back to indexDr. Lex Fridman: Machines, Creativity & Love | Huberman Lab Podcast #29
link |
Welcome to the Huberman Lab Podcast,
link |
where we discuss science and science-based tools
link |
for everyday life.
link |
I'm Andrew Huberman, and I'm a professor of neurobiology
link |
and ophthalmology at Stanford School of Medicine.
link |
Today, I have the pleasure of introducing Dr. Lex Friedman
link |
as our guest on the Huberman Lab Podcast.
link |
Dr. Friedman is a researcher at MIT
link |
specializing in machine learning, artificial intelligence,
link |
and human-robot interactions.
link |
I must say that the conversation with Lex
link |
was without question one of the most fascinating
link |
conversations that I've ever had,
link |
not just in my career, but in my lifetime.
link |
I knew that Lex worked on these topics,
link |
and I think many of you are probably familiar with Lex
link |
and his interest in these topics
link |
from his incredible podcast, the Lex Friedman Podcast.
link |
If you're not already watching that podcast,
link |
please subscribe to it, it is absolutely fantastic.
link |
But in holding this conversation with Lex,
link |
I realized something far more important.
link |
He revealed to us a bit of his dream,
link |
his dream about humans and robots,
link |
about humans and machines,
link |
and about how those interactions can change the way
link |
that we perceive ourselves
link |
and that we interact with the world.
link |
We discuss relationships of all kinds,
link |
relationships with animals, relationships with friends,
link |
relationships with family, and romantic relationships.
link |
And we discuss relationships with machines,
link |
machines that move and machines that don't move,
link |
and machines that come to understand us in ways
link |
that we could never understand for ourselves,
link |
and how those machines can educate us about ourselves.
link |
Before this conversation,
link |
I had no concept of the ways in which machines
link |
could inform me or anyone about themselves.
link |
By the end, I was absolutely taken with the idea,
link |
and I'm still taken with the idea
link |
that interactions with machines of a very particular kind,
link |
a kind that Lex understands and wants to bring to the world,
link |
can not only transform the self,
link |
but may very well transform humanity.
link |
So whether or not you're familiar
link |
with Dr. Lex Friedman or not,
link |
I'm certain you're going to learn a tremendous amount
link |
from him during the course of our discussion,
link |
and that it will transform the way
link |
that you think about yourself and about the world.
link |
Before we begin, I want to mention
link |
that this podcast is separate
link |
from my teaching and research roles at Stanford.
link |
It is, however, part of my desire and effort
link |
to bring zero cost to consumer information about science
link |
and science-related tools to the general public.
link |
In keeping with that theme,
link |
I'd like to thank the sponsors of today's podcast.
link |
Our first sponsor is Roca.
link |
Roca makes sunglasses and eyeglasses
link |
that are of absolutely phenomenal quality.
link |
The company was founded
link |
by two all-American swimmers from Stanford,
link |
and everything about the sunglasses and eyeglasses
link |
they've designed had performance in mind.
link |
Now, I've spent a career working on the visual system,
link |
and one of the fundamental issues
link |
that your visual system has to deal with
link |
is how to adjust what you see when it gets darker
link |
or brighter in your environment.
link |
With Roca sunglasses and eyeglasses,
link |
whether or not it's dim in the room or outside,
link |
whether or not there's cloud cover,
link |
or whether or not you walk into a shadow,
link |
you can always see the world with absolute clarity.
link |
And that just tells me that they really understand
link |
the way that the visual system works,
link |
processes like habituation and attenuation.
link |
All these things that work at a real mechanistic level
link |
have been built into these glasses.
link |
In addition, the glasses are very lightweight.
link |
You don't even notice really that they're on your face.
link |
And the quality of the lenses is terrific.
link |
Now, the glasses were also designed
link |
so that you could use them not just while working
link |
or at dinner, et cetera, but while exercising.
link |
They don't fall off your face or slip off your face
link |
if you're sweating.
link |
And as I mentioned, they're extremely lightweight,
link |
so you can use them while running,
link |
you can use them while cycling, and so forth.
link |
Also, the aesthetic of Roca glasses is terrific.
link |
Unlike a lot of performance glasses out there,
link |
which frankly make people look like cyborgs,
link |
these glasses look great.
link |
You can wear them out to dinner.
link |
You can wear them for essentially any occasion.
link |
If you'd like to try Roca glasses,
link |
you can go to roca.com, that's R-O-K-A.com,
link |
and enter the code Huberman
link |
to save 20% off your first order.
link |
That's Roca, R-O-K-A.com,
link |
and enter the code Huberman at checkout.
link |
Today's episode is also brought to us by InsideTracker.
link |
InsideTracker is a personalized nutrition platform
link |
that analyzes data from your blood and DNA
link |
to help you better understand your body
link |
and help you reach your health goals.
link |
I am a big believer in getting regular blood work done
link |
for the simple reason that many of the factors
link |
that impact our immediate and long-term health
link |
can only be assessed from a quality blood test.
link |
And now with the advent of quality DNA tests,
link |
we can also get insight into some of our genetic
link |
underpinnings of our current and long-term health.
link |
The problem with a lot of blood and DNA tests out there,
link |
however, is you get the data back
link |
and you don't know what to do with those data.
link |
You see that certain things are high
link |
or certain things are low,
link |
but you really don't know what the actionable items are,
link |
what to do with all that information.
link |
With InsideTracker, they make it very easy
link |
to act in the appropriate ways on the information
link |
that you get back from those blood and DNA tests.
link |
And that's through the use of their online platform.
link |
They have a really easy to use dashboard
link |
that tells you what sorts of things can bring the numbers
link |
for your metabolic factors, endocrine factors, et cetera,
link |
into the ranges that you want and need
link |
for immediate and long-term health.
link |
In fact, I know one individual just by way of example
link |
that was feeling good but decided to go
link |
with an InsideTracker test and discovered
link |
that they had high levels of what's called
link |
C-reactive protein.
link |
They would have never detected that otherwise.
link |
C-reactive protein is associated
link |
with a number of deleterious health conditions,
link |
some heart issues, eye issues, et cetera.
link |
And so they were able to take immediate action
link |
to try and resolve those CRP levels.
link |
And so with InsideTracker, you get that sort of insight.
link |
And as I mentioned before, without a blood or DNA test,
link |
there's no way you're going to get that sort of insight
link |
until symptoms start to show up.
link |
If you'd like to try InsideTracker,
link |
you can go to insidetracker.com slash Huberman
link |
to get 25% off any of InsideTracker's plans.
link |
You just use the code Huberman at checkout.
link |
That's insidetracker.com slash Huberman
link |
to get 25% off any of InsideTracker's plans.
link |
Today's podcast is brought to us by Athletic Greens.
link |
Athletic Greens is an all-in-one
link |
vitamin mineral probiotic drink.
link |
I started taking Athletic Greens way back in 2012.
link |
And so I'm delighted that they're sponsoring the podcast.
link |
The reason I started taking Athletic Greens
link |
and the reason I still take Athletic Greens
link |
is that it covers all of my vitamin mineral probiotic bases.
link |
In fact, when people ask me, what should I take?
link |
I always suggest that the first supplement people take
link |
is Athletic Greens for the simple reason
link |
is that the things it contains covers your bases
link |
for metabolic health, endocrine health,
link |
and all sorts of other systems in the body.
link |
And the inclusion of probiotics are essential
link |
for a healthy gut microbiome.
link |
There are now tons of data showing
link |
that we have neurons in our gut.
link |
And keeping those neurons healthy
link |
requires that they are exposed
link |
to what are called the correct microbiota,
link |
little microorganisms that live in our gut
link |
and keep us healthy.
link |
And those neurons in turn help keep our brain healthy.
link |
They influence things like mood, our ability to focus,
link |
and many, many other factors related to health.
link |
With Athletic Greens, it's terrific
link |
because it also tastes really good.
link |
I drink it once or twice a day.
link |
I mix mine with water and I add a little lemon juice
link |
or sometimes a little bit of lime juice.
link |
If you want to try Athletic Greens,
link |
you can go to athleticgreens.com slash Huberman.
link |
And if you do that, you can claim their special offer.
link |
They're giving away five free travel packs,
link |
the little packs that make it easy to mix up
link |
Athletic Greens while you're on the road.
link |
And they'll give you a year supply of vitamin D3 and K2.
link |
Again, go to athleticgreens.com slash Huberman
link |
to claim that special offer.
link |
And now my conversation with Dr. Lex Friedman.
link |
Thanks so much for sitting down with me.
link |
I have a question that I think is on a lot of people's minds
link |
or ought to be on a lot of people's minds
link |
because we hear these terms a lot these days,
link |
but I think most people, including most scientists
link |
and including me, don't know really
link |
what is artificial intelligence and how is it different
link |
from things like machine learning and robotics?
link |
So if you would be so kind as to explain to us
link |
what is artificial intelligence and what is machine learning?
link |
Well, I think that question is as complicated
link |
and as fascinating as the question of what is intelligence.
link |
So I think of artificial intelligence first
link |
as a big philosophical thing.
link |
Pamela McCordick said AI was the ancient wish
link |
to forge the gods or was born as an ancient wish
link |
to forge the gods.
link |
So I think at the big philosophical level,
link |
it's our longing to create other intelligent systems,
link |
perhaps systems more powerful than us.
link |
At the more narrow level, I think it's also a set of tools
link |
that are computational mathematical tools
link |
to automate different tasks.
link |
And then also it's our attempt to understand our own mind.
link |
So build systems that exhibit some intelligent behavior
link |
in order to understand what is intelligence
link |
in our own selves.
link |
So all those things are true.
link |
Of course, what AI really means as a community,
link |
as a set of researchers and engineers,
link |
it's a set of tools, a set of computational techniques
link |
that allow you to solve various problems.
link |
There's a long history that approaches the problem
link |
from different perspectives.
link |
What's always been throughout one of the threads,
link |
one of the communities goes under the flag
link |
of machine learning, which is emphasizing
link |
in the AI space, the task of learning.
link |
How do you make a machine that knows very little
link |
in the beginning, follow some kind of process
link |
and learns to become better and better
link |
at a particular task.
link |
What's been most very effective in the recent about 15 years
link |
is a set of techniques that fall under the flag
link |
of deep learning that utilize neural networks.
link |
What neural networks are, are these fascinating things
link |
inspired by the structure of the human brain very loosely,
link |
but they have, it's a network of these little basic
link |
computational units called neurons, artificial neurons.
link |
And they have, these architectures have an input
link |
and output, they know nothing in the beginning
link |
and they're tasked with learning something interesting.
link |
What that something interesting is,
link |
usually involves a particular task.
link |
The, there's a lot of ways to talk about this
link |
and break this down.
link |
Like one of them is how much human supervision
link |
is required to teach this thing.
link |
So supervised learning, this broad category,
link |
is the neural network knows nothing in the beginning
link |
and then it's given a bunch of examples of,
link |
in computer vision, that will be examples of cats,
link |
dogs, cars, traffic signs, and then you're given the image
link |
and you're given the ground truth of what's in that image.
link |
And when you get a large database of such image examples
link |
where you know the truth, the neural network
link |
is able to learn by example,
link |
that's called supervised learning.
link |
The question, there's a lot of fascinating questions
link |
within that, which is how do you provide the truth?
link |
When you've given an image of a cat,
link |
how do you provide to the computer
link |
that this image contains a cat?
link |
Do you just say the entire image is a picture of a cat?
link |
Do you do what's very commonly been done,
link |
which is a bounding box?
link |
You have a very crude box around the cat's face
link |
saying this is a cat.
link |
Do you do semantic segmentation?
link |
Mind you, this is a 2D image of a cat.
link |
So it's not a, the computer knows nothing
link |
about our three-dimensional world.
link |
It's just looking at a set of pixels.
link |
So semantic segmentation is drawing a nice,
link |
very crisp outline around the cat and saying that's a cat.
link |
That's really difficult to provide that truth.
link |
And one of the fundamental open questions
link |
in computer vision is,
link |
is that even a good representation of the truth?
link |
Now, there's another contrasting set of ideas.
link |
Their attention, their overlapping
link |
is what's used to be called unsupervised learning,
link |
what's commonly now called self-supervised learning,
link |
which is trying to get less and less
link |
and less human supervision into the task.
link |
So self-supervised learning is more,
link |
it's been very successful in the domain of a language model,
link |
natural language processing,
link |
and now more and more is being successful
link |
in computer vision tasks.
link |
And what's the idea there is,
link |
let the machine without any ground truth annotation,
link |
just look at pictures on the internet
link |
or look at texts on the internet
link |
and try to learn something generalizable
link |
about the ideas that are at the core of language
link |
or at the core of vision.
link |
And based on that,
link |
we humans at its best like to call that common sense.
link |
So we have this giant base of knowledge
link |
on top of which we build more sophisticated knowledge,
link |
but we have this kind of common sense knowledge.
link |
And so the idea with self-supervised learning
link |
is to build this common sense knowledge
link |
about what are the fundamental visual ideas
link |
that make up a cat and a dog and all those kinds of things
link |
without ever having human supervision.
link |
The dream there is you just let an AI system
link |
that's self-supervised run around the internet for a while,
link |
watch YouTube videos for millions and millions of hours,
link |
and without any supervision be primed and ready
link |
to actually learn with very few examples
link |
once the human is able to show up.
link |
We think of children in this way, human children,
link |
is your parents only give one or two examples
link |
to teach a concept.
link |
The dream with self-supervised learning
link |
is that would be the same with machines
link |
that they would watch millions of hours of YouTube videos
link |
and then come to a human and be able to understand
link |
when the human shows them this is a cat.
link |
Like, remember this is a cat.
link |
They will understand that a cat is not just a thing
link |
with pointy ears or a cat is a thing that's orange
link |
or is furry, they'll see something more fundamental
link |
that we humans might not actually be able
link |
to introspect and understand.
link |
Like if I asked you what makes a cat versus a dog,
link |
you wouldn't probably not be able to answer that.
link |
But if I showed you, brought to you a cat and a dog,
link |
you'll be able to tell the difference.
link |
What are the ideas that your brain uses
link |
to make that difference?
link |
That's the whole dream of self-supervised learning
link |
is it would be able to learn that on its own,
link |
that set of common sense knowledge
link |
that's able to tell the difference.
link |
And then there's like a lot of incredible uses
link |
of self-supervised learning,
link |
very weirdly called self-play mechanism.
link |
That's the mechanism behind the reinforcement learning
link |
successes of the systems that want to go
link |
at alpha zero, that want to chess.
link |
Oh, I see, that play games.
link |
That play games. Got it.
link |
So the idea of self-play is probably applies
link |
to other domains than just games,
link |
is a system that just plays against itself.
link |
And this is fascinating in all kinds of domains,
link |
but it knows nothing in the beginning.
link |
And the whole idea is it creates a bunch of mutations
link |
of itself and plays against those versions of itself.
link |
And the fascinating thing is when you play against systems
link |
that are a little bit better than you,
link |
you start to get better yourself.
link |
Like learning, that's how learning happens.
link |
That's true for martial arts,
link |
that's true in a lot of cases,
link |
where you want to be interacting with systems
link |
that are just a little better than you.
link |
And then through this process of interacting
link |
with systems just a little better than you,
link |
you start following this process where everybody
link |
starts getting better and better and better and better
link |
until you are several orders of magnitude better
link |
than the world champion in chess, for example.
link |
And it's fascinating because it's like a runaway system.
link |
One of the most terrifying and exciting things
link |
that David Silver, the creator of AlphaGo and AlphaZero,
link |
one of the leaders of the team said to me is
link |
they haven't found the ceiling for AlphaZero,
link |
meaning it could just arbitrarily keep improving.
link |
Now in the realm of chess, that doesn't matter to us,
link |
that it's like, it just ran away with the game of chess.
link |
Like it's like just so much better than humans.
link |
But the question is if you can create that in the realm
link |
that does have a bigger, deeper effect on human beings,
link |
on societies, that can be a terrifying process.
link |
To me, it's an exciting process
link |
if you supervise it correctly,
link |
if you inject what's called value alignment,
link |
you make sure that the goals that the AI is optimizing
link |
is aligned with human beings and human societies.
link |
There's a lot of fascinating things to talk about
link |
within the specifics of neural networks
link |
and all the problems that people are working on.
link |
But I would say the really big exciting one
link |
is self-supervised learning,
link |
where trying to get less and less human supervision,
link |
less and less human supervision of neural networks.
link |
And also just to comment and I'll shut up.
link |
No, please keep going.
link |
I have questions, but I'm learning.
link |
So please keep going.
link |
So to me, what's exciting is not the theory,
link |
it's always the application.
link |
One of the most exciting applications
link |
of artificial intelligence,
link |
specifically neural networks and machine learning
link |
is Tesla Autopilot.
link |
So these are systems that are working in the real world.
link |
This isn't an academic exercise.
link |
This is human lives at stake.
link |
This is safety critical.
link |
These are automated vehicles, autonomous vehicles.
link |
Semi-autonomous, we want to be.
link |
We've gone through wars on these topics.
link |
So even though it's called FSD, full self-driving,
link |
it is currently not fully autonomous,
link |
meaning human supervision is required.
link |
So human is tasked with overseeing the systems.
link |
In fact, liability wise, the human is always responsible.
link |
This is a human factor psychology question,
link |
which is fascinating.
link |
I'm fascinated by the whole space,
link |
which is a whole nother space of human robot interaction.
link |
When AI systems and humans work together
link |
to accomplish tasks.
link |
That dance to me is one of the smaller communities,
link |
but I think it will be one of the most important
link |
open problems once they're solved,
link |
is how do humans and robots dance together?
link |
To me, semi-autonomous driving is one of those spaces.
link |
So for Elon, for example, he doesn't see it that way.
link |
He sees semi-autonomous driving as a stepping stone
link |
towards fully autonomous driving.
link |
Like humans and robots can't dance well together.
link |
Like humans and humans dance and robots and robots dance.
link |
Like we need to, this is an engineering problem.
link |
We need to design a perfect robot that solves this problem.
link |
To me forever, maybe this is not the case with driving,
link |
but the world is going to be full of problems
link |
where it's always humans and robots have to interact
link |
because I think robots will always be flawed,
link |
just like humans are going to be flawed, are flawed.
link |
And that's what makes life beautiful, that they're flawed.
link |
That's where learning happens at the edge
link |
of your capabilities.
link |
So you always have to figure out how can flawed robots
link |
and flawed humans interact together,
link |
such that they, like the sum is bigger than the whole,
link |
as opposed to focusing on just building the perfect robot.
link |
So that's one of the most exciting applications,
link |
I would say, of artificial intelligence to me,
link |
is autonomous driving and semi-autonomous driving.
link |
And that's a really good example of machine learning
link |
because those systems are constantly learning.
link |
And there's a process there that maybe I can comment on.
link |
Andre Karpathy, who's the head of Autopilot,
link |
calls it the data engine.
link |
And this process applies for a lot of machine learning,
link |
which is you build a system
link |
that's pretty good at doing stuff.
link |
You send it out into the real world.
link |
It starts doing the stuff and then it runs
link |
into what are called edge cases, like failure cases,
link |
where it screws up.
link |
You know, we do this as kids that, you know, you have-
link |
We do this as adults.
link |
We do this as adults, exactly.
link |
But we learn really quickly.
link |
But the whole point,
link |
and this is the fascinating thing about driving,
link |
is you realize there's millions of edge cases.
link |
There's just like weird situations that you did not expect.
link |
And so the data engine process is
link |
you collect those edge cases,
link |
and then you go back to the drawing board
link |
and learn from them.
link |
And so you have to create this data pipeline
link |
where all these cars,
link |
hundreds of thousands of cars as you're driving around,
link |
and something weird happens.
link |
And so whenever this weird detector fires,
link |
it's another important concept,
link |
that piece of data goes back to the mothership
link |
for the training, for the retraining of the system.
link |
And through this data engine process,
link |
it keeps improving and getting better and better
link |
and better and better.
link |
So basically, you send out a pretty clever AI systems
link |
out into the world and let it find the edge cases.
link |
Let it screw up just enough
link |
to figure out where the edge cases are,
link |
and then go back and learn from them,
link |
and then send out that new version
link |
and keep updating that version.
link |
Is the updating done by humans?
link |
The annotation is done by humans.
link |
the weird examples come back, the edge cases,
link |
and you have to label what actually happened in there.
link |
There's also some mechanisms for automatically labeling,
link |
but mostly I think you always have to rely on humans
link |
to improve, to understand what's happening
link |
in the weird cases.
link |
And then there's a lot of debate.
link |
And that's the other thing,
link |
what is artificial intelligence?
link |
Which is a bunch of smart people
link |
having very different opinions about what is intelligence.
link |
So AI is basically a community of people
link |
who don't agree on anything.
link |
It seems to be the case.
link |
And first of all, this is a beautiful description of terms
link |
that I've heard many times among my colleagues at Stanford,
link |
at meetings, in the outside world.
link |
And there's so many fascinating things.
link |
I have so many questions,
link |
but I do want to ask one question about the culture of AI,
link |
because it does seem to be a community where,
link |
at least as an outsider,
link |
where it seems like there's very little consensus
link |
about what the terms
link |
and the operational definitions even mean.
link |
And there seems to be a lot of splitting happening now
link |
of not just supervised and unsupervised learning,
link |
but these sort of intermediate conditions
link |
where machines are autonomous,
link |
but then go back for more instruction,
link |
like kids go home from college during the summer
link |
and get a little, mom still feeds them,
link |
then eventually they leave the nest kind of thing.
link |
Is there something in particular about engineers
link |
or about people in this realm of engineering
link |
that you think lends itself to disagreement?
link |
Yeah, I think, so first of all, the more specific you get,
link |
the less disagreement there is.
link |
So there's a lot of disagreement
link |
about what is artificial intelligence,
link |
but there's less disagreement about what is machine learning
link |
and even less when you talk about active learning
link |
or machine teaching or self-supervised learning.
link |
And then when you get into NLP language models
link |
when you get into specific neural network architectures,
link |
there's less and less and less disagreement
link |
about those terms.
link |
So you might be hearing the disagreement
link |
from the high level terms,
link |
and that has to do with the fact that engineering,
link |
especially when you're talking about intelligent systems
link |
is a little bit of an art and a science.
link |
So the art part is the thing that creates disagreements
link |
because then you start having disagreements
link |
about how easy or difficult a particular problem is.
link |
For example, a lot of people disagree with Elon,
link |
how difficult the problem of autonomous driving is.
link |
And so, but nobody knows.
link |
So there's a lot of disagreement
link |
about what are the limits of these techniques.
link |
And through that, the terminology also contains within it,
link |
the disagreements.
link |
But overall, I think it's also a young science
link |
that also has to do with that.
link |
So like, it's not just engineering,
link |
it's that artificial intelligence truly
link |
as a large scale discipline where it's thousands,
link |
tens of thousands, hundreds of thousands of people
link |
working on it, huge amounts of money being made.
link |
That's a very recent thing.
link |
So we're trying to figure out those terms.
link |
And of course there's egos and personalities
link |
and a lot of fame to be made,
link |
like the term deep learning, for example.
link |
Neural networks have been around for many, many decades,
link |
since the sixties, you can argue since the forties.
link |
So there was a rebranding of neural networks
link |
into the word deep learning, the term deep learning
link |
that was part of the reinvigoration of the field.
link |
But it's really the same exact thing.
link |
I didn't know that.
link |
I mean, I grew up in the age of neuroscience
link |
when neural networks were discussed.
link |
Computational neuroscience and theoretical neuroscience,
link |
they had their own journals.
link |
It wasn't actually taken terribly seriously
link |
by experimentalists until a few years ago.
link |
I would say about five to seven years ago,
link |
excellent theoretical neuroscientists like Larry Abbott
link |
and other colleagues, certainly at Stanford as well,
link |
that people started paying attention
link |
to computational methods.
link |
But these terms, neural networks, computational methods,
link |
I actually didn't know that neural network works
link |
in deep learning where those have now become
link |
kind of synonymous.
link |
No, they were always, no, they're always the same thing.
link |
I'm a neuroscientist and I didn't know that.
link |
So, well, because neural networks probably mean
link |
something else in neuroscience, not something else,
link |
but a little different flavor depending on the field.
link |
And that's fascinating too, because neuroscience and AI,
link |
people have started working together
link |
and dancing a lot more in the recent,
link |
I would say probably decade.
link |
Oh, machines are going into the brain.
link |
I have a couple of questions,
link |
but one thing that I'm sort of fixated on
link |
that I find incredibly interesting is this example you gave
link |
of playing a game with a mutated version of yourself
link |
Yeah, I find that incredibly interesting
link |
as a kind of a parallel or a mirror for what happens
link |
when we try and learn as humans,
link |
which is we generate repetitions of whatever it is
link |
we're trying to learn and we make errors.
link |
Occasionally we succeed.
link |
In a simple example, for instance,
link |
of trying to throw bull's eyes on a dartboard.
link |
I'm going to have errors, errors, errors.
link |
I'll probably miss the dartboard
link |
and maybe occasionally hit a bull's eye.
link |
And I don't know exactly what I just did, right?
link |
But then let's say I was playing darts
link |
against a version of myself
link |
where I was wearing a visual prism,
link |
like my visual, I had a visual defect.
link |
You learn certain things in that mode as well.
link |
You're saying that a machine can sort of mutate itself.
link |
Does the mutation always cause a deficiency
link |
that it needs to overcome?
link |
Because mutations in biology
link |
sometimes give us superpowers, right?
link |
Occasionally you'll get somebody
link |
who has better than 20, 20 vision
link |
and they can see better than 99.9% of people out there.
link |
So when you talk about a machine playing a game
link |
against a mutated version of itself,
link |
is the mutation always what we call a negative mutation
link |
or an adaptive or a maladaptive mutation?
link |
No, you don't know until you mutate first
link |
and then figure out and they compete against each other.
link |
So you're evolving,
link |
the machine gets to evolve itself in real time.
link |
Yeah, and I think of it, which would be exciting,
link |
if you could actually do with humans,
link |
it's not just, so usually you freeze a version
link |
So really you take an Andrew of yesterday
link |
and you make 10 clones of them
link |
and then maybe you mutate, maybe not.
link |
And then you do a bunch of competitions
link |
of the Andrew of today.
link |
Like you fight to the death and who wins last.
link |
So I love that idea of like creating
link |
a bunch of clones of myself from each of the day
link |
for the past year and just seeing who's going to be better
link |
at like podcasting or science or picking up chicks at a bar
link |
or I don't know, or competing in jujitsu.
link |
That's one way to do it.
link |
I mean, a lot of Lexes would have to die for that process,
link |
but that's essentially what happens
link |
is in reinforcement learning
link |
through the self-play mechanisms,
link |
it's a graveyard of systems that didn't do that well.
link |
And the surviving, the good ones survive.
link |
Do you think that, I mean, Darwin's theory of evolution
link |
might have worked in some sense in this way,
link |
but at the population level.
link |
I mean, you get a bunch of birds with different shaped beaks
link |
and some birds have the shape beak
link |
that allows them to get the seeds.
link |
I mean, it's a trivially simple example
link |
of Darwinian evolution, but I think it's correct
link |
even though it's not exhaustive.
link |
Is that what you're referring to?
link |
You essentially that normally this is done
link |
between members of a different species.
link |
Lots of different members of species have different traits
link |
and some get selected for,
link |
but you could actually create multiple versions of yourself
link |
with different traits.
link |
So with, I should probably have said this,
link |
but perhaps it's implied,
link |
but the machine learning or the reinforcement learning
link |
through these processes,
link |
one of the big requirements
link |
is to have an objective function, a loss function,
link |
a utility function.
link |
Those are all different terms for the same thing.
link |
Is there's like an equation that says what's good.
link |
And then you're trying to optimize that equation.
link |
So there's a clear goal for these systems.
link |
Because it's a game, like with chess, there's a goal.
link |
But for anything, anything you want machine learning
link |
to solve, there needs to be an objective function.
link |
In machine learning, it's usually called loss function
link |
that you're optimizing.
link |
The interesting thing about evolution,
link |
complicated of course,
link |
but the goal also seems to be evolving.
link |
Like it's a, I guess adaptation to the environment
link |
is the goal, but it's unclear.
link |
You can convert that always.
link |
It's like survival of the fittest.
link |
It's unclear what the fittest is.
link |
In machine learning, the starting point,
link |
and this is like what human ingenuity provides,
link |
is that fitness function of what's good and what's bad,
link |
which it lets you know which of the systems is going to win.
link |
So you need to have a equation like that.
link |
One of the fascinating things about humans
link |
is we figure out objective functions for ourselves.
link |
Like we're, it's the meaning of life.
link |
Like why the hell are we here?
link |
And a machine currently has to have
link |
a hard-coded statement about why.
link |
It has to have a meaning of
link |
artificial intelligence-based life.
link |
So like there's a lot of interesting explorations
link |
about that function being more about curiosity,
link |
about learning new things and all that kind of stuff,
link |
but it's still hard-coded.
link |
If you want a machine to be able to be good at stuff,
link |
it has to be given very clear statements
link |
of what good at stuff means.
link |
That's one of the challenges of artificial intelligence
link |
is you have to formalize the,
link |
in order to solve a problem, you have to formalize it
link |
and you have to provide
link |
both like the full sensory information.
link |
You have to be very clear about
link |
what is the data that's being collected.
link |
And you have to also be clear about the objective function.
link |
What is the goal that you're trying to reach?
link |
And that's a very difficult thing
link |
for artificial intelligence.
link |
I love that you mentioned curiosity.
link |
I'm sure this definition falls short in many ways,
link |
but I define curiosity as a strong interest
link |
in knowing something,
link |
but without an attachment to the outcome.
link |
You know, it's sort of a,
link |
it's not, it could be a random search,
link |
but there's not really an emotional attachment.
link |
It's really just a desire to discover
link |
and unveil what's there
link |
without hoping it's a gold coin under a rock.
link |
You're just looking under rocks.
link |
Is that more or less how the machine,
link |
within machine learning,
link |
it sounds like there are elements of reward prediction
link |
and rewards the machine has to know
link |
when it's done the right thing.
link |
So can you make machines that are curious
link |
or are the sorts of machines
link |
that you are describing curious by design?
link |
Yeah, curiosity is a kind of a symptom,
link |
So what happens is one of the big trade-offs
link |
in reinforcement learning
link |
is this exploration versus exploitation.
link |
So when you know very little,
link |
it pays off to explore a lot,
link |
like even trajectories that seem like
link |
they're not going to lead anywhere.
link |
That's called exploration.
link |
The smarter and smarter and smarter you get,
link |
the more emphasis you put on exploitation,
link |
meaning you take the best solution,
link |
you take the best path.
link |
Now through that process,
link |
the exploration can look like curiosity by us humans,
link |
but it's really just trying to get out of the local optimal,
link |
the thing that's already discovered.
link |
It's from an AI perspective,
link |
it's always looking to optimize the objective function.
link |
It derives, and we can talk about this a lot more,
link |
but in terms of the tools of machine learning today,
link |
it derives no pleasure from just the curiosity of like,
link |
I don't know, discovery.
link |
So there's no dopamine for a machine.
link |
There's no dopamine.
link |
There's no reward system chemical
link |
or I guess electronic reward system.
link |
That said, if you look at machine learning literature
link |
and reinforcement learning literature,
link |
they will use like deep mind,
link |
we use terms like dopamine.
link |
We're constantly trying to use the human brain
link |
to inspire totally new solutions to these problems.
link |
So they'll think like,
link |
how does dopamine function in the human brain?
link |
And how can that lead to more interesting ways
link |
to discover optimal solutions?
link |
But ultimately, currently,
link |
there has to be a formal objective function.
link |
Now you could argue that humans
link |
also has a set of objective functions
link |
we're trying to optimize.
link |
We're just not able to introspect them.
link |
Yeah, we don't actually know what we're looking for
link |
and seeking and doing.
link |
Well, like Lisa Feldman Barrett,
link |
she's spoken with at least on Instagram.
link |
I hope you get her through you.
link |
Yeah, I hope you actually have her on this podcast.
link |
That'd be terrific.
link |
So she has a very,
link |
it has to do with homeostasis like that.
link |
Basically there's a very dumb objective function
link |
that the brain is trying to optimize,
link |
like to keep like body temperature the same.
link |
Like there's a very dumb
link |
kind of optimization function happening.
link |
And then what we humans do with our fancy consciousness
link |
and cognitive abilities is we tell stories to ourselves
link |
so we can have nice podcasts,
link |
but really it's the brain trying to maintain
link |
just like healthy state, I guess.
link |
That's fascinating.
link |
I also see the human brain
link |
and I hope artificial intelligence systems
link |
as not just systems that solve problems or optimize a goal,
link |
but also storytellers.
link |
I think there's a power to telling stories.
link |
We tell stories to each other.
link |
That's what communication is.
link |
Like when you're alone, that's when you solve problems.
link |
That's when it makes sense to talk about solving problems.
link |
But when you're a community,
link |
the capability to communicate, tell stories,
link |
share ideas in such a way that those ideas are stable
link |
over a long period of time,
link |
that's like, that's being a charismatic storyteller.
link |
And I think both humans are very good at this.
link |
Arguably, I would argue that's why we are who we are
link |
is we're great storytellers.
link |
And then AI, I hope will also become that.
link |
So it's not just about being able to solve problems
link |
with a clear objective function.
link |
It's afterwards be able to tell like a way better,
link |
like make up a way better story
link |
about why you did something or why you failed.
link |
So you think that robots and or machines of some sort
link |
are going to start telling humans stories?
link |
So the technical field for that is called explainable AI,
link |
explainable artificial intelligence
link |
is trying to figure out how you get the AI system
link |
to explain to us humans why the hell it failed
link |
or why it succeeded.
link |
Or there's a lot of different sort of versions of this
link |
or to visualize how it understands the world.
link |
That's a really difficult problem,
link |
especially with neural networks that are famously opaque,
link |
that they, we don't understand in many cases
link |
why a particular neural network does what it does so well.
link |
And to try to figure out where it's going to fail,
link |
that requires the AI to explain itself.
link |
There's a huge amount of money,
link |
like there's a huge amount of money in this,
link |
especially from government funding and so on.
link |
Because if you want to deploy AI systems in the real world,
link |
we humans at least want to ask it a question,
link |
like why the hell did you do that?
link |
Like in a dark way, why did you just kill that person?
link |
Right, like if a car ran over a person,
link |
we wouldn't understand why that happened.
link |
And now again, we're sometimes very unfair to AI systems
link |
because we humans can often not explain why very well.
link |
But that's the field of explainable AI.
link |
That's very, people are very interested in
link |
because the more and more we rely on AI systems,
link |
like the Twitter recommender system, that AI algorithm,
link |
that's I would say impacting elections,
link |
perhaps starting wars or at least military conflict.
link |
That's that algorithm.
link |
We want to ask that algorithm, first of all,
link |
do you know what the hell you're doing?
link |
Do you know, do you understand
link |
the society level effects you're having?
link |
And can you explain the possible other trajectories?
link |
Like we would have that kind of conversation with a human.
link |
We want to be able to do that with an AI.
link |
And on my own personal level,
link |
I think it would be nice to talk to AI systems
link |
for stupid stuff, like robots, when they fail to-
link |
Why'd you fall down the stairs?
link |
Yeah, but not an engineering question,
link |
but almost like a endearing question.
link |
Like I'm looking for, if I fell
link |
and you and I were hanging out,
link |
I don't think you need an explanation
link |
exactly what were the dynamic,
link |
like what was the under actuated system problem here?
link |
Like what was the texture of the floor or so on?
link |
Or like what was the-
link |
I want to know what you're thinking.
link |
That, or you might joke about like,
link |
you're drunk again, go home or something.
link |
Like there could be humor in it.
link |
That's an opportunity,
link |
like storytelling isn't just explanation of what happened.
link |
It's something that makes people laugh,
link |
makes people fall in love,
link |
makes people dream and understand things
link |
in a way that poetry makes people understand things
link |
as opposed to a rigorous log of where every sensor was,
link |
where every actuator was.
link |
I mean, I find this incredible because,
link |
one of the hallmarks of severe autism spectrum disorders
link |
is a report of experience from the autistic person
link |
that is very much a catalog of action steps.
link |
It's like, how do you feel today?
link |
And they'll say, well, I got up and I did this
link |
and then I did this and I did this.
link |
And it's not at all the way that a person with,
link |
who doesn't have autism spectrum disorder would respond.
link |
And the way you describe these machines
link |
has so much humanism or so much of a human
link |
and biological element.
link |
But I realized that we were talking about machines.
link |
I want to make sure that I understand
link |
if there's a distinction between a machine that learns,
link |
a machine with artificial intelligence and a robot.
link |
At what point does a machine become a robot?
link |
So if I have a ballpoint pen,
link |
I'm assuming I wouldn't call that a robot,
link |
but if my ballpoint pen can come to me
link |
when I moved to the opposite side of the table,
link |
if it moves by whatever mechanism,
link |
at that point, does it become a robot?
link |
Okay, there's a million ways to explore this question.
link |
It's a fascinating one.
link |
So first of all, there's a question of what is life?
link |
Like how do you know something is a living form and not?
link |
And it's similar to the question of when does sort of a,
link |
maybe a cold computational system becomes a,
link |
well, we're already loading these words
link |
with a lot of meaning, robot and machine.
link |
But so one, I think movement is important,
link |
but that's kind of a boring idea
link |
that a robot is just a machine
link |
that's able to act in the world.
link |
So one, artificial intelligence could be
link |
both just the thinking thing,
link |
which I think is what machine learning is,
link |
and also the acting thing,
link |
which is what we usually think about robots.
link |
So robots are the things that have a perception system
link |
that's able to take in the world,
link |
however you define the world,
link |
is able to think and learn
link |
and do whatever the hell it does inside
link |
and then act on the world.
link |
So that's the difference between maybe an AI system
link |
or a machine and a robot.
link |
It's something that's able,
link |
a robot is something that's able to perceive the world
link |
and act in the world.
link |
So it could be through language or sound,
link |
or it could be through movement or both.
link |
Yeah, and I think it could also be in the digital space,
link |
as long as there's a aspect of entity
link |
that's inside the machine
link |
and a world that's outside the machine,
link |
and there's a sense in which the machine
link |
is sensing that world and acting in it.
link |
So we could, for instance,
link |
there could be a version of a robot,
link |
according to the definition that I think you're providing,
link |
where the robot, where I go to sleep at night
link |
and this robot goes and forges for information
link |
that it thinks I want to see
link |
loaded onto my desktop in the morning.
link |
There was no movement of that machine.
link |
There was no language,
link |
but it essentially has movement in cyberspace.
link |
Yeah, there's a distinction that I think is important
link |
in that there's an element of it being an entity,
link |
whether it's in the digital or the physical space.
link |
So when you have something like Alexa in your home,
link |
most of the speech recognition,
link |
most of what Alexa is doing
link |
is constantly being sent back to the mothership.
link |
When Alexa is there on its own,
link |
that's, to me, a robot,
link |
when it's there interacting with the world.
link |
When it's simply a finger of the main mothership,
link |
then Alexa is not a robot.
link |
Then it's just an interaction device.
link |
Then maybe the main Amazon Alexa AI,
link |
big, big system is the robot.
link |
So that's important because there's some element
link |
to us humans, I think,
link |
where we want there to be an entity,
link |
whether in the digital or the physical space.
link |
That's where ideas of consciousness come in
link |
and all those kinds of things
link |
that we project our understanding
link |
what it means to be a being.
link |
And so to take that further,
link |
when does a machine become a robot?
link |
I think there's a special moment.
link |
There's a special moment in a person's life,
link |
in a robot's life where it surprises you.
link |
I think surprise is a really powerful thing
link |
where you know how the thing works
link |
and yet it surprises you.
link |
That's a magical moment for us humans.
link |
So whether it's a chess playing program
link |
that does something that you haven't seen before
link |
that makes people smile,
link |
like, huh, those moments happen with alpha zero
link |
for the first time in chess playing
link |
or grandmasters were really surprised by a move.
link |
They didn't understand the move
link |
and then they studied and study
link |
and then they understood it.
link |
But that moment of surprise,
link |
that's for grandmasters in chess.
link |
I find that moment of surprise really powerful,
link |
really magical in just everyday life.
link |
Because it supersedes the human brain in that moment?
link |
Not supersedes like outperforms,
link |
but surprises you in a positive sense.
link |
Like I didn't think you could do that.
link |
I didn't think that you had that in you.
link |
And I think that moment is a big transition for a robot
link |
from a moment of being a servant
link |
that accomplishes a particular task
link |
with some level of accuracy,
link |
with some rate of failure to an entity,
link |
a being that's struggling just like you are in this world.
link |
And that's a really important moment
link |
that I think you're not gonna find many people
link |
in the AI community that talk like I just did.
link |
I'm not speaking like some philosopher or some hippie.
link |
I'm speaking from purely engineering perspective.
link |
I think it's really important for robots to become entities
link |
and explore that as a real engineering problem,
link |
as opposed to everybody treats robots
link |
in the robotics community.
link |
They don't even call them a he or she.
link |
They don't give them, try to avoid giving them names.
link |
They really want to see it like a system, like a servant.
link |
They see it as a servant that's trying to accomplish a task.
link |
To me, I don't think I'm just romanticizing the notion.
link |
I think it's a being.
link |
It's currently perhaps a dumb being,
link |
but in the long arc of history,
link |
humans are pretty dumb beings too, so.
link |
I would agree with that statement.
link |
So I tend to really want to explore
link |
this treating robots really as entities.
link |
So like anthropomorphization,
link |
which is the sort of the act of looking at a inanimate object
link |
and projecting onto it lifelike features,
link |
I think robotics generally sees that as a negative.
link |
I see it as a superpower.
link |
Like that, we need to use that.
link |
Well, I'm struck by how that really grabs on
link |
to the relationship between human and machine
link |
or human and robot.
link |
So it's the simple question is,
link |
and I think you've already told us the answer,
link |
but does interacting with a robot change you?
link |
Does it, in other words, do we develop relationships
link |
Yeah, I definitely think so.
link |
I think the moment you see a robot or AI systems
link |
as more than just servants,
link |
but entities, they begin to change,
link |
just like good friends do, just like relationships,
link |
just like other humans.
link |
I think for that, you have to have certain aspects
link |
of that interaction,
link |
like the robot's ability to say no,
link |
to have its own sense of identity,
link |
to have its own set of goals
link |
that's not constantly serving you,
link |
but instead trying to understand the world
link |
and do that dance of understanding
link |
through communication with you.
link |
So I definitely think there's a,
link |
I mean, I have a lot of thoughts about this, as you may know,
link |
and that's at the core of my lifelong dream, actually,
link |
of what I want to do,
link |
which is I believe that most people have
link |
a notion of loneliness in them that we haven't discovered,
link |
that we haven't explored, I should say.
link |
And I see AI systems as helping us explore that
link |
so that we can become better humans,
link |
better people towards each other.
link |
So I think that connection between human and AI,
link |
human and robot is not only possible,
link |
but will help us understand ourselves
link |
in ways that are like several orders of magnitude,
link |
deeper than we ever could have imagined.
link |
I tend to believe that,
link |
well, I have very wild levels of belief
link |
in terms of how impactful that would be.
link |
All right, so when I think about human relationships,
link |
I don't always break them down into variables,
link |
but we could explore a few of those variables
link |
and see how they map to human-robot relationships.
link |
One is just time, right?
link |
If you spend zero time with another person at all
link |
in cyberspace or on the phone or in person,
link |
you essentially have no relationship to them.
link |
If you spend a lot of time, you have a relationship.
link |
This is obvious, but I guess one variable would be time,
link |
how much time you spend with the other entity,
link |
The other would be wins and successes.
link |
You know, you enjoy successes together.
link |
I'll give an absolutely trivial example of this in a moment,
link |
but the other would be failures.
link |
When you struggle with somebody,
link |
whether or not you struggle between one another,
link |
you disagree, like I was really struck
link |
by the fact that you said that robot's saying no.
link |
I've never thought about a robot saying no to me,
link |
I look forward to you being one of the first people
link |
to send this robot to.
link |
So there's struggle.
link |
You grow, you know, when you struggle with somebody,
link |
Sometimes the struggles are imposed
link |
between those two people, so-called trauma bonding.
link |
They call it in the whole psychology literature
link |
and pop psychology literature.
link |
But in any case, I could imagine,
link |
so time, successes together, struggle together,
link |
and then just peaceful time, hanging out at home,
link |
watching movies, waking up near one another.
link |
Here, we're breaking down the kind of elements
link |
of relationships of any kind.
link |
So do you think that these elements apply
link |
to robot-human relationships?
link |
And if so, then I could see how if the robot
link |
is its own entity and has some autonomy
link |
in terms of how it reacts to you,
link |
it's not just there just to serve you.
link |
It's not just a servant.
link |
It actually has opinions and can tell you
link |
when maybe your thinking is flawed
link |
or your actions are flawed.
link |
It can also leave.
link |
It could also leave.
link |
So I've never conceptualized
link |
robot-human interactions this way.
link |
So tell me more about how this might look.
link |
Are we thinking about a human-appearing robot?
link |
I know you and I have both had intense relationships
link |
to our, we have separate dogs, obviously, but to animals.
link |
This sounds a lot like human-animal interaction.
link |
So what is the ideal human-robot relationship?
link |
So there's a lot to be said here,
link |
but you actually pinpointed one of the big,
link |
big first steps, which is this idea of time.
link |
And it's a huge limitation
link |
in machine learning community currently.
link |
Now we're back to like the actual details.
link |
Lifelong learning is a problem space
link |
that focuses on how AI systems can learn
link |
over a long period of time.
link |
What's currently most machine learning systems
link |
are not able to do is to,
link |
all of the things you've listed under time,
link |
the successes, the failures,
link |
or just chilling together watching movies,
link |
AI systems are not able to do that,
link |
which is all the beautiful, magical moments
link |
that I believe are the days filled with.
link |
They're not able to keep track of those together with you.
link |
Because they can't move with you and be with you.
link |
No, no, like literally we don't have the techniques
link |
to do the learning,
link |
the actual learning of containing those moments.
link |
Current machine learning systems are really focused
link |
on understanding the world in the following way.
link |
It's more like the perception system,
link |
like looking around, understand like what's in the scene,
link |
that there's a bunch of people sitting down,
link |
that there is cameras and microphones,
link |
that there's a table, understand that.
link |
But the fact that we shared this moment of talking today
link |
and still remember that for next time you're,
link |
for like next time you're doing something,
link |
remember that this moment happened.
link |
We don't know how to do that technique wise.
link |
This is what I'm hoping to innovate on
link |
as I think it's a very, very important component
link |
of what it means to create a deep relationship,
link |
that sharing of moments together.
link |
Could you post a photo of you and the robot,
link |
like selfie with robot,
link |
and then the robot sees that image
link |
and recognizes that was time spent,
link |
there were smiles or there were tears,
link |
and create some sort of metric of emotional depth
link |
in the relationship and update its behavior?
link |
So could it text you in the middle of the night and say,
link |
why haven't you texted me back?
link |
Well, yes, all of those things,
link |
but we can dig into that.
link |
But I think that time element, forget everything else,
link |
just sharing moments together, that changes everything.
link |
I believe that changes everything.
link |
There's specific things that are more in terms of systems
link |
that I can explain you.
link |
It's more technical and probably a little bit off line
link |
because I have kind of wild ideas
link |
how that can revolutionize social networks
link |
and operating systems.
link |
But the point is that element alone,
link |
forget all the other things we're talking about,
link |
like emotions, saying no, all that,
link |
just remember sharing moments together
link |
would change everything.
link |
We don't currently have systems
link |
that share moments together.
link |
Like even just you and your fridge,
link |
just all those times you went late at night
link |
and ate the thing you shouldn't have eaten,
link |
that was a secret moment you had with your refrigerator.
link |
You shared that moment,
link |
that darkness or that beautiful moment
link |
where you just like heartbroken for some reason,
link |
you're eating that ice cream or whatever,
link |
that's a special moment.
link |
And that refrigerator was there for you.
link |
And the fact that it missed the opportunity
link |
to remember that is tragic.
link |
And once it does remember that,
link |
I think you're gonna be very attached to the refrigerator.
link |
You're gonna go through some hell with that refrigerator.
link |
Most of us have like in the developed world
link |
have weird relationships with food, right?
link |
So you can go through some deep moments
link |
of trauma and triumph with food.
link |
And at the core of that is the refrigerator.
link |
So a smart refrigerator, I believe would change society,
link |
not just the refrigerator,
link |
but these ideas in the systems all around us.
link |
So I just wanna comment on how powerful
link |
the idea of time is.
link |
And then there's a bunch of elements of actual interaction
link |
of allowing you as a human to feel like you're being heard,
link |
truly heard, truly understood that we human,
link |
like deep friendship is like that, I think,
link |
but we're still, there's still an element of selfishness.
link |
There's still an element of not really being able
link |
to understand another human.
link |
And a lot of the times when you're going through
link |
trauma together through difficult times and through
link |
successes, you're actually starting to get that inkling
link |
of understanding of each other.
link |
But I think that can be done more aggressively,
link |
Like if you think of a great therapist,
link |
I think I've never actually been to a therapist,
link |
but I'm a believer.
link |
I used to want to be a psychiatrist.
link |
Do Russians go to therapists?
link |
No, they don't, they don't.
link |
And if they do the therapist don't live to tell the story.
link |
No, I do believe in talk there,
link |
which friendship is to me is talk therapy or like it's like,
link |
it's you don't necessarily need to talk.
link |
It's like just connecting through in the space of ideas
link |
and the space of experiences.
link |
And I think there's a lot of ideas of how to make AI
link |
systems to be able to ask the right questions
link |
and truly hear another human.
link |
This is what we try to do with podcasting, right?
link |
I think there's ways to do that with AI,
link |
but above all else, just remembering the collection
link |
of moments that make up the day, the week, the months.
link |
I think you maybe have some of this as well.
link |
Some of my closest friends still
link |
are the friends from high school.
link |
That's time we've been through a bunch of shit together.
link |
And that like we're very different people,
link |
but just the fact that we've been through that
link |
and we remember those moments and those moments somehow
link |
create a depth of connection like nothing else,
link |
like you and your refrigerator.
link |
I love that because my graduate advisor,
link |
unfortunately she passed away, but when she passed away,
link |
somebody said at her memorial,
link |
all these amazing things she had done, et cetera.
link |
And then her kids got up there and she had young children
link |
that I knew as they were when she was pregnant with them.
link |
And so it was really, even now I can feel like
link |
your heart gets heavy thinking about this.
link |
They're going to grow up without their mother.
link |
And it was really amazing.
link |
Very, very strong young girls and now young women.
link |
And what they said was incredible.
link |
They said what they really appreciated most
link |
about their mother, who was an amazing person,
link |
is all the unstructured time they spent together.
link |
So it wasn't the trips to the zoo.
link |
It wasn't, oh, she woke up at five in the morning
link |
and drove us to school.
link |
She did all those things too.
link |
She had two hour commute in each direction.
link |
It was incredible, ran a lab, et cetera.
link |
But it was the unstructured time.
link |
So on the passing of their mother,
link |
that's what they remembered was the biggest give
link |
and what bonded them to her was all the time
link |
where they just kind of hung out.
link |
And the way you described the relationship
link |
to a refrigerator is so, I want to say human-like,
link |
but I'm almost reluctant to say that
link |
because what I'm realizing as we're talking
link |
is that what we think of as human-like
link |
might actually be a lower form of relationship.
link |
There may be relationships that are far better
link |
than the sorts of relationships
link |
that we can conceive in our minds right now
link |
based on what these machine relationship interactions
link |
Do I have that right?
link |
I think there's no reason to see machines
link |
as somehow incapable of teaching us something
link |
that's deeply human.
link |
I don't think humans have a monopoly on that.
link |
I think we understand ourselves very poorly
link |
and we need to have the kind of prompting from a machine.
link |
And definitely part of that is just remembering the moments.
link |
Remembering the moments.
link |
I think the unstructured time together,
link |
I wonder if it's quite so unstructured.
link |
That's like calling this podcast unstructured time.
link |
Maybe what they meant was it wasn't a big outing.
link |
There was no specific goal,
link |
but a goal was created through the lack of a goal.
link |
Like where you just hang out
link |
and then you start playing thumb war
link |
and you end up playing thumb war for an hour.
link |
So the structure emerges from lack of structure.
link |
No, but the thing is the moments,
link |
there's something about those times
link |
that create special moments.
link |
And I think that those could be optimized for.
link |
I think we think of like a big outing
link |
as I don't know, going to Six Flags or something,
link |
or some big, the Grand Canyon or go into some,
link |
I don't know, I think we would need to,
link |
we don't quite yet understand as humans
link |
what creates magical moments.
link |
I think there's possible to optimize a lot of those things.
link |
And perhaps like podcasting is helping people discover that
link |
like maybe the thing we want to optimize for
link |
isn't necessarily like some sexy, like quick clips.
link |
Maybe what we want is long form authenticity, depth, depth.
link |
So we were trying to figure that out,
link |
certainly from a deep connection between humans
link |
and humans and AI systems, I think long conversations
link |
or long periods of communication over a series of moments
link |
like my new, perhaps seemingly insignificant
link |
to the big ones, the big successes,
link |
the big failures, those are all,
link |
just stitching those together and talking throughout.
link |
I think that's the formula
link |
for a really, really deep connection
link |
that from like a very specific engineering perspective
link |
is I think a fascinating open problem
link |
that hasn't been really worked on very much.
link |
And for me, from a, if I have the guts
link |
and I mean, there's a lot of things to say,
link |
but one of it is guts as I'll build a startup around it.
link |
Yeah, so let's talk about this startup
link |
and let's talk about the dream.
link |
You've mentioned this dream before
link |
in our previous conversations,
link |
always as little hints dropped here and there,
link |
just for anyone listening,
link |
there's never been an offline conversation about this dream.
link |
I'm not privy to anything except what Lex says now.
link |
And I realized that there's no way
link |
to capture the full essence of a dream
link |
in any kind of verbal statement in a way
link |
that captures all of it.
link |
But what is this dream that you've referred to now
link |
several times when we've sat down together
link |
and talked on the phone?
link |
Maybe it's this company, maybe it's something distinct.
link |
If you feel comfortable, it'd be great
link |
if you could share a little bit about what that is.
link |
Sure, so the way people express long-term vision,
link |
I've noticed is quite different.
link |
Like Elon is an example of somebody who can very crisply
link |
say exactly what the goal is.
link |
Also has to do with the fact the problems he's solving
link |
have nothing to do with humans.
link |
So my long-term vision is a little bit more difficult
link |
to express in words, I've noticed, as I've tried.
link |
It could be my brain's failure.
link |
But there's a ways to sneak up to it.
link |
So let me just say a few things.
link |
Early on in life, and also in the recent years,
link |
I've interacted with a few robots
link |
where I understood there's magic there.
link |
And that magic could be shared by millions
link |
if it's brought to light.
link |
When I first met Spot from Boston Dynamics,
link |
I realized there's magic there that nobody else is seeing.
link |
Is the dog, sorry.
link |
The Spot is the four-legged robot from Boston Dynamics.
link |
Some people might have seen it, it's this yellow dog.
link |
And sometimes in life, you just notice something
link |
that just grabs you.
link |
And I believe that this is something that,
link |
this magic is something that could be
link |
every single device in the world.
link |
The way that I think maybe Steve Jobs
link |
thought about the personal computer.
link |
Woz didn't think about the personal computer this way,
link |
Which is like, he thought that the personal computer
link |
should be as thin as a sheet of paper
link |
and everybody should have one.
link |
I mean, this idea, I think it is heartbreaking
link |
that we're getting, the world is being filled up
link |
with machines that are soulless.
link |
And I think every one of them can have that same magic.
link |
One of the things that also inspired me
link |
in terms of a startup is that magic can be engineered
link |
much easier than I thought.
link |
That's my intuition with everything I've ever built
link |
So the dream is to add a bit of that magic
link |
in every single computing system in the world.
link |
So the way that Windows operating system for a long time
link |
was the primary operating system everybody interacted with.
link |
They built apps on top of it.
link |
I think this is something that should be as a layer.
link |
It's almost as an operating system
link |
in every device that humans interacted with in the world.
link |
Now what that actually looks like,
link |
the actual dream when I was especially a kid,
link |
it didn't have this concrete form of a business.
link |
It had more of a dream of exploring your own loneliness
link |
by interacting with machines, robots.
link |
This deep connection between humans and robots
link |
was always a dream.
link |
And so for me, I'd love to see a world
link |
where there's every home has a robot
link |
and not a robot that washes the dishes or a sex robot,
link |
or I don't know, I think of any kind of activity
link |
the robot can do, but more like a companion.
link |
A family member, the way a dog is,
link |
but a dog that's able to speak your language too.
link |
So not just connect the way a dog does
link |
by looking at you and looking away
link |
and almost like smiling with its soul in that kind of way,
link |
but also to actually understand what the hell,
link |
like why are you so excited about the successes?
link |
Like understand the details, understand the traumas.
link |
And I just think that has always filled me
link |
with the excitement that I could,
link |
with artificial intelligence, bring joy to a lot of people.
link |
More recently, I've been more and more
link |
heartbroken to see the kind of division, derision,
link |
even hate that's boiling up on the internet
link |
through social networks.
link |
And I thought this kind of mechanism is exactly applicable
link |
in the context of social networks as well.
link |
So it's an operating system that serves
link |
as your guide on the internet.
link |
One of the biggest problems with YouTube
link |
and social networks currently
link |
is they're optimizing for engagement.
link |
I think if you create AI systems
link |
that know each individual person,
link |
you're able to optimize for long-term growth,
link |
for long-term happiness.
link |
Of the individual?
link |
Of the individual, of the individual.
link |
And there's a lot of other things to say, which is the,
link |
in order for AI systems to learn everything about you,
link |
they need to collect, they need to,
link |
just like you and I, when we talk offline,
link |
we're collecting data about each other,
link |
secrets about each other.
link |
The same way AI has to do that.
link |
And that allows you to,
link |
and that requires you to rethink ideas of ownership of data.
link |
I think each individual should own all of their data
link |
and very easily be able to leave.
link |
Just like AI systems can leave,
link |
humans can disappear and delete all of their data
link |
in a moment's notice,
link |
which is actually better than we humans can do.
link |
Once we load the data into each other, it's there.
link |
I think it's very important to be both,
link |
give people complete control over their data
link |
in order to establish trust that they can trust you.
link |
And the second part of trust is transparency.
link |
Whenever the data is used to make it very clear
link |
what it's being used for.
link |
And not clear in a lawyerly legal sense,
link |
but clear in a way that people really understand
link |
what it's used for.
link |
I believe when people have the ability
link |
to delete all their data and walk away
link |
and know how the data is being used, I think they'll stay.
link |
The possibility of a clean breakup
link |
is actually what will keep people together.
link |
I think, yeah, exactly.
link |
I think a happy marriage requires the ability
link |
to divorce easily without the divorce industrial complex
link |
or whatever is currently going on.
link |
There's so much money to be made from lawyers and divorce.
link |
But yeah, the ability to leave is what enables love,
link |
It's interesting, I've heard the phrase
link |
from a semi-cynical friend
link |
that marriage is the leading cause of divorce.
link |
But now we've heard that divorce
link |
or the possibility of divorce
link |
could be the leading cause of marriage.
link |
Of a happy marriage.
link |
Of a happy marriage.
link |
So yeah, but there's a lot of details there.
link |
But the big dream is that connection
link |
between AI system and a human.
link |
there's so much fear
link |
about artificial intelligence systems and about robots
link |
that I haven't quite found the right words
link |
to express that vision
link |
because the vision I have is one,
link |
it's not like some naive delusional vision
link |
of technology is gonna save everybody.
link |
I really do just have a positive view
link |
of ways AI systems can help humans explore themselves.
link |
I love that positivity and I agree
link |
that the stance everything is doomed is equally bad
link |
to say that everything's gonna turn out all right.
link |
There has to be a dedicated effort.
link |
And clearly you're thinking
link |
about what that dedicated effort would look like.
link |
You mentioned two aspects to this dream.
link |
And I wanna make sure that I understand
link |
where they connect if they do
link |
or if these are independent streams.
link |
One was this hypothetical robot family member
link |
or some other form of robot
link |
that would allow people to experience
link |
the kind of delight that you experienced many times
link |
and that you would like the world to be able to have.
link |
And it's such a beautiful idea of this give.
link |
And the other is social media
link |
or social network platforms
link |
that really serve individuals
link |
and their best selves and their happiness and their growth.
link |
Is there crossover between those
link |
or are these two parallel dreams?
link |
It's 100% the same thing.
link |
It's difficult to kind of explain
link |
without going through details,
link |
but maybe one easy way to explain
link |
the way I think about social networks
link |
is to create an AI system that's yours.
link |
It's not like Amazon Alexa that's centralized.
link |
It's like your little friend
link |
that becomes your representative on Twitter
link |
that helps you find things that will make you feel good,
link |
that will also challenge your thinking to make you grow,
link |
but not let you get lost in the negative spiral of dopamine
link |
that gets you to be angry
link |
or most just get you to be not open to learning.
link |
And so that little representative
link |
is optimizing your long-term health.
link |
And I believe that that is not only good for human beings,
link |
it's also good for business.
link |
I think long-term you can make a lot of money
link |
by challenging this idea
link |
that the only way to make money is maximizing engagement.
link |
And one of the things that people disagree with me on
link |
is they think Twitter's always going to win.
link |
Like maximizing engagement is always going to win.
link |
I think people have woken up now to understanding
link |
that they don't always feel good,
link |
the ones who are on Twitter a lot,
link |
that they don't always feel good at the end of the week.
link |
I would love feedback from whatever this creature,
link |
whatever, I can't, I don't know what to call it,
link |
as to maybe at the end of the week,
link |
it would automatically unfollow
link |
some of the people that I follow
link |
because it realized through some really smart data
link |
about how I was feeling inside
link |
or how I was sleeping or something
link |
that that just wasn't good for me,
link |
but it might also put things and people in front of me
link |
that I ought to see.
link |
Is that kind of a sliver of what this looks like?
link |
The whole point, because of the interaction,
link |
because of sharing the moments
link |
and learning a lot about you,
link |
you're now able to understand
link |
what interactions led you to become
link |
a better version of yourself,
link |
like the person you yourself are happy with.
link |
This isn't, if you're into flat earth
link |
and you feel very good about it,
link |
that you believe the earth is flat,
link |
like the idea that you should censor, that is ridiculous.
link |
If it makes you feel good
link |
and you're becoming the best version of yourself,
link |
I think you should be getting
link |
as much flat earth as possible.
link |
Now, it's also good to challenge your ideas,
link |
but not because the centralized committee decided,
link |
but because you tell to the system
link |
that you like challenging your ideas.
link |
I think all of us do.
link |
And then, which actually YouTube doesn't do that well,
link |
once you go down the flat earth rabbit hole,
link |
that's all you're gonna see.
link |
It's nice to get some really powerful communicators
link |
to argue against flat earth.
link |
And it's nice to see that for you
link |
and potentially at least long-term to expand your horizons,
link |
maybe the earth is not flat.
link |
But if you continue to live your whole life
link |
thinking the earth is flat, I think,
link |
and you're being a good father or son or daughter,
link |
and like you're being the best version of yourself
link |
and you're happy with yourself, I think the earth is flat.
link |
So like, I think this kind of idea,
link |
and I'm just using that kind of silly, ridiculous example,
link |
because I don't like the idea of centralized forces
link |
controlling what you can and can't see.
link |
But I also don't like this idea of like,
link |
not censoring anything,
link |
because that's always the biggest problem with that
link |
is there's a central decider.
link |
I think you yourself can decide what you wanna see and not.
link |
And it's good to have a companion that reminds you
link |
that you felt shitty last time you did this,
link |
or you felt good last time you did this.
link |
I mean, I feel like in every good story,
link |
there's a guide or a companion that flies out
link |
or forages a little bit further or a little bit differently
link |
and brings back information that helps us,
link |
or at least tries to steer us in the right direction.
link |
So that's exactly what I'm thinking
link |
and what I've been working on.
link |
I should mention there's a bunch of difficulties here.
link |
You see me up and down a little bit recently.
link |
So there's technically a lot of challenges here.
link |
Like with a lot of technologies,
link |
and the reason I'm talking about it on a podcast comfortably
link |
as opposed to working in secret is it's really hard.
link |
And maybe it's time has not come.
link |
And that's something you have to constantly struggle with
link |
in terms of like entrepreneurially as a startup.
link |
Like I've also mentioned to you maybe offline,
link |
I really don't care about money.
link |
I don't care about business success,
link |
all those kinds of things.
link |
So it's a difficult decision to make
link |
how much of your time do you want to go all in here
link |
and give everything to this?
link |
It's a big roll of the dice because I've also realized
link |
that working on some of these problems,
link |
both with the robotics and the technical side
link |
in terms of the machine learning system
link |
that I'm describing, it's lonely, it's really lonely
link |
because both on a personal level and a technical level.
link |
So on the technical level,
link |
I'm surrounded by people that kind of doubt me,
link |
which I think all entrepreneurs go through.
link |
And they doubt you in the following sense.
link |
They know how difficult it is.
link |
Like the people that, the colleagues of mine,
link |
they know how difficult lifelong learning is.
link |
They also know how difficult it is
link |
to build a system like this,
link |
to build a competitive social network.
link |
And in general, there's a kind of a loneliness
link |
to just working on something on your own
link |
for long periods of time.
link |
And you start to doubt whether,
link |
given that you don't have a track record of success,
link |
like that's a big one.
link |
When you look in the mirror, especially when you're young,
link |
but I still have that on most things,
link |
you look in the mirror and you have these big dreams.
link |
How do you know you're actually as smart
link |
as you think you are?
link |
Like, how do you know you're going to be able
link |
to accomplish this dream?
link |
You have this ambition.
link |
You sort of don't, but you're kind of pulling on a string
link |
hoping that there's a bigger ball of yarn.
link |
Yeah, but you have this kind of intuition.
link |
I think I pride myself in knowing what I'm good at
link |
because the reason I have that intuition
link |
is because I think I'm very good at knowing
link |
all the things I suck at, which is basically everything.
link |
So like, whenever I notice like, wait a minute,
link |
I'm kind of good at this, which is very rare for me.
link |
I think like that might be a ball of yarn worth pulling at.
link |
And the thing with, in terms of engineering systems
link |
that are able to interact with humans,
link |
I think I'm very good at that.
link |
And it's because we talk about podcasting and so on.
link |
I don't know if I'm very good at podcasting.
link |
You're very good at podcasting.
link |
But I certainly don't.
link |
I think maybe it is compelling for people
link |
to watch a kindhearted idiot struggle with this form.
link |
Maybe that's what's compelling.
link |
But in terms of like actual being a good engineer
link |
of human-robot interaction systems, I think I'm good.
link |
But it's hard to know until you do it
link |
and then the world keeps telling you you're not.
link |
And it's just, it's full of doubt and it's really hard.
link |
And I've been struggling with that recently.
link |
It's kind of a fascinating struggle.
link |
But then that's where the Goggins thing comes in,
link |
is like, aside from the stay hard motherfucker,
link |
is the like, whenever you're struggling,
link |
that's a good sign that if you keep going,
link |
that you're going to be alone in the success, right?
link |
Well, in your case, however, I agree.
link |
And actually David had a post recently
link |
that I thought was among his many brilliant posts
link |
was one of the more brilliant about how,
link |
he talked about this myth of the light
link |
at the end of the tunnel.
link |
And instead what he replaced that myth with
link |
was a concept that eventually your eyes adapt to the dark.
link |
That the tunnel, it's not about a light at the end,
link |
that it's really about adapting to the dark of the tunnel.
link |
It's very Goggins-
link |
I love him so much.
link |
Yeah, you guys share a lot in common.
link |
Knowing you both a bit, you share a lot in common.
link |
But in this loneliness and the pursuit of this dream,
link |
it seems to me it has a certain component to it
link |
that is extremely valuable,
link |
which is that the loneliness itself could serve as a driver
link |
to build the companion for the journey.
link |
Well, I'm very deeply aware of that.
link |
So like some people can make,
link |
cause I talk about love a lot.
link |
I really love everything in this world,
link |
but I also love humans, friendship and romantic,
link |
you know, like even the cheesy stuff, just-
link |
You like romantic movies.
link |
Well, I got so much shit from Rogan about like,
link |
what is it, the tango scene from a Scent of a Woman.
link |
But yeah, I find like there's nothing better
link |
than a woman in a red dress.
link |
Like, you know, just like classy.
link |
You should move to Argentina, Mike.
link |
You know, my father's Argentine.
link |
And you know what he said when I went on your podcast
link |
for the first time, he said, he dresses well.
link |
Because in Argentina, the men go to a wedding
link |
or a party or something.
link |
You know, in the U.S. by halfway through the night,
link |
10 minutes in the night, all the jackets are off.
link |
It looks like everyone's undressing for the party
link |
they just got dressed up for.
link |
And he said, you know, I like the way he dresses.
link |
And then when I started, he was talking about you.
link |
And then when I started my podcast, he said,
link |
why don't you wear a real suit like your friend Lex?
link |
But let's talk about this pursuit just a bit more.
link |
Because I think what you're talking about
link |
is building a, not just a solution for loneliness,
link |
but you've alluded to the loneliness
link |
as itself an important thing.
link |
And I think you're right.
link |
I think within people, there's like caverns of thoughts
link |
and shame, but also just the desire to be,
link |
to have resonance, to be seen and heard.
link |
And I don't even know that it's seen
link |
and heard through language.
link |
But these reservoirs of loneliness, I think they're,
link |
well, they're interesting.
link |
Maybe you could comment a little bit about it
link |
because just as often as you talk about love,
link |
I haven't quantified it,
link |
but it seems that you talk about this loneliness.
link |
And maybe you just would, if you're willing,
link |
you could share a little bit more about that
link |
and what that feels like now in the pursuit
link |
of building this robot-human relationship.
link |
And you've been, let me be direct.
link |
You've been spending a lot of time on building
link |
a robot-human relationship.
link |
Oh, well, in terms of business and in terms of systems.
link |
No, I'm talking about a specific robot.
link |
Oh, so, okay, I should mention a few things.
link |
So one is there's a startup where there's an idea
link |
where I hope millions of people can use.
link |
And then there's my own personal,
link |
almost like Frankenstein explorations
link |
with particular robots.
link |
So I'm very fascinated with the legged robots
link |
in my own private sounds like dark,
link |
but like N of one experiments
link |
to see if I can recreate the magic.
link |
And that's been, I have a lot of really good already,
link |
perception systems and control systems
link |
that are able to communicate affection
link |
in a dog-like fashion.
link |
So I'm in a really good place there.
link |
The stumbling blocks,
link |
which also have been part of my sadness recently
link |
is that I also have to work with robotics companies
link |
that I gave so much of my heart, soul,
link |
and love and appreciation towards Boston Dynamics.
link |
But Boston Dynamics is also as a company
link |
that has to make a lot of money
link |
and they have marketing teams.
link |
And they're like looking at this silly Russian kid
link |
in a suit and tie.
link |
It's like, what's he trying to do with all this love
link |
and robot interaction and dancing and so on?
link |
So there was, I think, let's say for now,
link |
it's like when you break up with a girlfriend or something.
link |
Right now we decided to part ways on this particular thing.
link |
They're huge supporters of mine, they're huge fans.
link |
But on this particular thing,
link |
Boston Dynamics is not focusing on
link |
or interested in human robot interaction.
link |
In fact, their whole business currently
link |
is keep the robot as far away from humans as possible.
link |
Because it's in the industrial setting
link |
where it's doing monitoring in dangerous environments.
link |
It's almost like a remote security camera
link |
essentially is its application.
link |
To me, I thought it's still even in those applications
link |
exceptionally useful for the robot
link |
to be able to perceive humans, like see humans
link |
and to be able to, in a big map,
link |
localize where those humans are and have human intention.
link |
For example, like this,
link |
I did this a lot of work with pedestrians
link |
for a robot to be able to anticipate
link |
what the hell the human is doing, like where it's walking.
link |
Humans are not ballistics object.
link |
They're not, just because you're walking this way
link |
one moment doesn't mean you'll keep walking that direction.
link |
You have to infer a lot of signals,
link |
especially the head movement and the eye movement.
link |
So I thought that's super interesting to explore,
link |
but they didn't feel that.
link |
So I'll be working with a few other robotics companies
link |
that are much more open to that kind of stuff.
link |
And they're super excited and fans of mine.
link |
And hopefully Boston Dynamics, my first love
link |
that getting back with an ex-girlfriend will come around.
link |
But so the algorithmically it's,
link |
I'm basically done there.
link |
The rest is actually getting
link |
some of these companies to work with.
link |
And then there's, for people who'd work with robots know
link |
that one thing is to write software that works.
link |
And the other is to have a real machine that actually works.
link |
And it breaks down in all kinds of different ways
link |
that are fascinating.
link |
And so there's a big challenge there.
link |
But that's almost, it may sound a little bit confusing
link |
in the context of our previous discussion,
link |
because the previous discussion was more
link |
about the big dream, how I hoped to have millions of people
link |
enjoy this moment of magic.
link |
This current discussion about a robot
link |
is something I personally really enjoy.
link |
It just brings me happiness.
link |
I really try to do now everything that just brings me joy.
link |
I maximize that because robots are awesome.
link |
But two, given my like a little bit growing platform,
link |
I want to use the opportunity to educate people.
link |
It's just like robots are cool.
link |
And if I think they're cool, I'll be able to,
link |
I hope be able to communicate why they're cool to others.
link |
So this little robot experiment
link |
is a little bit of a research project too.
link |
There's a couple of publications with MIT folks around that.
link |
But the other is just to make some cool videos
link |
and explain to people how they actually work.
link |
And as opposed to people being scared of robots,
link |
they can still be scared, but also excited.
link |
Like see the dark side, the beautiful side,
link |
the magic of what it means to bring, you know,
link |
for a machine to become a robot.
link |
I want to inspire people with that, but that's less,
link |
it's interesting because I think the big impact
link |
in terms of the dream does not have to do with embodied AI.
link |
So it does not need to have a body.
link |
I think the refrigerator is enough that for an AI system,
link |
just to have a voice and to hear you,
link |
that's enough for loneliness.
link |
The embodiment is just-
link |
By embodiment, you mean the physical structure.
link |
Physical instantiation of intelligence.
link |
So it's a legged robot or even just a thing.
link |
I have a few other humanoid robot,
link |
a little humanoid robot, maybe I'll keep them on the table.
link |
Just like walks around or even just like a mobile platform
link |
that can just like turn around and look at you.
link |
It's like we mentioned with the pen,
link |
something that moves and can look at you.
link |
It's like that butter robot that asks, what is my purpose?
link |
That is really, it's almost like art.
link |
There's something about a physical entity that moves around,
link |
that's able to look at you and interact with you,
link |
that makes you wonder what it means to be human.
link |
It like challenges you to think,
link |
if that thing looks like it has consciousness,
link |
what the hell am I?
link |
And I like that feeling.
link |
I think that's really useful for us.
link |
It's humbling for us humans, but that's less about research.
link |
It's certainly less about business
link |
and more about exploring our own selves
link |
and challenging others to think like,
link |
to think about what makes them human.
link |
I love this desire to share the delight
link |
of an interaction with a robot.
link |
And as you describe it,
link |
I actually, I find myself starting to crave that
link |
because we all have those elements from childhood where,
link |
or from adulthood where we experience something
link |
we want other people to feel that.
link |
And I think that you're right.
link |
I think a lot of people are scared of AI.
link |
I think a lot of people are scared of robots.
link |
My only experience of a robotic like thing
link |
is my Roomba vacuum where it goes about,
link |
actually was pretty good at picking up Costello's hair
link |
And then, and I was grateful for it.
link |
But then when I was on a call or something
link |
and it would get caught on a wire or something,
link |
I would find myself getting upset with the Roomba.
link |
In that moment, I'm like, what are you doing?
link |
And obviously it's just doing what it does.
link |
But that's a kind of mostly positive
link |
but slightly negative interaction.
link |
But what you're describing,
link |
it has so much more richness and layers of detail
link |
that I can only imagine what those relationships are like.
link |
Well, there's a few, just a quick comment.
link |
So I've had, they're currently in Boston.
link |
I have a bunch of Roombas for my robot.
link |
And I did this experiment.
link |
Wait, how many Roombas?
link |
Sounds like a fleet of Roombas.
link |
Yeah, so probably seven or eight.
link |
Well, that's a lot of Roombas.
link |
So- This place is very clean.
link |
Well, so this, I'm kind of waiting.
link |
This is the place we're currently in in Austin
link |
is way larger than I need.
link |
But I basically got it to make sure I have room for robots.
link |
So you have these seven or so Roombas.
link |
You deploy all seven at once?
link |
Oh no, I do different experiments with them.
link |
So one of the things I want to mention is this is,
link |
I think there was a YouTube video
link |
that inspired me to try this,
link |
is I got them to scream in pain and moan in pain
link |
whenever they were kicked or contacted.
link |
And I did that experiment to see how I would feel.
link |
I meant to do like a YouTube video on it,
link |
but then it just seemed very cruel.
link |
Did any Roomba rights activists come at you?
link |
Like, I think if I release that video,
link |
I think it's going to make me look insane,
link |
which I know people know I'm already insane.
link |
Now you have to release the video.
link |
I think maybe if I contextualize it
link |
by showing other robots like to show why this is fascinating
link |
because ultimately I felt like they were human
link |
almost immediately.
link |
And that display of pain was what did that.
link |
Giving them a voice.
link |
Giving them a voice, especially a voice of dislike of pain.
link |
I have to connect you to my friend, Eddie Chang.
link |
He studies speech and language.
link |
He's a neurosurgeon and we're lifelong friends.
link |
He studies speech and language,
link |
but he describes some of these more primitive,
link |
visceral vocalizations, cries, groans, moans of delight,
link |
other sounds as well, use your imagination,
link |
as such powerful rudders for the other,
link |
for the emotions of other people.
link |
And so I find it fascinating.
link |
I can't wait to see this video.
link |
Is that, so is the video available online?
link |
No, I haven't recorded it, I just had a bunch of Roombas
link |
that are able to scream in pain in my Boston place.
link |
Like people are ready as well.
link |
Next podcast episode with Lex, maybe we'll have that one.
link |
So the thing is like people,
link |
I've noticed because I talk so much about love
link |
and it's really who I am.
link |
I think they wanna, to a lot of people,
link |
it seems like there gotta be a dark person
link |
in there somewhere.
link |
And I thought if I release videos and Roombas screaming
link |
and they're like, yep, yep, that guy's definitely insane.
link |
What about like shouts of glee and delight?
link |
You could do that too, right?
link |
Well, I don't know how to,
link |
to me delight is quiet, right?
link |
Like you're Russian.
link |
Americans are much louder than Russians.
link |
But I don't, I mean, unless you're talking about like,
link |
I don't know how you would have sexual relations
link |
I wasn't necessarily saying sexual delight, but-
link |
Trust me, I tried, I'm just kidding.
link |
That's a joke, internet.
link |
Okay, but I was fascinated in the psychology
link |
of how little it took.
link |
Cause you mentioned you had a negative relationship
link |
Well, I'd find that mostly I took it for granted.
link |
It just served me, it collected Costello's hair.
link |
And then when it would do something I didn't like,
link |
I would get upset with it.
link |
So that's not a good relationship.
link |
It was taken for granted and I would get upset
link |
and then I'd park it again.
link |
And I just like, you're in the corner.
link |
Yeah, but there's a way to frame it being quite dumb
link |
as a almost cute, almost connecting with it
link |
And I think that's a artificial intelligence problem.
link |
I think flaws should be a feature, not a bug.
link |
So along the lines of this,
link |
the different sorts of relationships that one could have
link |
with robots and the fear,
link |
but also some of the positive relationships
link |
that one could have.
link |
There's so much dimensionality.
link |
There's so much to explore,
link |
but power dynamics in relationships are very interesting
link |
because the obvious ones that the unsophisticated view
link |
of this is one, there's a master and a servant, right?
link |
But there's also manipulation.
link |
There's benevolent manipulation.
link |
Children do this with parents.
link |
Puppies turn their head and look cute
link |
and maybe give out a little noise.
link |
And parents always think that they're doing this
link |
because they love the parent.
link |
But in many ways, studies show that those coos are ways
link |
to extract the sorts of behaviors and expressions
link |
from the parent that they want.
link |
The child doesn't know it's doing this.
link |
It's completely subconscious,
link |
but it's benevolent manipulation.
link |
So there's one version of fear of robots
link |
that I hear a lot about that I think most people
link |
can relate to where the robots take over
link |
and they become the masters and we become the servants.
link |
But there could be another version
link |
that in certain communities that I'm certainly not a part of
link |
but they call topping from the bottom
link |
where the robot is actually manipulating you
link |
into doing things, but you are under the belief
link |
that you are in charge, but actually they're in charge.
link |
And so I think that's one that if we could explore
link |
that for a second, you could imagine
link |
it wouldn't necessarily be bad,
link |
although it could lead to bad things.
link |
The reason I want to explore this is I think people
link |
always default to the extreme, like the robots take over
link |
and we're in little jail cells and they're out having fun
link |
and ruling the universe.
link |
What sorts of manipulation can a robot
link |
potentially carry out, good or bad?
link |
Yeah, just so there's a lot of good and bad manipulation
link |
between humans, right, just like you said.
link |
To me, especially like you said,
link |
topping from the bottom, is that the term?
link |
I think someone from MIT told me that term, wasn't Lex.
link |
So first of all, there's power dynamics in bed
link |
and power dynamics in relationships
link |
and power dynamics on the street
link |
and in the work environment, those are all very different.
link |
I think power dynamics can make human relationships,
link |
especially romantic relationships, fascinating and rich
link |
and fulfilling and exciting and all those kinds of things.
link |
So I don't think in themselves they're bad
link |
and the same goes with robots.
link |
I really love the idea that a robot would be a top
link |
or a bottom in terms of like power dynamics
link |
and I think everybody should be aware of that
link |
and the manipulation is not so much manipulation
link |
but a dance of like pulling away, a push and pull
link |
and all those kinds of things.
link |
In terms of control, I think we're very, very, very far away
link |
from AI systems that are able to lock us up.
link |
They, to lock us up in, like to have so much control
link |
that we basically cannot live our lives
link |
in the way that we want.
link |
I think there's, in terms of dangers of AI systems,
link |
there's much more dangers that have to do
link |
with autonomous weapon systems and all those kinds of things.
link |
So the power dynamics as exercised in the struggle
link |
between nations and war and all those kinds of things.
link |
But in terms of personal relationships,
link |
I think power dynamics are a beautiful thing.
link |
Now there is, of course, going to be all those kinds
link |
of discussions about consent and rights
link |
and all those kinds of things.
link |
Well, here we're talking, I always say,
link |
in any discussion around this,
link |
if we need to define really the context,
link |
it's always, it always should be consensual,
link |
age appropriate, context appropriate, species appropriate.
link |
But now we're talking about human robot interactions
link |
and so I guess that-
link |
No, I actually was trying to make a different point
link |
which is I do believe that robots will have rights
link |
down the line and I think in order for us
link |
to have deep meaningful relationship with robots,
link |
we would have to consider them as entities in themselves
link |
that deserve respect.
link |
And that's a really interesting concept
link |
that I think people are starting to talk about
link |
a little bit more, but it's very difficult for us
link |
to understand how entities that are other than human,
link |
I mean, the same as with dogs and other animals
link |
can have rights on a level as humans.
link |
Well, yeah, I mean, we can't and nor should we
link |
do whatever we want with animals.
link |
We have a USDA, we have departments of agriculture
link |
that deal with animal care and use committees
link |
for research, for farming and ranching and all that.
link |
So while when you first said it, I thought,
link |
wait, why would there be a bill of robotic rights?
link |
But it absolutely makes sense in the context of everything
link |
we've been talking about up until now.
link |
Let's, if you're willing, I'd love to talk about dogs
link |
because you've mentioned dogs a couple of times,
link |
a robot dog, you had a biological dog, yeah.
link |
Yeah, I had a Newfoundland named Homer
link |
for many years growing up.
link |
In Russia or in the US?
link |
In the United States.
link |
And he was about, he's over 200 pounds, that's a big dog.
link |
If people know Newfoundland, so he's this black dog
link |
that's a really long hair and just a kind soul.
link |
I think perhaps that's true for a lot of large dogs,
link |
but he thought he was a small dog.
link |
So he moved like that and-
link |
So you had him since he was fairly young?
link |
Oh, since, yeah, since the very, very beginning
link |
to the very, very end.
link |
And one of the things, I mean, he had this kind of,
link |
we mentioned like the Roombas, he had a kindhearted
link |
dumbness about him that was just overwhelming.
link |
It's part of the reason I named him Homer
link |
because it's after Homer Simpson,
link |
in case people are wondering which Homer I'm referring to.
link |
I'm not, you know.
link |
There's a clumsiness that was just something
link |
that immediately led to a deep love for each other.
link |
And one of the, I mean, he was always,
link |
it's the shared moments.
link |
He was always there for so many nights together.
link |
That's a powerful thing about a dog that he was there
link |
through all the loneliness, through all the tough times,
link |
through the successes and all those kinds of things.
link |
And I remember, I mean,
link |
that was a really moving moment for me.
link |
I still miss him to this day.
link |
How long ago did he die?
link |
Maybe 15 years ago.
link |
So it's been a while,
link |
but it was the first time I've really experienced
link |
like the feeling of death.
link |
So what happened is he got cancer
link |
and so he was dying slowly.
link |
And then at a certain point he couldn't get up anymore.
link |
There's a lot of things I could say here,
link |
you know, that I struggle with.
link |
That maybe he suffered much longer than he needed to.
link |
That's something I really think about a lot.
link |
But I remember I had to take him to the hospital
link |
and the nurses couldn't carry him, right?
link |
So you talk about 200 pound dog.
link |
I was really into powerlifting at the time.
link |
I remember like they tried to figure out
link |
all these kinds of ways to,
link |
so in order to put them to sleep,
link |
they had to take them into a room.
link |
And so I had to carry him everywhere.
link |
And here's this dying friend of mine
link |
that I just had to,
link |
first of all, it was really difficult to carry
link |
somebody that heavy when they're not helping you out.
link |
And yeah, so I remember it was the first time
link |
seeing a friend laying there
link |
and seeing wife drained from his body and that realization
link |
that we're here for a short time was made so real
link |
that here's a friend that was there for me
link |
the week before, the day before, and now he's gone.
link |
And that was, I don't know,
link |
that spoke to the fact that he could be deeply connected
link |
with a dog, also spoke to the fact
link |
that the shared moments together
link |
that led to that deep friendship
link |
will make life so amazing,
link |
but also spoke to the fact that death is a motherfucker.
link |
So I know you've lost Costello recently
link |
and you've been going.
link |
And as you're saying this,
link |
I'm definitely fighting back the tears.
link |
Thank you for sharing that,
link |
that I guess we're about to both cry over our dead dogs,
link |
that it was bound to happen
link |
just given when this is happening.
link |
How long did you know that Costello was not doing well?
link |
Well, let's see, a year ago, during the start of,
link |
about six months into the pandemic,
link |
he started getting abscesses and he was not,
link |
his behavior changed and something really changed.
link |
And then I put him on testosterone
link |
because, which helped a lot of things,
link |
it certainly didn't cure everything,
link |
but it helped a lot of things he was dealing with,
link |
joint pain, sleep issues.
link |
And then it just became a very slow decline
link |
to the point where two, three weeks ago,
link |
he had a closet full of medication.
link |
I mean, this dog was, it was like a pharmacy.
link |
It's amazing to me when I looked at it the other day,
link |
I still haven't cleaned up and removed all his things
link |
because I can't quite bring myself to do it.
link |
Do you think he was suffering?
link |
Well, so what happened was about a week ago,
link |
it was really just about a week ago, it's amazing.
link |
He was going up the stairs, I saw him slip.
link |
And he was a big dog.
link |
He wasn't 200 pounds, but he was about 90 pounds.
link |
But he's a bulldog, that's pretty big.
link |
And then I noticed that he wasn't carrying a foot
link |
in the back like it was injured.
link |
It had no feeling at all.
link |
He never liked me to touch his hind paws.
link |
And I could do, that thing was just flopping there.
link |
And then the vet found some spinal degeneration
link |
and I was told that the next one would go.
link |
But something changed in his eyes.
link |
It's the eyes again.
link |
I know you and I spend long hours on the phone
link |
and talking about like the eyes and how,
link |
what they convey and what they mean about internal states
link |
and forsaken robots and biology of other kinds.
link |
Do you think something about him was gone in his eyes?
link |
I think he was real.
link |
Here I am anthropomorphizing.
link |
I think he was realizing that one of his great joys in life,
link |
which was to walk and sniff and pee on things.
link |
This dog loved to pee on things.
link |
I've wondered where he put it.
link |
He was like a reservoir of urine.
link |
It was incredible.
link |
I think, oh, that's it.
link |
He's just, he'd put like one drop on the 50 millionth plant.
link |
And then we get to the 50 millionth in one plant
link |
and he just have, you know, leave a puddle.
link |
And here I am talking about Costello peeing.
link |
He was losing that ability to stand up and do that.
link |
He was falling down while he was doing that.
link |
And I do think he started to realize,
link |
and the passage was easy and peaceful,
link |
but you know, I'll say this.
link |
I'm not ashamed to say it.
link |
I mean, I wake up every morning since then just,
link |
I don't even make the conscious decision
link |
to allow myself to cry.
link |
And I'm fortunately able to make it through the day,
link |
thanks to the great support of my friends
link |
and you and my family.
link |
But I miss him, man.
link |
And I feel like he, you know, Homer, Costello,
link |
you know, the relationship to one's dog is so specific, but.
link |
So that part of you is gone.
link |
That's the hard thing.
link |
what I think is different is that I made the mistake,
link |
I hope it was a good decision,
link |
but sometimes I think I made the mistake
link |
of I brought Costello a little bit to the world
link |
through the podcast, through posting about him.
link |
I gave, I anthropomorphized about him in public.
link |
I have no idea what his mental life was
link |
or his relationship to me.
link |
And I'm just exploring all this for the first time
link |
because he was my first dog,
link |
but I raised him since he was seven weeks.
link |
Yeah, you got to hold it together.
link |
I noticed the episode you released on Monday,
link |
you mentioned Costello.
link |
Like you brought him back to life for me
link |
for that brief moment.
link |
Yeah, but he's gone.
link |
he's going to be gone for a lot of people too.
link |
Well, this is what I'm struggling with.
link |
I think that maybe you're pretty good at this, Lila.
link |
Wait, have you done this before?
link |
This is the challenge is I actually, part of me,
link |
I know how to take care of myself pretty well.
link |
Not perfectly, but pretty well.
link |
And I have good support.
link |
I do worry a little bit about how it's going to land
link |
and how people will feel.
link |
I'm concerned about their internalization.
link |
So that's something I'm still iterating on.
link |
And you have to, they have to watch you struggle,
link |
which is fascinating.
link |
Right, and I've mostly been shielding them from this,
link |
but what would make me happiest
link |
if people would internalize some of Costello's best traits
link |
and his best traits were that he was incredibly tough.
link |
I mean, he was a 22 inch neck, bulldog, the whole thing.
link |
He was just born that way.
link |
But what was so beautiful is that his toughness
link |
is never what he rolled forward.
link |
It was just how sweet and kind he was.
link |
And so if people can take that,
link |
then there's a win in there someplace.
link |
So I think there's some ways in which
link |
he should probably live on in your podcast too.
link |
You should, I mean, it's such a,
link |
one of the things I loved about his role in your podcast
link |
is that he brought so much joy to you.
link |
I mentioned the robots, right?
link |
I think that's such a powerful thing to bring that joy
link |
into like allowing yourself to experience that joy,
link |
to bring that joy to others, to share it with others.
link |
That's really powerful.
link |
And I mean, not to, this is like the Russian thing is,
link |
it touched me when Louis CK had that moment
link |
that I keep thinking about in his show, Louis,
link |
where like an old man was criticizing Louis
link |
for whining about breaking up with his girlfriend.
link |
And he was saying like the most beautiful thing
link |
about love, they made a song that's catchy now
link |
that's not making me feel horrible saying it,
link |
but like is the loss.
link |
The loss really also is making you realize
link |
how much that person, that dog meant to you.
link |
And like allowing yourself to feel that loss
link |
and not run away from that loss is really powerful.
link |
And in some ways that's also sweet, just like the love was,
link |
the loss is also sweet because you know
link |
that you felt a lot for that, for your friend.
link |
So I, you know, and then continue bringing that joy.
link |
I think it would be amazing to the podcast.
link |
I hope to do the same with robots
link |
or whatever else is the source of joy, right?
link |
And maybe, do you think about one day getting another dog?
link |
Yeah, in time, you're hitting on all the key buttons here.
link |
I want that to, we're thinking about, you know,
link |
ways to kind of immortalize Costello in a way that's real,
link |
not just, you know, creating some little logo
link |
or something silly.
link |
You know, Costello, much like David Goggins is a person,
link |
but Goggins also has grown into a kind of a verb.
link |
You're going to Goggins this or you're going to,
link |
and there's an adjective, like that's extreme, like it.
link |
I think that for me, Costello was all those things.
link |
He was a being, he was his own being.
link |
He was a noun, a verb, and an adjective.
link |
So, and he had this amazing superpower
link |
that I wish I could get, which is this ability
link |
to get everyone else to do things for you
link |
without doing a damn thing.
link |
The Costello effect, as I call it.
link |
So it's an idea, I hope he lives on.
link |
Yes, thank you for that.
link |
This actually has been very therapeutic for me.
link |
Which actually brings me to a question, we're friends.
link |
We're not just co-scientists, colleagues,
link |
working on a project together,
link |
and in the world that's somewhat similar.
link |
Just two dogs, just two dogs, basically.
link |
But let's talk about friendship.
link |
Because I think that, I certainly know as a scientist
link |
that there are elements that are very lonely
link |
of the scientific pursuit.
link |
There are elements of many pursuits that are lonely.
link |
Music, math, always seemed to me
link |
like they're like the loneliest people.
link |
Who knows if that's true or not.
link |
Also people work in teams,
link |
and sometimes people are surrounded by people
link |
interacting with people and they feel very lonely.
link |
But for me, and I think as well for you,
link |
friendship is an incredibly strong force
link |
in making one feel like certain things are possible
link |
or worth reaching for.
link |
Maybe even making us compulsively reach for them.
link |
So when you were growing up,
link |
you grew up in Russia until what age?
link |
Okay, and then you moved directly to Philadelphia?
link |
And then Philadelphia and San Francisco and Boston
link |
and so on, but really to Chicago.
link |
That's why I went to high school.
link |
Do you have siblings?
link |
Most people don't know that.
link |
Yeah, he is a very different person,
link |
but somebody I definitely look up to.
link |
So he's a wild man.
link |
He's, he was into, I mean,
link |
so he's also a scientist, a bio engineer,
link |
but when we were growing up and he was the person
link |
who did drink and did every drug,
link |
but also was the life of the party.
link |
And I just thought he was the,
link |
when you're the older brother, five years older,
link |
he was the coolest person that I always wanted to be him.
link |
So to that, he definitely had a big influence.
link |
But I think for me, in terms of friendship growing up,
link |
I had one really close friend.
link |
And then when I came here, I had another close friend,
link |
but I'm very, I believe, I don't know if I believe,
link |
but I draw a lot of strength from deep connections
link |
with other people and just a small number of people,
link |
just a really small number of people.
link |
That's when I moved to this country,
link |
I was really surprised how like there would be
link |
these large groups of friends, quote unquote,
link |
but the depth of connection was not there at all
link |
from my sort of perspective.
link |
Now I moved to the suburb of Chicago was Naperville.
link |
It's more like a middle-class, maybe upper middle-class.
link |
So it's like people that cared more
link |
about material possessions than deep human connection.
link |
So that added to the thing.
link |
But I drew more meaning than almost anything else
link |
was from friendship early on.
link |
I had a best friend, his name was, his name is Yura.
link |
I don't know how to say it in English.
link |
How do you say it in Russian?
link |
What's his last name?
link |
Mirkulov, Yura Mirkulov.
link |
So we just spent all our time together.
link |
There's also a group of friends.
link |
Like, I don't know, it's like eight guys.
link |
In Russia, growing up, it's like parents didn't care
link |
if you're coming back at a certain hour.
link |
So we would spend all day, all night,
link |
just playing soccer, usually called football
link |
and just talking about life and all those kinds of things.
link |
Even at that young age, I think people in Russia
link |
and the Soviet Union grow up much quicker.
link |
I think the education system at the university level
link |
is world-class in the United States
link |
in terms of really creating really big, powerful minds,
link |
at least it used to be, but I think that they aspire to that.
link |
But the education system for younger kids
link |
in the Soviet Union was incredible.
link |
They did not treat us as kids.
link |
The level of literature, Tolstoy, Dostoyevsky.
link |
When you were just a small child?
link |
The level of mathematics and you're made to feel like shit
link |
if you're not good at mathematics.
link |
Like we, I think in this country, there's more,
link |
like especially young kids, cause they're so cute.
link |
Like they're being babied.
link |
We only start to really push adults later in life.
link |
Like, so if you want to be the best in the world at this,
link |
then you get to be pushed.
link |
But we were pushed at a young age, everybody was pushed.
link |
And that brought out the best in people.
link |
I think they really forced people to discover like,
link |
discover themselves in the Goggin style,
link |
but also discover what they're actually passionate about,
link |
Was it true for boys and girls
link |
where they pushed equally there?
link |
Yeah, they were pushed.
link |
Yeah, they were pushed equally, I would say.
link |
There was, obviously there was more, not obviously,
link |
but there, at least from my memories, more of a,
link |
what's the right way to put it,
link |
but there was like gender roles,
link |
but not in a negative connotation.
link |
It was the red dress versus the suit and tie
link |
kind of connotation, which is like,
link |
there's like guys like lifting heavy things
link |
and girls like creating beautiful art.
link |
A more traditional view of gender, more 1950s, 60s.
link |
But we didn't think in terms of, at least at that age,
link |
in terms of like roles and then like a homemaker or something
link |
like that, or no, it was more about what people care about.
link |
Like girls cared about this set of things
link |
and guys cared about this set of things.
link |
I think mathematics and engineering was something
link |
that guys cared about and sort of,
link |
at least my perception of that time.
link |
And then girls cared about beauty.
link |
So like guys want to create machines,
link |
girls want to create beautiful stuff.
link |
And now, of course, that I don't take that forward
link |
in some kind of philosophy of life,
link |
but it's just the way I grew up and the way I remember it.
link |
But all, everyone worked hard.
link |
The value of hard work was instilled in everybody.
link |
And through that, I think it's a little bit of hardship.
link |
Of course, also economically, everybody was poor,
link |
especially with the collapse of the Soviet Union.
link |
There's poverty everywhere.
link |
You didn't notice it as much,
link |
but there was a, because there's not much material
link |
possessions, there was a huge value placed
link |
on human connection.
link |
Just meeting with neighbors, everybody knew each other.
link |
We lived in an apartment building,
link |
very different than you have in the United States
link |
these days, everybody knew each other.
link |
You would get together, drink vodka, smoke cigarettes
link |
and play guitar and sing sad songs about life.
link |
What's with the sad songs and the Russian thing?
link |
I mean, Russians do express joy from time to time.
link |
But what do you think that's about?
link |
Is it because it's cold there?
link |
But it's cold other places too.
link |
I think, so first of all, the Soviet Union,
link |
the echoes of World War II and the millions
link |
and millions and millions of people there,
link |
civilians that were slaughtered
link |
and also starvation is there, right?
link |
So like the echoes of that, of the ideas,
link |
the literature, the art is there.
link |
Like that's grandparents, that's parents, that's all there.
link |
So that contributes to it, that life can be absurdly,
link |
unexplainably cruel.
link |
At any moment, everything can change.
link |
So that's in there.
link |
Then I think there's an empowering aspect
link |
to finding beauty and suffering,
link |
but then everything else is beautiful too.
link |
Like if you just linger or it's like
link |
why you meditate on death.
link |
It's like if you just think about the worst possible case
link |
and find beauty in that, then everything else beautiful too.
link |
And so you write songs about the dark stuff
link |
and that somehow helps you deal with whatever comes.
link |
There's a hopelessness to the Soviet Union
link |
that like inflation, all those kinds of things
link |
where people were sold dreams and never delivered.
link |
And so like, if you don't sing songs about sad things,
link |
you're going to become cynical about this world.
link |
So they don't want to give into cynicism.
link |
Now, a lot of people did, one of the,
link |
but it's the battle against cynicism.
link |
One of the things that may be common in Russia
link |
is the kind of cynicism about,
link |
like if I told you the thing I said earlier
link |
about dreaming about robots,
link |
it's very common for people to dismiss that dream
link |
of saying, nah, that's not, that's too wild.
link |
Like who else do you know that did that?
link |
Or you want to start a podcast, like who else?
link |
Like nobody's making money on podcasts.
link |
Like why do you want to start a podcast?
link |
That kind of mindset I think is quite common,
link |
which is why I would say entrepreneurship in Russia
link |
is still not very good, which to be a business,
link |
like to be an entrepreneur, you have to dream big
link |
and you have to have others around you,
link |
like friends and support group that makes you dream big.
link |
But if you don't give into cynicism
link |
and appreciate the beauty in the unfairness of life,
link |
the absurd unfairness of life,
link |
then I think it just makes you appreciative of everything.
link |
It's like a, it's a prerequisite for gratitude.
link |
And so, yeah, I think that instilled in me
link |
ability to appreciate everything,
link |
just like everything, everything's amazing.
link |
And then also there's a culture
link |
of like romanticizing everything.
link |
Like it's almost like romantic relationships
link |
were very like soap opera like,
link |
is very like over the top dramatic.
link |
And I think that was instilled in me too.
link |
Not only do I appreciate everything about life,
link |
but I get like emotional about it.
link |
In a sense, like I get like a visceral feeling of joy
link |
for everything and the same with friends
link |
or people of the opposite sex.
link |
Like there's a deep, like emotional connection there
link |
that like, that's like way too dramatic to like,
link |
I guess relative to what the actual moment is.
link |
But I derive so much deep, like dramatic joy
link |
from so many things in life.
link |
And I think I would attribute that to bringing in Russia.
link |
But the thing that sticks most of all is the friendship
link |
and have now since then had one other friend like that
link |
in the United States, he lives in Chicago.
link |
And slowly here and there accumulating
link |
really fascinating people,
link |
but I'm very selective with that.
link |
Funny enough, the few times,
link |
it's not few, it's a lot of times now
link |
interacting with Joe Rogan.
link |
It sounds surreal to say,
link |
but there was a kindred spirit there.
link |
Like I've connected with him.
link |
And there's been people like that
link |
also in the grappling sports
link |
that are really connected with.
link |
I've actually struggled,
link |
which is why I'm so glad to be your friend
link |
is I've struggled to connect with scientists.
link |
They can be a little bit wooden sometimes.
link |
Even the biologists.
link |
I mean, one thing that I-
link |
Even the biologists.
link |
Well, I'm so struck by the fact that you work with robots,
link |
you're an engineer, AI.
link |
Science, technology.
link |
And that all sounds like hardware, right?
link |
But what you're describing,
link |
and I know is true about you,
link |
is this deep emotional life and this resonance.
link |
And it's really wonderful.
link |
I actually think it's one of the reasons why
link |
so many people, scientists and otherwise,
link |
have gravitated towards you and your podcast
link |
is because you hold both elements.
link |
In Hermann Hesse's book,
link |
I don't know if you were at Narcissus and Goldman, right?
link |
It's about these elements of the logical, rational mind
link |
and the emotional mind
link |
and how those are woven together.
link |
And if people haven't read it, they should.
link |
And you embody the full picture.
link |
And I think that's so much of what draws people to you.
link |
I've read every Hermann Hesse book, by the way.
link |
As usual, I've done about 9% of what Lexi said.
link |
You mentioned Joe, who is a phenomenal human being,
link |
not just for his amazing accomplishments,
link |
but for how he shows up to the world one-on-one.
link |
I think I heard him say the other day on an interview,
link |
he said, there is no public or private version of him.
link |
He's like, this is me.
link |
He said the word, it was beautiful.
link |
He said, I'm like the fish that got through the net.
link |
There is no on-stage, off-stage version.
link |
And you're absolutely right.
link |
And I, so, well, you guys, I have a question actually about-
link |
But that's a really good point
link |
about public and private life.
link |
He was a huge, if I could just comment real quick.
link |
Like that, he was, I've been a fan of Joe for a long time,
link |
but he's been an inspiration
link |
to not have any difference between public and private life.
link |
I actually had a conversation with Naval about this.
link |
And he said that you can't have a rich life,
link |
like an exciting life
link |
if you're the same person publicly and privately.
link |
And I think I understand that idea,
link |
but I don't agree with it.
link |
I think it's really fulfilling and exciting
link |
to be the same person privately and publicly,
link |
with very few exceptions.
link |
Now, that said, I don't have any really strange sex kinks.
link |
So like, I feel like I can be open with basically everything.
link |
I don't have anything I'm ashamed of.
link |
There's some things that could be perceived poorly,
link |
like the screaming Arumbas, but I'm not ashamed of them.
link |
I just have to present them in the right context.
link |
But there's freedom to being the same person in private
link |
And that Joe made me realize that you can be that
link |
and also to be kind to others.
link |
It sounds kind of absurd,
link |
but I really always enjoyed like being good to others.
link |
Like just being kind towards others.
link |
But I always felt like the world didn't want me to be.
link |
Like there's so much negativity when I was growing up,
link |
like just around people.
link |
If you actually just notice how people talk,
link |
from like complaining about the weather,
link |
this could be just like the big cities that I visited in,
link |
but there's a general negativity
link |
and positivity is kind of suppressed.
link |
One, you're not seen as very intelligent.
link |
And two, you're seen as like a little bit of a weirdo.
link |
And so I always felt like I had to hide that.
link |
And what Joe made me realize,
link |
one, I could be fully just the same person,
link |
private and public.
link |
And two, I can embrace being kind
link |
and just in the way that I like,
link |
in the way I know how to do.
link |
And sort of for me on like on Twitter
link |
or like publicly, whenever I say stuff,
link |
that means saying stuff simply,
link |
almost to the point of cliche.
link |
And like, I have the strength now to say it,
link |
even if I'm being mocked, you know what I mean?
link |
Like just, it's okay.
link |
If everything's going to be okay.
link |
Okay, some people will think you're dumb.
link |
They're probably right.
link |
The point is like, it's just enjoy being yourself.
link |
And Joe more than almost anybody else,
link |
because he's so successful at it, inspired me to do that.
link |
Be kind and be the same person, private and public.
link |
I love it, and I love the idea that authenticity
link |
doesn't have to be oversharing, right?
link |
That it doesn't mean you reveal every detail of your life,
link |
what, you know, it's a way of being true
link |
to an essence of oneself.
link |
Right, there's never a feeling
link |
when you deeply think and introspect
link |
that you're hiding something from the world
link |
or you're being dishonest in some fundamental way.
link |
So yeah, that's truly liberating.
link |
It allows you to think, it allows you to,
link |
like think freely, to speak freely,
link |
to just to be freely.
link |
That said, it's not like, you know,
link |
it's not like there's not still a responsibility
link |
to be the best version of yourself.
link |
So, you know, I'm very careful with the way I say something.
link |
So the whole point, it's not so simple
link |
to express the spirit that's inside you with words.
link |
It depends, I mean, some people are much better
link |
I struggle, like oftentimes when I say something
link |
and I hear myself say it, it sounds really dumb
link |
and not at all what I meant.
link |
So that's the responsibility you have.
link |
It's not just like being the same person publicly
link |
and privately means you can just say whatever the hell.
link |
It means there's still a responsibility to try to be,
link |
to express who you truly are.
link |
It is hard and I think that, you know, we have this pressure,
link |
all people, when I say we, I mean all humans,
link |
and maybe robots too, feel this pressure
link |
to be able to express ourselves in that one moment,
link |
And it is beautiful when somebody, for instance,
link |
can capture some essence of love or sadness
link |
or anger or something in a song or in a poem
link |
or in a short quote.
link |
But perhaps it's also possible to do it in aggregate.
link |
You know, all the things, you know, how you show up.
link |
For instance, one of the things that initially drew me
link |
to want to get to know you as a human being
link |
and a scientist and eventually we became friends
link |
was the level of respect that you brought
link |
to your podcast listeners by wearing a suit.
link |
I'm being serious here.
link |
You know, I was raised thinking that if you overdress
link |
a little bit, overdress by American,
link |
certainly by American standards,
link |
you're overdressed for a podcast,
link |
but this is, but it's genuine.
link |
You're not doing it for any reason,
link |
except I have to assume, and I assumed at the time,
link |
that it was because you have a respect for your audience.
link |
You respect them enough to show up a certain way for them.
link |
It's for you also, but it's for them.
link |
And I think between that and your commitment
link |
to your friendships, the way that you talk about friendships
link |
and love and the way you hold up these higher ideals,
link |
I think at least as a consumer of your content
link |
and as your friend, what I find is that in aggregate,
link |
you're communicating who you are.
link |
It doesn't have to be one quote or something.
link |
And I think that we were sort of obsessed
link |
by like the one Einstein quote
link |
or the one line of poetry or something,
link |
but I think you so embody the way that, and Joe as well,
link |
it's about how you live your life and how you show up
link |
as a collection of things and said and done.
link |
Yeah, that's fascinating, so the aggregate is the goal.
link |
The tricky thing, and Jordan Peterson talks about this
link |
because he's under attack way more than you and I
link |
will ever be, but that-
link |
This is very true for now.
link |
That the people who attack on the internet,
link |
this is one of the problems with Twitter,
link |
is they don't consider the aggregate.
link |
They take a single statements.
link |
And so one of the defense mechanisms,
link |
like again, why Joe has been an inspiration
link |
is that when you in aggregate are a good person,
link |
a lot of people will know that.
link |
And so that makes you much more immune
link |
to the attacks of people
link |
that bring out an individual statement
link |
that might be a misstatement of some kind
link |
or doesn't express who you are.
link |
And so that, I like that idea is the aggregate
link |
and the power of the podcast is you have hundreds
link |
of hours out there and being yourself
link |
and people get to know who you are.
link |
And once they do and you post pictures
link |
of screaming Roombas as you kick them,
link |
they will understand that you don't mean well.
link |
By the way, as a side comment,
link |
I don't know if I want to release this
link |
because it's not just the Roombas.
link |
You have a whole dungeon of robots.
link |
Okay, so this is a problem.
link |
Boston Dynamics came up against this problem,
link |
but let me just, let me work this out,
link |
like workshop this out with you.
link |
And maybe because we'll post this,
link |
people will let me know.
link |
So there's legged robots.
link |
They look like a dog.
link |
I'm trying to create a very real human robot connection,
link |
but like they're also incredible
link |
because you can throw them like off of a building
link |
and it'll land fine.
link |
And it's beautiful.
link |
I've seen the Instagram videos of like cats
link |
jumping off of like fifth story buildings
link |
and then walking away.
link |
No one should throw their cat out of a window.
link |
This is the problem I'm experiencing.
link |
I'll certainly kicking the robots.
link |
It's really fascinating how they recover from those kicks,
link |
but like just seeing myself do it
link |
and also seeing others do it, it just does not look good.
link |
And I don't know what to do with that.
link |
Cause I, it's such a-
link |
See, but you don't, I, you, cause you-
link |
Robot, no, I'm kidding.
link |
Now I'm, you know what's interesting?
link |
Before today's conversation, I probably could do it.
link |
And now I think I'm thinking about robots,
link |
bills of rights and things I'm actually,
link |
and not for any, not to satisfy you
link |
or to satisfy anything, except that if I,
link |
if they have some sentient aspect to their being,
link |
then I would loathe to kick it.
link |
I don't think you'd be able to kick it.
link |
You might be able to get the first time, but not the second.
link |
This is, this is the problem I've experienced.
link |
One of the cool things is one of the robots I'm,
link |
you can pick it up by one leg and it's dangling.
link |
You can throw it in any kind of way
link |
and it'll land correctly.
link |
I had a friend who had a cat like that.
link |
Oh man, we look forward to the letters from the cat.
link |
Oh no, I'm not suggesting anyone did that,
link |
but he had this cat and the cat, he would just, you know,
link |
throw it onto the bed from across the room
link |
and then it would run back for more.
link |
Somehow they had, that was the nature of the relationship.
link |
I think most, no one should do that to an animal,
link |
but this cat seemed to, you know,
link |
return for it for whatever reason.
link |
The robot is a robot and it's fascinating to me
link |
how hard it is for me to do that.
link |
So it's unfortunate,
link |
but I don't think I can do that to a robot.
link |
Like I struggle with that.
link |
So for me to be able to do that with a robot,
link |
I have to almost get like into the state
link |
that I imagine like doctors get into
link |
when they're doing surgery.
link |
Like I have to start,
link |
I have to do what robotics colleagues of mine do,
link |
which is like start seeing it as an object.
link |
So it was just fascinating that I have to do that
link |
in order to do that with a robot.
link |
I just wanted to take that a little bit of a tangent.
link |
No, I think it's an important thing.
link |
I mean, I am not, I'm not shy about the fact
link |
that for many years I've worked on experimental animals
link |
and that's been a very challenging aspect
link |
to being a biologist, mostly mice,
link |
but in the past, no longer, thank goodness,
link |
cause I just don't like doing it,
link |
larger animals as well.
link |
And now I work on humans,
link |
which I can give consent, verbal consent.
link |
So I think that it's extremely important
link |
to have an understanding of what the guidelines are
link |
and where one's own boundaries are around this.
link |
It's not just an important question.
link |
It might be the most important question
link |
before any work can progress.
link |
So you asked me about friendship.
link |
I know you have a lot of thoughts about friendship.
link |
What do you think is the value of friendship in life?
link |
Well, for me personally,
link |
just because of my life trajectory and arc of friendship,
link |
and I should say, I do have some female friends
link |
that are just friends,
link |
they're completely platonic relationships,
link |
but it's been mostly male friendship to me has been-
link |
It's been all male friendships to me, actually.
link |
Interesting, yeah.
link |
It's been an absolute lifeline.
link |
They are my family.
link |
I have a biological family and I have great respect
link |
and love for them and an appreciation for them,
link |
but it's provided me the,
link |
I wouldn't even say confidence
link |
because there's always an anxiety in taking any good risk
link |
or any risk worth taking.
link |
It's given me the sense that I should go for certain things
link |
and try certain things to take risks,
link |
to weather that anxiety.
link |
And I don't consider myself
link |
a particularly competitive person,
link |
but I would sooner die than disappoint
link |
or let down one of my friends.
link |
I can think of nothing worse, actually,
link |
than disappointing one of my friends.
link |
Everything else is secondary to me.
link |
Well, disappointment-
link |
Disappointing meaning not,
link |
I mean, certainly I strive always to show up
link |
as best I can for the friendship,
link |
and that can be in small ways.
link |
That can mean making sure the phone is away.
link |
Sometimes it's about,
link |
I'm terrible with punctuality because I'm an academic
link |
and so I just get lost in time and I don't mean anything by,
link |
but striving to listen, to enjoy good times and to make time.
link |
It kind of goes back to this first variable we talked about,
link |
to make sure that I spend time
link |
and to get time in person and check in.
link |
I think there's so many ways
link |
in which friendship is vital to me.
link |
It's actually, to me, what makes life worth living.
link |
I am surprised, like with the high school friends,
link |
how we don't actually talk that often these days
link |
in terms of time, but every time we see each other,
link |
it's immediately right back to where we started.
link |
So I struggle with that,
link |
how much time you really allocate
link |
for the friendship to be deeply meaningful
link |
because they're always there with me,
link |
even if we don't talk often.
link |
So there's a kind of loyalty.
link |
I think maybe it's a different style,
link |
but I think to me,
link |
friendship is being there in the hard times, I think.
link |
I'm much more reliable when you're going through shit.
link |
You're pretty reliable anyway.
link |
No, but if you're like a wedding or something like that,
link |
or I don't know, you want an award of some kind,
link |
yeah, I'll congratulate the shit out of you,
link |
but that's not, and I'll be there,
link |
but that's not as important to me as being there
link |
when nobody else is,
link |
just being there when shit hits the fan
link |
or something's tough where the world turns their back
link |
on you, all those kinds of things.
link |
That to me, that's where friendship is meaningful.
link |
Well, I know that to be true about you,
link |
and that's a felt thing and a real thing with you.
link |
Let me ask one more thing about that actually,
link |
because I'm not a practitioner of jujitsu.
link |
I know you are, Joe is,
link |
but years ago I read a book that I really enjoyed,
link |
which is Sam Sheridan's book, A Fighter's Heart.
link |
He talks about all these different forms of martial arts,
link |
and maybe it was in the book, maybe it was in an interview,
link |
but he said that, you know,
link |
fighting or being in physical battle with somebody,
link |
jujitsu, boxing, or some other form of physical,
link |
direct physical contact between two individuals
link |
creates this bond unlike any other,
link |
because he said, it's like a one night stand.
link |
You're sharing bodily fluids with somebody
link |
that you barely know.
link |
And I, you know, and I chuckled about it
link |
because it's kind of funny and a kind of tongue in cheek,
link |
but at the same time, I think this is a fundamental way
link |
in which members of a species bond
link |
is through physical contact.
link |
And certainly there are other forms.
link |
There's cuddling and there's hand holding
link |
and there's sexual intercourse
link |
and there's all sorts of things.
link |
I haven't heard of it.
link |
I heard this recently.
link |
I didn't know this term, but there's a term.
link |
They've turned the noun cupcake into a verb.
link |
Cupcaking, it turns out, I just learned about this.
link |
Cupcaking is when you spend time just cuddling.
link |
I didn't know about this.
link |
You heard it here first,
link |
although I heard it first just the other day,
link |
cupcaking is actually a verb.
link |
So cuddling is everything.
link |
It's not just like, is it in bed or is it on the couch?
link |
Like what's cuddling?
link |
I need to look up what cuddling is.
link |
We need to look at this stuff
link |
and we need to define the variables.
link |
I think it definitely has to do with physical contact,
link |
I am told, but in terms of battle, competition,
link |
you know, and the Sheridan quote, I'm just curious.
link |
So do you get close or feel a bond with people
link |
that, for instance, you rolled jujitsu with,
link |
even though you don't know anything else about them?
link |
Is he, was he right about this?
link |
Yeah, I mean, on many levels.
link |
He also has the book, what,
link |
A Fighter's Mind and The First Heart.
link |
He's actually an excellent writer.
link |
What's interesting about him, just briefly about Sheridan,
link |
I don't know, but I did a little bit of research.
link |
He went to Harvard.
link |
He was an art major at Harvard.
link |
He claims all he did was smoke cigarettes and do art.
link |
I don't know if his art was any good.
link |
And I think his father was in the SEAL teams.
link |
And then when he got out of Harvard, graduated,
link |
he took off around the world,
link |
learning all the forms of martial arts
link |
and was early to the kind of ultimate fighting
link |
to kind of mix martial arts and things.
link |
Great, great book.
link |
Yeah, it's amazing.
link |
I don't actually remember it, but I read it.
link |
I remember thinking there was an amazing encapsulation
link |
of what makes fighting the art,
link |
like what makes it compelling.
link |
I would say that there's so many ways that jiu-jitsu,
link |
grappling, wrestling, combat sports in general,
link |
is like one of the most intimate things you could do.
link |
I don't know if I would describe it
link |
in terms of bodily liquids and all those kinds of things.
link |
I think he was more or less joking, but.
link |
I think there's a few ways that it does that.
link |
So one, because you're so vulnerable.
link |
So that the honesty of stepping on the mat
link |
and often all of us have ego thinking
link |
we're better than we are at this particular art.
link |
And then the honesty of being submitted
link |
or being worse than you thought you are
link |
and just sitting with that knowledge.
link |
That kind of honesty,
link |
we don't get to experience it in most of daily life.
link |
We can continue living somewhat of an illusion
link |
of our conceptions of ourselves
link |
because people are not going to hit us with the reality.
link |
The mat speaks only the truth,
link |
that the reality just hits you.
link |
And that vulnerability is the same
link |
as like the loss of a loved one.
link |
It's the loss of a reality that you knew before.
link |
You now have to deal with this new reality.
link |
And when you're sitting there in that vulnerability
link |
and there's these other people
link |
that are also sitting in that vulnerability,
link |
you get to really connect like, fuck.
link |
Like I'm not as special as I thought I was
link |
and life is like not,
link |
life is harsher than I thought I was
link |
and we're just sitting there with that reality.
link |
Some of us can put words to them, some of them can't.
link |
So I think that definitely is the thing
link |
that leads to intimacy.
link |
The other thing is the human contact.
link |
There is something about, I mean, like a big hug.
link |
Like during COVID, very few people hugged me
link |
and I hugged them and I always felt good when they did.
link |
Like we're all tested and especially now we're vaccinated,
link |
but there's still people, this is true of San Francisco,
link |
this is true in Boston.
link |
They want to keep not only six feet away,
link |
but stay at home and never touch you.
link |
That was, that loss of basic humanity
link |
is the opposite of what I feel in jiu-jitsu
link |
where it was like that contact where you're like,
link |
I don't give a shit about whatever rules
link |
we're supposed to have in society where you're not,
link |
you have to keep a distance and all that kind of stuff.
link |
Just the hug, like the intimacy of a hug
link |
that's like a good bear hug
link |
and you're like just controlling another person.
link |
And also there is some kind of love communicating
link |
through just trying to break each other's arms.
link |
I don't exactly understand why violence
link |
is such a close neighbor to love, but it is.
link |
Well, in the hypothalamus,
link |
the neurons that control sexual behavior,
link |
but also non-sexual contact are not just nearby
link |
the neurons that control aggression and fighting,
link |
they are salt and pepper with those neurons.
link |
It's a very interesting and it almost sounds
link |
kind of risque and controversial and stuff.
link |
I'm not anthropomorphizing about what this means,
link |
but in the brain, those structures are interdigitated.
link |
You can't separate them except at a very fine level.
link |
And here the way you describe it is the same
link |
I do want to make an interesting comment.
link |
Again, these are the things
link |
that could be taken out of context,
link |
but one of the amazing things about jiu-jitsu
link |
is both guys and girls train it.
link |
And I was surprised.
link |
So like I'm a big fan of yoga pants,
link |
at the gym kind of thing.
link |
It reveals the beauty of the female form.
link |
But the thing is like girls are dressed
link |
in skintight clothes in jiu-jitsu often.
link |
And I found myself like not at all thinking like that at all
link |
when training with girls.
link |
Well, the context is very non-sexual.
link |
But I was surprised to learn that.
link |
When I first started jiu-jitsu,
link |
I thought wouldn't that be kind of weird
link |
to train with the opposites in something so intimate?
link |
Boys and girls, men and women,
link |
they rolled jiu-jitsu together completely.
link |
And the only times girls kind of try to stay away from guys,
link |
I mean, there's two contexts.
link |
Of course, there's always going to be creeps in this world.
link |
So everyone knows who to stay away from.
link |
And the other is like, there's a size disparity.
link |
So girls will often try to roll with people
link |
a little bit closer weight-wise.
link |
But no, that's one of the things
link |
that are empowering to women.
link |
That's what they fall in love with
link |
when they start doing jiu-jitsu is I can,
link |
first of all, they gain an awareness
link |
and a pride over their body, which is great.
link |
And then second, they get, especially later on,
link |
start submitting big dudes,
link |
like these like bros that come in
link |
who are all shredded and like muscular.
link |
And they get through technique to exercise dominance
link |
And that's a powerful feeling.
link |
You've seen women force a larger guy to tap
link |
or even choke him out.
link |
Well, I was deadlifting for,
link |
oh boy, I think it's 495.
link |
So I was really into powerlifting when I started jiu-jitsu.
link |
And I remember being submitted by,
link |
I thought I walked in feeling like I'm going to be,
link |
if not the greatest fighter of at least top three.
link |
And so as a white belt, you roll in like all happy.
link |
And then you realize that as long as you're not applying
link |
too much force that you're having,
link |
I remember being submitted many times
link |
by like 130, 120 pound girls
link |
at the Balance Studios in Philadelphia,
link |
that a lot of incredible female jiu-jitsu players.
link |
And that's really humbling too.
link |
The technique can overpower in combat
link |
and pure strength.
link |
And that's the other thing that there is something
link |
about combat that's primal.
link |
Like it just feels, it feels like we were born to do this.
link |
Like that there's-
link |
But we have circuits in our brain that are dedicated
link |
to this kind of interaction.
link |
There's no question.
link |
And like, that's what it felt like.
link |
It wasn't that I'm learning a new skill.
link |
It was like somehow I am remembering echoes
link |
of something I've learned in the past.
link |
It's like hitting puberty.
link |
A child before puberty has no concept
link |
of boys and girls having this attraction,
link |
regardless of whether or not they're attracted
link |
to boys or girls, doesn't matter.
link |
At some point, most people, not all, but certainly,
link |
but most people, when they hit puberty,
link |
suddenly people appear differently.
link |
And certain people take on a romantic or sexual interest
link |
for the very first time.
link |
And so it's like, it's revealing a circuitry in the brain.
link |
It's not like they learned that, it's innate.
link |
And I think when I hear the way you describe jiu-jitsu
link |
and rolling jiu-jitsu, it reminds me a little bit,
link |
Joe was telling me recently about the first time
link |
he went hunting and he felt like it revealed a circuit
link |
that was in him all along,
link |
but he hadn't experienced before.
link |
Yeah, that's definitely there.
link |
And of course there's the physical activity.
link |
One of the interesting things about jiu-jitsu
link |
is it's one of the really strenuous exercises
link |
that you can do late into your adult life,
link |
like into your 50s, 60s, 70s, 80s.
link |
When I came up, there's a few people in their 80s
link |
that were training.
link |
And as long as you're smart,
link |
as long as you practice techniques
link |
and pick your partners correctly,
link |
you can do that kind of art.
link |
It's late into life and so you're getting exercise.
link |
There's not many activities I find
link |
that are amenable to that.
link |
So because it's such a thinking game,
link |
the jiu-jitsu in particular is an art
link |
where technique pays off a lot.
link |
So you can still maintain, first of all,
link |
remain injury free if you use good technique
link |
and also through good technique,
link |
be able to go, be active with people
link |
that are much, much younger.
link |
And so that was to me,
link |
that and running are the two activities
link |
you can kind of do late in life
link |
because to me, a healthy life has exercise
link |
as a piece of the puzzle.
link |
And I'm glad that we're on the physical component
link |
because I know that there's, for you,
link |
you've talked before about the crossover
link |
between the physical and the intellectual and the mental.
link |
Are you still running at ridiculous hours of the night
link |
for ridiculously long?
link |
Yeah, so definitely.
link |
I've been running late at night here in Austin.
link |
People tell me, the area we're in now,
link |
people say is a dangerous area,
link |
which I find laughable coming from the bigger cities.
link |
No, I've run late at night.
link |
There's something.
link |
If you see a guy running through Austin at 2 a.m.
link |
in a suit and tie, it's probably.
link |
Well, yeah, I mean, I do think about that
link |
because I get recognized more and more in Austin.
link |
I worry that, not really,
link |
that I get recognized late at night.
link |
But there is something about the night
link |
that brings out those deep philosophical thoughts
link |
and self-reflection that I really enjoy.
link |
But recently I started getting back to the grind.
link |
So I'm gonna be competing or hoping to be compete
link |
in September and October.
link |
In jiu-jitsu, yeah, to get back to competition.
link |
And so that requires getting back into great cardio shape.
link |
And so I've been getting running
link |
as part of my daily routine.
link |
Well, I always know I can reach you
link |
regardless of time zone in the middle of the night,
link |
wherever that happens.
link |
Well, part of that has to be just being single
link |
and being a programmer.
link |
Those two things just don't work well
link |
in terms of a steady sleep schedule.
link |
It's not banker's hours kind of work, nine to five.
link |
I want to, you mentioned single.
link |
I want to ask you a little bit
link |
about the other form of relationship,
link |
which is a romantic love.
link |
So your parents are still married?
link |
Still married, still happily married.
link |
That's impressive.
link |
A rare thing nowadays.
link |
So you grew up with that example.
link |
Yeah, I guess that's a powerful thing, right?
link |
If there's an example that I think can work.
link |
Yeah, I didn't have that in my own family,
link |
but when I see it, it's inspiring and it's beautiful.
link |
The fact that they have that and that was the norm for you,
link |
I think is really wonderful.
link |
In the case of my parents, it was interesting to watch
link |
because there's obviously tension.
link |
Like there'll be times where they fought
link |
and all those kinds of things.
link |
They obviously get frustrated with each other
link |
and they like, but they find mechanisms
link |
how to communicate that to each other,
link |
like to make fun of each other a little bit,
link |
like to tease, to get some of that frustration out
link |
and then ultimately to reunite
link |
and find their joyful moments and be that the energy.
link |
I think it's clear because I got together in there,
link |
I think early twenties, like very, very young.
link |
I think you grow together as people.
link |
Yeah, you're still in the critical period
link |
of brain plasticity.
link |
And also, I mean, it's just like divorce
link |
was so frowned upon that you stick it out.
link |
And I think a lot of couples,
link |
especially from that time in the Soviet Union,
link |
that's probably applies to a lot of cultures.
link |
You stick it out and you put in the work.
link |
You learn how to put in the work.
link |
And once you do, you start to get to some of those
link |
rewarding aspects of being like through time,
link |
sharing so many moments together.
link |
That's definitely something that was an inspiration to me,
link |
but maybe that's where I have,
link |
so I have a similar kind of longing
link |
to have a lifelong partner like that,
link |
have that kind of view where same with friendship,
link |
lifelong friendship is the most meaningful kind
link |
that there is something with that time
link |
of sharing all that time together,
link |
like till death do us part as a powerful thing,
link |
not by force, not because the religion said it
link |
or the government said it or your culture said it,
link |
but because you want to.
link |
Do you want children?
link |
Definitely want children.
link |
How many Roombas do you have?
link |
Oh, I thought you should know human children.
link |
No, human children.
link |
Because I already have the children.
link |
Exactly, well, I was saying you probably need
link |
at least as many human children as you do Roombas,
link |
big family, small family.
link |
In your mind's eye, is there a big,
link |
are there a bunch of freedman's running around?
link |
So I'll tell you like realistically,
link |
I can explain exactly my thinking.
link |
And this is similar to the robotics work
link |
is if I'm like purely logical right now,
link |
my answer would be I don't want kids
link |
because I just don't have enough time.
link |
I have so much going on.
link |
But when I'm using the same kind of vision
link |
I use for the robots is I know my life
link |
will be transformed with the first.
link |
Like I know I would love being a father.
link |
And so the question of how many,
link |
that's on the other side of that hill.
link |
It could be some ridiculous number.
link |
So I just know that-
link |
I have a feeling and I don't have a crystal ball,
link |
but I don't know, I see an upwards
link |
of certainly three or more comes to mind.
link |
So so much of that has to do
link |
with the partner you're with too.
link |
So like that's such an open question,
link |
especially in this society of what the right partnership is.
link |
Because I'm deeply empathetic.
link |
I want to see, like to me,
link |
what I look for in a relationship is
link |
for me to be really excited about the passions
link |
of another person, like whatever they're into.
link |
It doesn't have to be a career success,
link |
any kind of success, just to be excited for them.
link |
And for them to be excited for me
link |
and they can share in that excitement
link |
and build and build and build.
link |
But there was also practical aspects of like,
link |
what kind of shit do you enjoy doing together?
link |
And I think family is a real serious undertaking.
link |
Oh, it certainly is.
link |
I mean, I think that I have a friend who said it,
link |
I think best, which is that you first half,
link |
he's in a very successful relationship and has a family.
link |
And he said, you first have to define the role
link |
and then you have to cast the right person for the role.
link |
Well, yeah, there's some deep aspect to that,
link |
but there's also an aspect to which you're not smart enough
link |
from this side of it to define the role.
link |
There's part of it that has to be a leap
link |
that you have to take.
link |
And I see having kids that way.
link |
You just have to go with it and figure it out also,
link |
as long as there's love there.
link |
Like what the hell is life for even?
link |
So there's so many incredibly successful people that I know,
link |
that I've gotten to know, that all have kids.
link |
And the presence of kids for the most part
link |
has only been something that energized them,
link |
something that gave them meaning,
link |
something that made them the best version of themselves,
link |
like made them more productive, not less,
link |
which is fascinating to me.
link |
It is fascinating.
link |
I mean, you can imagine if the way that you felt about Homer,
link |
the way that I feel and felt about Costello
link |
is at all a glimpse of what that must be like then.
link |
The downside, the thing I worry more about
link |
is the partner side of that.
link |
I've seen the kids are almost universally
link |
a source of increased productivity and joy and happiness.
link |
Like, yeah, they're a pain in the ass.
link |
Yeah, it's complicated.
link |
Yeah, so on and so forth.
link |
People like to complain about kids.
link |
But when you actually look past that little shallow layer
link |
of complaint, kids are great.
link |
The source of pain for a lot of people
link |
is when the relationship doesn't work.
link |
And so I'm very kind of concerned about,
link |
dating is very difficult and I'm a complicated person.
link |
And so it's been very difficult
link |
to find the right kind of person.
link |
But that statement doesn't even make sense
link |
because I'm not on dating apps.
link |
I don't see people.
link |
You're like the first person I saw in a while.
link |
It's like you, Michael Malice and like Joe.
link |
So I don't think I've seen like a female, what is it?
link |
An element of the female species in quite a while.
link |
So I think you have to put yourself out there.
link |
Daniel Johnson says, true love will find you,
link |
but only if you're looking.
link |
So there's some element of really taking the leap
link |
and putting yourself out there
link |
in kind of different situations.
link |
And I don't know how to do that
link |
when you're behind a computer all the time.
link |
Well, you're a builder and you're a problem solver
link |
and you find solutions and I'm confident this solution is,
link |
and the solution is out there.
link |
I think you're implying that I'm going to build
link |
the girlfriend, which I think-
link |
Or that you, well, and maybe we shouldn't separate
link |
this friendship, the notion of friendship and community.
link |
And if we go back to this concept of the aggregate,
link |
maybe you'll meet this woman through a friend
link |
or maybe or something of that sort.
link |
So one of the things, I don't know if you feel the same way.
link |
I definitely one of those people that just falls in love
link |
Yeah, I can't say I'm like that.
link |
With Costello, it was instantaneous.
link |
I mean, I know it's not romantic love,
link |
but it was instantaneous.
link |
No, but that's me.
link |
And I think that if you know, you know,
link |
because that's a good thing that you have that.
link |
Well, I'm very careful of that
link |
because you don't want to fall in love with the wrong person.
link |
So I try to be very kind of careful with,
link |
I've noticed this because I fall in love with everything,
link |
like this mug, everything.
link |
I fall in love with things in this world.
link |
So like, you have to be really careful
link |
because a girl comes up to you and says,
link |
she loves Dostoevsky.
link |
That doesn't necessarily mean you need to marry her tonight.
link |
Yes, and I liked the way you said that out loud
link |
so that you heard it.
link |
It doesn't mean you need to marry her tonight.
link |
But people are amazing and people are beautiful.
link |
And that's, so I'm fully embraced that,
link |
but I also, you have to be careful with relationships.
link |
And at the same time, like I mentioned to you offline,
link |
I don't, there's something about me that appreciates
link |
swinging for the fences and not dating,
link |
like doing serial dating or dating around.
link |
Yeah, you're a one guy, one girl kind of guy.
link |
And it's tricky because you want to be careful
link |
with that kind of stuff.
link |
Especially now there's a growing platform
link |
that have a ridiculous amount of female interest
link |
of a certain kind, but I'm looking for deep connection.
link |
And I'm looking by sitting home alone
link |
and every once in a while, talking to Stanford professors.
link |
It's going to work out great.
link |
It's well incorporated.
link |
It's part of, that constitutes machine learning of sorts.
link |
I do, you mentioned what has now become a quite extensive
link |
and expansive public platform, which is incredible.
link |
I mean, the number of people out,
link |
first time I saw your podcast, I noticed the suit.
link |
I was like, he respects his audience, which was great.
link |
But I also thought, this is amazing.
link |
People are showing up for science and engineering
link |
and technology information and those discussions
link |
and other sorts of discussions.
link |
Now, I do want to talk for a moment about the podcast.
link |
So my two questions about the podcast are,
link |
when you started it, did you have a plan?
link |
And regardless of what that answer is,
link |
do you know where you're taking it?
link |
Or would you like to leave us?
link |
I do believe in an element of surprise is always fun.
link |
But what about the podcast?
link |
Do you enjoy the podcast?
link |
I mean, your audience certainly includes me,
link |
really enjoys the podcast.
link |
So I love talking to people
link |
and there's something about microphones
link |
that really bring out the best in people.
link |
Like you don't get a chance to talk like this.
link |
If you and I were just hanging out,
link |
we would have a very different conversation
link |
in the amount of focus we allocate to each other.
link |
We would be having fun talking about other stuff
link |
and doing other things.
link |
There'd be a lot of distraction.
link |
There would be some phone use and all that kind of stuff.
link |
But here we're 100% focused on each other
link |
and focused on the idea.
link |
And sometimes playing with ideas
link |
that we both don't know the answer to,
link |
like a question we don't know the answer to.
link |
We're both like fumbling with it, trying to figure out,
link |
trying to get some insights
link |
at something we haven't really figured out before
link |
and together arriving at that.
link |
I think that's magical.
link |
I don't know why we need microphones for that,
link |
but we somehow do.
link |
It feels like doing science.
link |
It feels like doing science for me, definitely.
link |
That's exactly it.
link |
Then, and I'm really glad you said that
link |
because I don't actually often say this,
link |
but that's exactly what I felt like.
link |
I wanted to talk to friends and colleagues at MIT
link |
to do real science together.
link |
That's how I felt about it.
link |
Like to really talk to problems
link |
that are actually interesting
link |
as opposed to like incremental work
link |
that we're currently working for a particular conference.
link |
So really asking questions like, what are we doing?
link |
Like, where's this headed to?
link |
Like, what are the big,
link |
is this really going to help us solve,
link |
in the case of AI, solve intelligence?
link |
Like, is this even working on intelligence?
link |
There's a certain sense,
link |
which is why I initially called it artificial intelligence
link |
is like most of us are not working
link |
on artificial intelligence.
link |
You're working on some very specific problem
link |
and a set of techniques.
link |
At the time, it's machine learning
link |
to solve this particular problem.
link |
This is not going to take us to a system
link |
that is anywhere close to the generalizability
link |
of the human mind.
link |
Like the kind of stuff the human mind can do
link |
in terms of memory, in terms of cognition,
link |
in terms of reasoning, common sense reasoning.
link |
This doesn't seem to take us there.
link |
So the initial impulse was,
link |
can I talk to these folks,
link |
do science together through conversation?
link |
And I also thought that there was not enough,
link |
now, I didn't think there was enough good conversations
link |
with world-class minds that I got to meet
link |
and not the ones with a book or this was the thing.
link |
Oftentimes you go on this tour when you have a book,
link |
but there's a lot of minds that don't write books.
link |
And the books constrain the conversation too,
link |
because then you're talking about this thing, this book.
link |
But there's, I've noticed that with people
link |
that haven't written a book who are brilliant,
link |
we get to talk about ideas in a new way.
link |
We both haven't actually,
link |
when we raise a question,
link |
we don't know the answer to it
link |
when the question is raised and we try to arrive there.
link |
Like, I don't know.
link |
I remember asking questions
link |
of world-class researchers in deep learning
link |
of why do neural networks work as well as they do?
link |
That question is often loosely asked,
link |
but like when you have microphones
link |
and you have to think through it
link |
and you have 30 minutes to an hour
link |
to think through it together, I think that's science.
link |
I think that's really powerful.
link |
So that was the one goal.
link |
I again don't usually talk about this,
link |
but there's some sense in which I wanted
link |
to have dangerous conversations.
link |
Part of the reasons I wanted to wear a suit is like,
link |
I want it to be fearless.
link |
Now, the reason I don't usually talk about it
link |
is because I feel like I'm not good at conversation.
link |
So it looks like it doesn't match the current skill level,
link |
but I wanted to have really dangerous conversations
link |
that I uniquely would be able to do.
link |
Not completely uniquely, but like I'm a huge fan
link |
of Joe Rogan and I had to ask myself,
link |
what conversations can I do that Joe Rogan can't?
link |
For me, I know I bring this up,
link |
but for me, that person I thought about
link |
at the time was Putin.
link |
Like that's why I bring him up.
link |
He's just like with Costello, he's not just a person.
link |
He's also an idea to me for what I strive for,
link |
just to have those dangerous conversations.
link |
And the reason I'm uniquely qualified is both the Russian,
link |
but also there's the judo and the martial arts.
link |
There's a lot of elements that make me have a conversation
link |
he hasn't had before.
link |
And there's a few other people that I kept in mind,
link |
like Don Knuth, he's a computer scientist from Stanford
link |
that I thought is one of the most beautiful minds ever.
link |
And nobody really talked to him, like really talked to him.
link |
He did a few lectures, which people love,
link |
but really just have a conversation with him.
link |
There's a few people like that.
link |
One of them passed away, John Conway, that I never got.
link |
We agreed to talk, but he died before we didn't.
link |
There's a few people like that that I thought like,
link |
it's such a crime to not hear those folks.
link |
And I have the unique ability to know how to purchase
link |
a microphone on Amazon and plug it into a device
link |
that records audio and then publish it,
link |
which seems relatively unique.
link |
Like that's not easy in the scientific community.
link |
People knowing how to plug in a microphone.
link |
No, they can build Faraday cages and two photon microscopes
link |
and bio-engineer, all sorts of things.
link |
But the idea that you could take ideas and export them
link |
into a structure or a pseudo structure
link |
that people would benefit from
link |
seems like a cosmic achievement to them.
link |
I don't know if it's a fear or just basically
link |
they haven't tried it,
link |
so they haven't learned the skill level.
link |
I think they're not trained, I mean,
link |
we could riff on this for a while,
link |
but I think that, but it's important and maybe we should,
link |
which is that it's, they're not trained to do it.
link |
They're trained to think in specific aims
link |
and specific hypotheses.
link |
And many of them don't care to, right?
link |
They became scientists because that's where they felt safe.
link |
And so why would they leave that haven of safety?
link |
Well, they also don't necessarily always see
link |
We're all together learning,
link |
you and I are learning the value of this.
link |
I think you're probably,
link |
you have an exceptionally successful and amazing podcast
link |
that you started just recently.
link |
Thanks to your encouragement.
link |
Well, but there's a raw skill there
link |
that you're definitely an inspiration to me
link |
in how you do the podcast
link |
in the level of excellence you reach.
link |
But I think you've discovered
link |
that that's also an impactful way to do science,
link |
And I think a lot of scientists have not yet discovered
link |
that this is, if they apply same kind of rigor
link |
as they do to academic publication
link |
or to even conference presentations,
link |
and they do that rigor and effort to podcast,
link |
whatever that is, that could be a five minute podcast,
link |
a two hour podcast, it could be conversational,
link |
or it can be more like lecture-like.
link |
If they apply that effort,
link |
you have the potential to reach over time,
link |
tens of thousands, hundreds of thousands,
link |
millions of people.
link |
And that's really, really powerful.
link |
giving a platform to a few of those folks,
link |
especially for me personally,
link |
so maybe you can speak to what fields you're drawn to,
link |
but I thought computer scientists
link |
were especially bad at this.
link |
So there's brilliant computer scientists
link |
that I thought it would be amazing to explore their mind,
link |
explore their thinking.
link |
And so I took that almost on as an effort.
link |
And at the same time, I had other guests in mind
link |
or people that connect to my own interests.
link |
wrestling, music, football,
link |
both American football and soccer.
link |
I have a few particular people
link |
that I'm really interested in.
link |
Bovisar Satiev, the Satiev brothers,
link |
even Khabib for wrestling,
link |
just to talk to them.
link |
Because you guys can communicate.
link |
In Russian and in wrestling,
link |
as wrestlers and as Russians.
link |
And so that little,
link |
it's like an opportunity to explore a mind
link |
that I'm able to bring to the world.
link |
And also, I feel like it makes me a better person
link |
just that being that vulnerable
link |
and exploring ideas together.
link |
I don't know, like good conversation.
link |
I don't know how often you have a really good conversation
link |
with friends, but like podcasts are like that.
link |
And it's deeply moving.
link |
And what you brought through,
link |
I mean, when I saw you sit down with Penrose,
link |
Nobel Prize winning physicists and these other folks,
link |
it's not just because he has a Nobel,
link |
it's what comes out of his mouth is incredible.
link |
And what you were able to hold in that conversation
link |
was so much better.
link |
Light years beyond what he had any other interviewer,
link |
I don't want to even call you an interviewer
link |
because it's really about conversation.
link |
Light years beyond what anyone else had been able
link |
to engage with him was such a beacon of what's possible.
link |
And I know that, I think that's what people are drawn to.
link |
And there's a certain intimacy
link |
that certainly if two people are friends as we are
link |
and they know each other, that there's more of that,
link |
but there's an intimacy in those kinds
link |
of private conversations that are made public.
link |
Well, that's the, with you,
link |
you're probably starting to realize, and Costello,
link |
it's like part of it, because you're authentic
link |
and you're putting yourself out there completely,
link |
people are almost not just consuming
link |
the words you're saying, they also enjoy watching you,
link |
Andrew, struggle with these ideas
link |
or try to communicate these ideas.
link |
They like the flaws.
link |
They like a human being exploring ideas.
link |
Well, that's good, because I got plenty of those.
link |
Well, they like the self-critical aspects,
link |
like where you're very careful,
link |
where you're very self-critical about your flaws.
link |
I mean, in that same way, it's interesting,
link |
I think, for people to watch me talk to Penrose,
link |
not just because Penrose is communicating ideas,
link |
but here's this like silly kid trying to explore ideas.
link |
Like they know this kid,
link |
that there's a human connection that is really powerful.
link |
Same, I think, with Putin, right?
link |
Like it's not just a good interview with Putin.
link |
It's also, here's this kid struggling
link |
to talk with one of the most powerful,
link |
some would argue dangerous people in the world,
link |
that they love that, the authenticity that led up to that.
link |
Like, and in return, I get to connect everybody I run to
link |
in the street and all those kinds of things.
link |
There's a depth of connection there,
link |
almost within like a minute or two,
link |
that's unlike any other.
link |
Yeah, there's an intimacy that you've formed with them.
link |
Yeah, we've been on this like journey together.
link |
I mean, I have the same thing with Joe Rogan
link |
before I ever met him, right?
link |
Like I was, because I was a fan of Joe for so many years,
link |
there's something, there's a kind of friendship
link |
as absurd as it might be to say in podcasting
link |
and listening to podcasts.
link |
Yeah, maybe it fills in a little bit of that,
link |
or solves a little bit of that loneliness
link |
that you're talking about.
link |
Until the robots are here.
link |
I have just a couple more questions,
link |
but one of them is on behalf of your audience,
link |
which is, I'm not going to ask you
link |
the meaning of the hedgehog,
link |
but I just want to know, does it have a name?
link |
And you don't have to tell us the name,
link |
but just does it have a name, yes or no?
link |
Well, there's a name he likes to be referred to as,
link |
and then there's a private name
link |
in the privacy of our own company that we call each other.
link |
No, I'm not that insane.
link |
No, his name is Hedgy.
link |
I don't like stuffed animals,
link |
but his story is one of minimalism.
link |
So I gave away everything I own now three times in my life.
link |
By everything, I mean, almost everything,
link |
kept jeans and shirt and a laptop.
link |
And recently it's also been guitar, things like that.
link |
But he survived because he was always in the,
link |
at least in the first two times, was in the laptop bag,
link |
and he just got lucky.
link |
And so I just liked the perseverance of that.
link |
And I first saw him in the,
link |
the reason I got a stuffed animal
link |
and I don't have other stuffed animals
link |
is it was in a thrift store
link |
in this like giant pile of stuffed animals.
link |
And he jumped out at me because unlike all the rest of them,
link |
he has this intense mean look about him,
link |
that he's just, he's upset at life,
link |
at the cruelty of life.
link |
And just, especially in the contrast of the other
link |
stuffed animals, they have this dumb smile on their face.
link |
If you look at most stuffed animals,
link |
they have this dumb look on their face
link |
and they're just happy.
link |
It's like Pleasantville.
link |
It's what we say in neuroscience,
link |
they have a smooth cortex, not many fold.
link |
And this, like, Hegyi like saw through all of it.
link |
He was like Dostoevsky's man from underground.
link |
I mean, there's a sense that he saw the darkness
link |
of the world and persevered.
link |
So I got, and there's also a famous Russian cartoon,
link |
Hedgehog in the Fog, that I grew up with,
link |
There's people who know of that cartoon.
link |
You can see it on YouTube.
link |
Hedgehog in the Fog.
link |
It's just as you would expect,
link |
especially from like early Soviet cartoons.
link |
It's a hedgehog, like sad, walking through the fog,
link |
exploring like loneliness and sadness.
link |
It's like, but it's beautiful.
link |
It's like a piece of art.
link |
People should, even if you don't speak Russian,
link |
you'll see, you'll understand.
link |
Oh, it's like the moment you said that I was gonna ask.
link |
So it's in Russian, but of course it's in Russian.
link |
It's in Russian, but it's more,
link |
there's very little speaking in it.
link |
It's almost, there's an interesting exploration
link |
of how you make sense of the world
link |
when you see it only vaguely through the fog.
link |
So he's trying to understand the world.
link |
Here we have Mickey Mouse.
link |
We have Bugs Bunny.
link |
We have all these crazy animals,
link |
and you have the hedgehog in the fog.
link |
So there's a certain period, and this is, again,
link |
I don't know what to attribute it to,
link |
but it was really powerful,
link |
which there's a period in Soviet history,
link |
I think probably 70s and 80s,
link |
where like, especially kids were treated very seriously.
link |
Like they were treated like they're able to deal
link |
with the weightiness of life.
link |
And that was reflected in the cartoons.
link |
And it was allowed to have really artistic content,
link |
not like dumb cartoons that are trying
link |
to get you to be like smile and run around,
link |
but like create art.
link |
Like stuff that, you know how like short cartoons
link |
or short films can win Oscars?
link |
Like that's what they're swinging for.
link |
So what strikes me about this is a little bit
link |
how we were talking about the suit earlier.
link |
It's almost like they treat kids with respect.
link |
Like they have an intelligence
link |
and they honor that intelligence.
link |
Yeah, they're really just adult in a small body.
link |
Like you want to protect them
link |
from the true cruelty of the world,
link |
but in terms of their intellectual capacity
link |
or like philosophical capacity,
link |
they're right there with you.
link |
And so the cartoons reflected that,
link |
the art that they consumed, the education reflected that.
link |
So he represents that.
link |
I mean, there's a sense of because he survived so long
link |
and because I don't like stuffed animals,
link |
that it's like, we've been through all of this together
link |
and it's the same sharing the moments together.
link |
It's the friendship.
link |
And there's a sense in which, you know,
link |
if all the world turns on you and goes to hell,
link |
at least we got each other.
link |
That, and he doesn't die because he's an inanimate object.
link |
Until you animate him.
link |
Until you animate him.
link |
And then I probably wouldn't want to know
link |
what he was thinking about this whole time.
link |
He's probably really into Taylor Swift
link |
or something like that.
link |
It's like that I wouldn't even want to know.
link |
Well, I now feel a connection to Hedgy the Hedgehog
link |
that I certainly didn't have before.
link |
And I think that encapsulates the kind of possibility
link |
of connection that is possible
link |
between human and other object
link |
and through robotics, certainly.
link |
There's a saying that I heard when I was a graduate student
link |
that's just been ringing in my mind
link |
throughout this conversation in such a,
link |
I think, appropriate way,
link |
which is that Lex, you are in a minority of one.
link |
You are truly extraordinary in your ability
link |
to encapsulate so many aspects of science, engineering,
link |
public communication about so many topics,
link |
martial arts, and the emotional depth that you bring to it,
link |
and just the purposefulness.
link |
And I think if it's not clear to people,
link |
it absolutely should be stated,
link |
but I think it's abundantly clear
link |
that just the amount of time and thinking
link |
that you put into things is,
link |
it is the ultimate mark of respect.
link |
So I'm just extraordinarily grateful for your friendship
link |
and for this conversation.
link |
I'm proud to be your friend.
link |
And I just wish you showed me the same kind of respect
link |
by wearing a suit and make your father proud,
link |
Next time, indeed.
link |
Thanks so much, my friend.
link |
Thank you. Thank you, Andrew.
link |
Thank you for joining me for my discussion
link |
with Dr. Lex Friedman.
link |
If you're enjoying this podcast and learning from it,
link |
please consider subscribing on YouTube.
link |
As well, you can subscribe to us on Spotify or Apple.
link |
Please leave any questions and comments and suggestions
link |
that you have for future podcast episodes and guests
link |
in the comment section on YouTube.
link |
At Apple, you can also leave us up to a five-star review.
link |
If you'd like to support this podcast, we have a Patreon.
link |
That's patreon.com slash Andrew Huberman.
link |
And there, you can support us at any level that you like.
link |
Also, please check out our sponsors mentioned
link |
at the beginning of the podcast episode.
link |
That's the best way to support this podcast.
link |
Links to our sponsors can be found in the show notes.
link |
And finally, thank you for your interest in science.
link |
Letter on the prefers"]