentions
link |
of people because some people will use strategies if they get ahead with them that feel a particular
link |
kind of ecological niche.
link |
So you need to destroy that ecological niche.
link |
And if an effective government has a monopoly on violence, it can create a world where nobody
link |
is able to use violence and get ahead.
link |
So you want to use that monopoly on violence not to exert violence, but to make violence
link |
impossible, to raise the cost of violence.
link |
So people need to get ahead with nonviolent means.
link |
So the idea is that you might be able to achieve that in an anarchist state with companies.
link |
So with the forces of capitalism is create security companies where the one that's most
link |
ethically sound rises to the top, basically, it would be a much better representative of
link |
the people because there is less sort of stickiness to the big military force sticking around,
link |
even though it's long outlived.
link |
So you have groups of militants that are hopefully officially organized because otherwise they're
link |
going to lose against the other groups of militants.
link |
And they are coordinating themselves with the rest of society until they are having
link |
a monopoly on violence.
link |
How is that different for a government?
link |
So it's basically converging to the same thing.
link |
So I think it always, in my, as I was trying to argue with Malice, I feel like it always
link |
converges towards government at scale.
link |
But I think the idea is you can have a lot of collectives that are, you basically never
link |
let anything scale too big.
link |
So one of the problems with governments is it gets too big in terms of the size of the
link |
group over which it has control.
link |
My sense is that would happen anyway.
link |
So a successful company like Amazon or Facebook, I mean, it starts forming a monopoly over
link |
entire populations, not over just hundreds of millions, but billions of people.
link |
So I don't know, but there is something about the abuses of power the government can have
link |
when it has a monopoly on violence, right?
link |
And so that's, that's attention there.
link |
But so the question is, how can you set the incentives for government correctly?
link |
And this mostly applies at the highest levels of government.
link |
And we, because we haven't found a way to set them correctly, we made the highest levels
link |
of government relatively weak.
link |
And this is, I think, part of the reason why we had difficulty to coordinate the pandemic
link |
And China didn't have that much difficulty.
link |
And there is, of course, a much higher risk of the abuse of power that exists in China
link |
because the power is largely unchecked.
link |
And that's basically what happens in the next generation, for instance.
link |
Imagine that we would agree that the current government of China is largely correct and
link |
And maybe we don't agree on this, but if we did, how can we make sure that this stays
link |
And if you don't have checks and balances and division of power, it's hard to achieve.
link |
You don't have a solution for that problem.
link |
But the abolishment of government basically would remove the control structure.
link |
From a cybernetic perspective, there is an optimal point in the system that the regulation
link |
should be happening, where you can measure the current incentives, and the regulator
link |
would probably incentivize to make the right decisions and change the payout metrics of
link |
everything below it in such a way that the local prisoner's dilemmas get resolved.
link |
You cannot resolve the prisoner's dilemma without some kind of eternal control that
link |
emulates an infinite game in a way.
link |
I mean, there is a sense in which it seems like the reason the parts of government that
link |
don't work well currently is because there's not good mechanisms through which to interact
link |
for the citizenry to interact with government.
link |
It's basically, it hasn't caught up in terms of technology.
link |
And I think once you integrate some of the digital revolution of being able to have a
link |
lot of access to data, be able to vote on different ideas at a local level, at all levels,
link |
at the optimal level, like you're saying, that can resolve the prisoner dilemmas.
link |
And to integrate AI to help you out, automate things that don't require the human ingenuity
link |
– I feel like that's where government can operate that well – and can also break
link |
apart the inefficient bureaucracies if needed.
link |
There'll be a strong incentive to be efficient and successful.
link |
So out human history, we see an evolutionary competition of modes of government and of
link |
individual governments is in these modes.
link |
And every nation state in some sense is some kind of organism that has found different
link |
solutions for the problem of government.
link |
And you could look at all these different models and the different scales at which it
link |
exists as empirical attempts to validate the idea of how to build a better government.
link |
And I suspect that the idea of anarchism, similar to the idea of communism, is the result
link |
of being disenchanted with the ugliness of the real existing solutions and the attempt
link |
to get to an utopia.
link |
And I suspect that communism originally was not a utopia.
link |
I think that in the same way as original Christianity, it had a particular kind of vision.
link |
And this vision is a society, a mode of organization within the society in which humans can coexist
link |
at scale without coercion, in the same way as we do in a healthy family.
link |
In a good family, you don't terrorize each other into compliance, but you understand
link |
what everybody needs and what everybody is able to contribute and what the intended future
link |
of the whole thing is.
link |
And everybody coordinates their behavior in the right way and informs each other about
link |
how to do this and all the interactions that happen are instrumental to making that happen.
link |
Could this happen at scale?
link |
And I think this is the idea of communism.
link |
Communism is opposed to the idea that we need economic terror or other forms of terror to
link |
But in practice, what happened is that the proto communist countries, the real existing
link |
socialism, replaced a part of the economic terror with moral terror.
link |
So we were told to do the right thing for moral reasons.
link |
And of course, it didn't really work.
link |
And the economy eventually collapsed.
link |
And the moral terror had actual real cost.
link |
People were imprisoned because they were morally noncompliant.
link |
And the other thing is that the idea of communism became a utopia.
link |
So it basically was projected into the afterlife.
link |
We were told in my childhood that communism was a hypothetical society to which we were
link |
in a permanent revolution that justified everything that was presently wrong with society morally.
link |
But it was something that our grandchildren probably would not ever see because it was
link |
too ideal and too far in the future to make it happen right now.
link |
And people were just not there yet morally.
link |
And the same thing happened with Christianity, right?
link |
This notion of heaven was mythologized and projected into an afterlife.
link |
And I think this was just the idea of God's kingdom, of this world in which we instantiate
link |
the next level transcendental agent in the perfect form.
link |
So everything goes smoothly and without violence and without conflict and without this human
link |
messiness on this economic messiness and the terror and coercion that existed in the present
link |
And the idea of that the humans can exist at scale in a harmonious way and noncoercively
link |
And both people tested it but didn't get it to work so far.
link |
And the utopia is a world in where you get all the good things without any of the bad
link |
And you are, I think, very susceptible to believe in utopias when you are very young
link |
and don't understand that everything has to happen in causal patterns, that there is always
link |
feedback loops that ultimately are closed.
link |
There's nothing that just happens because it's good or bad.
link |
Good or bad don't exist in isolation.
link |
They only exist with respect to larger systems.
link |
So can you intuit why utopias fail as systems?
link |
So having a utopia that's out there beyond the horizon, is it because then it's not only
link |
because it's impossible to achieve utopias, but it's because what certain humans, certain
link |
small number of humans start to sort of greedily attain power and money and control and influence
link |
as they become, as they see the power in using this idea of a utopia for propaganda.
link |
That's a little like saying, why is my garden not perfect?
link |
It's because some evil weeds are overgrowing it and they always do.
link |
So this is not how it works.
link |
A good garden is a system that is in balance and requires minimal interactions by the gardener.
link |
So you need to create a system that is designed to self stabilize and the design of social
link |
systems requires not just the implementation of the desired functionality, but the next
link |
level design, also in biological systems.
link |
You need to create a system that wants to converge to the intended function.
link |
So instead of just creating an institution like the FDA that is performing a particular
link |
kind of role in society, you need to make sure that the FDA is actually driven by a system
link |
that wants to do this optimally, that is incentivize to do it optimally, and then makes the performance
link |
that is actually enacted in every generation instrumental to that thing, that actual goal.
link |
That is much harder to design and to achieve.
link |
See if the design a system where I'm English and communism also was quote unquote incentivized
link |
to be a feedback loop system that achieves that utopia, it wasn't working given human
link |
The incentives were not correct given human nature.
link |
So how do you incentivize people when they are getting coal off the ground to work as
link |
Because it's a terrible job and it's very bad for your health and how do you do this?
link |
And you can give them prices and metals and status to some degree, there's only so much
link |
status to give for that, and most people will not fall for this, or you can pay them.
link |
And you probably have to pay them in an asymmetric way because if you pay everybody the same
link |
and you nationalize the coal mines, eventually people will figure out that they can game
link |
So you're describing capitalism.
link |
Capitalism is the present solution to the system.
link |
And what we also noticed that I think that Marx was correct in saying that capitalism
link |
is prone to crisis, that capitalism is a system that in its dynamics is not convergent
link |
It's not a stable system.
link |
And that eventually it produces an enormous potential for productivity, but it also is
link |
systematically misallocating resources.
link |
So a lot of people cannot participate in the production and consumption anymore.
link |
And this is what we observe.
link |
We observe that the middle class in the US is tiny.
link |
That's a lot of people think that they're middle class, but if you are still flying
link |
economy, you're not middle class, right?
link |
Every class is a magnitude smaller than the previous class, and I think about classes
link |
is really like airline classes, a lot of people are economy, business class and very few
link |
are first class and summer budget.
link |
I mean, I understand.
link |
I think there is, yeah, maybe some people probably I would push back against that definition
link |
of the middle class.
link |
It does feel like the middle class is pretty large, but yes, there's a discrepancy in terms
link |
So if you think about the terms of the productivity that our society could have, there is no reason
link |
for anybody to fly economy, right?
link |
We would be able to let everybody travel in style.
link |
But also some people like to be frugal even when they're billionaires, okay?
link |
So let's take that into account.
link |
Yes, but I mean, we probably don't need to be traveling lavish, but you also don't need
link |
to be tortured, right?
link |
There is a difference between frugal and subjecting yourself to torture.
link |
Listen, I love economy.
link |
I don't understand why you're comparing a flying economy to torture.
link |
Although the flight here, there's two crying babies next to me, so that, but that has nothing
link |
to do with crying babies.
link |
They're very cute though.
link |
Yeah, I have two kids and sometimes I have to go back to visit the grandparents and back
link |
means going from the west coast to Germany and that's a long flight.
link |
Is it true that when you're a father, you grow immune to the crying and all that kind
link |
Because like me just not having kids, other people's kids can be quite annoying when they're
link |
crying and screaming and all that kind of stuff.
link |
When you have children and you're wired up in the default natural way, you're lucky in
link |
this regard, you fall in love with them.
link |
And this falling in love with them means that you basically start to see the world through
link |
their eyes and you understand that in a given situation, they cannot do anything but expressing
link |
And so it becomes more differentiated.
link |
I noticed that for instance, my son is typically acting on a pure experience of what things
link |
are like right now and he has to do this right now and you have this small child that is
link |
when he was a baby and so on where he was just immediately expressing what he felt and
link |
if you cannot regulate this from the outside, there's no point to be upset about it.
link |
It's like dealing with weather or something like this.
link |
You all have to get through it and it's not easy for him either.
link |
But if you also have a daughter, maybe she is planning for that.
link |
Maybe she understands that she's sitting in the car behind you and she's screaming at
link |
the top of her lungs and you're almost doing an accident and you really don't know what
link |
What should I have done to make you stop screaming?
link |
You could have given me candy.
link |
I think that's like a cat versus dog discussion.
link |
Because you said like a fundamental aspect of that is love that makes it all worth it.
link |
What in this monkey riding an elephant in a dream world, what role does love play in
link |
the human condition?
link |
I think that love is the facilitator of nontransactual interaction and you are observing your own
link |
They go beyond your ego, they go beyond the particular organism that you are and your
link |
That's what you mean by nontransactual.
link |
So basically when you are acting in a transactional way, it means that you are respecting something
link |
in return for you from the one that you're interacting with.
link |
You are interacting with a random stranger, you buy something from them on eBay, you expect
link |
a fair value for the money that you sent them and vice versa.
link |
Because you don't know that person, you don't have any kind of relationship to them.
link |
And when you know this person a little bit better and you know the situation that they
link |
are in and you understand what they try to achieve in their life and you approve because
link |
you realize that they are in some sense serving the same human sacredness as you are.
link |
And they need to think that you have, maybe you give it to them as a present.
link |
But the feeling itself of joy is a kind of benefit, is a kind of transaction.
link |
Yes, but the joy is not the point.
link |
The joy is the signal that you get, it's the reinforcement signal that your brain sends
link |
to you because you are acting on the incentives of the agent that you're part of.
link |
We are meant to be part of something larger.
link |
That is the way in which we outcompeted other hominins.
link |
Take that Neanderthals.
link |
And also other humans.
link |
There was a population bottleneck for human society that leads to an extreme lack of genetic
link |
diversity among humans.
link |
If you look at Bushmen in the Kalahari that basically tribes that are not that far distant
link |
to each other have more genetic diversity than exists between Europeans and Chinese.
link |
And it's because basically the out of Africa population at some point had a bottleneck
link |
of just a few thousand individuals.
link |
And what probably happened is not that at any time the number of people shrunk below
link |
a few hundred thousand.
link |
What probably happened is that there was a small group that had a decisive mutation
link |
that produced an advantage.
link |
And this group multiplied and killed everybody else.
link |
And we are descendants of that group.
link |
Yeah, I wonder what the peculiar characteristics of that group.
link |
I mean, we can never know.
link |
And a lot of people do.
link |
We can only just listen to the echoes and the ripples that are still within us.
link |
So I suspect what eventually made a big difference was the ability to organize at scale, to program
link |
That we became programmable, that we are willing to work on lockstep, that we went below, above
link |
the tribal level, that we no longer were groups of a few hundred individuals and acted on direct
link |
reputation systems transactionally, but that we basically evolved an adaptation to become
link |
To form collectives outside of the direct collectives.
link |
And that's basically a part of us became committed to serving something outside of what we know.
link |
Then that's kind of what love is.
link |
And it's terrifying because it meant that we eradicated the others.
link |
It's an adaptive force that gets us ahead in evolution, which means we displaced something
link |
else that doesn't have that.
link |
Oh, so we had to murder a lot of people that weren't about love.
link |
So love led to destruction.
link |
They didn't have the same strong love as we did.
link |
That's why I mentioned this thing, this fascism.
link |
When you see these speeches, do you want total war?
link |
And everybody says, yes, this is this big, oh my God, be a part of something that is
link |
more important than me, that gives meaning to my existence.
link |
Do you have advice for young people today in high school, in college, that are thinking
link |
about what to do with their career, with their life, so that at the end of the whole thing,
link |
they could be proud of what they did?
link |
So what does integrity look like when you're at the river or at the leaf or the fat frog
link |
It basically means that you try to figure out what the thing is that is the most right.
link |
And this doesn't mean that you have to look for what other people tell you what's right,
link |
but you have to aim for moral autonomy.
link |
So things need to be right independently of what other people say.
link |
I always felt that when people told me to listen to what others say, read the room,
link |
build your ideas of what's true based on the high status people of your ingroup, that does
link |
not protect me from fascism.
link |
The only way to protect yourself from fascism is to decide it's the world that is being
link |
built here, the world that I want to be in.
link |
And so in some sense, try to make your behavior sustainable, act in such a way that you would
link |
feel comfortable on all sides of the transaction.
link |
Realize that everybody is you in a different timeline, but is seeing things differently
link |
and has reasons to do so.
link |
Yeah, I've come to realize this recently.
link |
There is an inner voice that tells you what's right and wrong.
link |
And speaking of reading the room, what integrity looks like is there's times when a lot of
link |
people are doing something wrong and what integrity looks like is not going on Twitter
link |
and tweeting about it, but not participating quietly, not doing, so it's not like signaling
link |
or not all this kind of stuff, but actually living your what you think is right, like
link |
That's also sometimes this expectation that others are like us.
link |
So imagine the possibility that some of the people around you are space aliens that only
link |
So they don't have the same prayers as you do.
link |
They don't have the same impulses that's right and wrong.
link |
There's a large diversity in these basic impulses that people can have in a given situation.
link |
And now realize that you are a space alien.
link |
You are not actually human.
link |
You think that you're human, but you don't know what it means, like what it's like to
link |
You just make it up as you go along like everybody else.
link |
And you have to figure that out what it means that you are a full human being, what it means
link |
to be human in the world and how to connect with others on that.
link |
And there is also something, don't be afraid in the sense that if you do this, you're not
link |
good enough because if you are acting on these incentives of integrity, you become trustworthy.
link |
That's the way in which you can recognize each other.
link |
There is a particular place where you can meet and you can figure out what that place
link |
is, where you will give support to people because you realize that they act with integrity
link |
and they will also do that.
link |
So in some sense, you are safe if you do that.
link |
You're not always protected.
link |
There are people which will abuse you and that might, that are bad actors in a way that
link |
it's hard to imagine before you meet them.
link |
But there is also people which will try to protect you.
link |
Yeah, that's such a, thank you for saying that, that's such a hopeful message that no
link |
matter what happens to you, there will be a place.
link |
There's people you'll meet that also have what you have and you will find happiness there
link |
Yeah, but it doesn't need to end well.
link |
It can also all go wrong.
link |
So there's no guarantees in this life.
link |
So you can do everything right and you still can fail and horrible things happen to you
link |
that traumatize you and mutilate you and you have to be grateful if it doesn't happen.
link |
And ultimately be grateful no matter what happens because even just being alive is pretty damn
link |
Yeah, even that, you know, the gratefulness in some sense is also just generated by your
link |
brain to keep you going.
link |
It's all the trick.
link |
Speaking of which, Camus said, I see many people die because they judge that life is
link |
I see others paradoxically getting killed for their ideas or illusions that give them
link |
a reason for living.
link |
What is called the reason for living is also an excellent reason for dying.
link |
I therefore conclude that the meaning of life is the most urgent of questions.
link |
So I have to ask what Joshua Bach is the meaning of life?
link |
It is an urgent question according to Camus.
link |
I don't think that there's a single answer to this.
link |
Nothing makes sense unless a mind makes it so.
link |
So you basically have to project a purpose.
link |
And if you zoom out far enough, there is the heat test of the universe and everything is
link |
Everything is just a blip in between.
link |
And the question is, do you find meaning in this blip in between?
link |
Do you find meaning in observing squirrels?
link |
Do you find meaning in raising children and projecting a multigenerational organism into
link |
Do you find meaning in projecting an aesthetic of the world that you like to the future and
link |
trying to serve that aesthetic?
link |
And if you do, then life has that meaning.
link |
And if you don't, then it doesn't.
link |
I kind of enjoy the idea that you just create the most vibrant, the most weird, the most
link |
unique kind of blip you can, given your environment, given your set of skills.
link |
Just be the most weird set of like local pocket of complexity you can be.
link |
So that when people study the universe, they'll pause and be like, uh, that's weird.
link |
It looks like a useful strategy, but of course it's still motivated reasoning.
link |
You're obviously acting on your incentives here.
link |
It's still a story we tell ourselves when then a dream that's hardly in touch with
link |
It's definitely a good strategy if you are a podcaster.
link |
And the human, which I'm still trying to figure out if I am, there's a mutual relationship
link |
Josh, you're you're one of the most incredible people I know.
link |
I really love talking to you.
link |
I love talking to you again, and it's really an honor that you spend your valuable time
link |
I hope we get to talk many times throughout our, uh, through our short and meaningless
link |
I enjoyed this conversation very much.
link |
Thanks for listening to this conversation with Yosha Bach and thank you to Coinbase,
link |
Codecademy, Linode, Netsuite, and ExpressVPN.
link |
Check them out in the description to support this podcast.
link |
Now, let me leave you with some words from Carl Jung.
link |
People will do anything, no matter how absurd in order to avoid facing their own souls.
link |
One does not become enlightened by imagining figures of light, but by making the darkness
link |
Thank you for listening and hope to see you next time.