Let’s talk about free speech and censorship.

You don’t build a company like this

unless you believe that people expressing themselves

is a good thing.

Let me ask you as a father,

there’s a weight heavy on you

that people get bullied on social networks.

I care a lot about how people feel

when they use our products

and I don’t want to build products that make people angry.

Why do you think so many people dislike you?

Some even hate you.

And how do you regain their trust and support?

The following is a conversation with Mark Zuckerberg,

CEO of Facebook, now called Meta.

Please allow me to say a few words

about this conversation with Mark Zuckerberg,

about social media,

and about what troubles me in the world today,

and what gives me hope.

If this is not interesting to you,

I understand, please skip.

I believe that at its best,

social media puts a mirror to humanity

and reveals the full complexity of our world,

shining a light on the dark aspects of human nature

and giving us hope, a way out,

through compassionate but tense chaos of conversation

that eventually can turn into understanding,

friendship, and even love.

But this is not simple.

Our world is not simple.

It is full of human suffering.

I think about the hundreds of millions of people

who are starving and who live in extreme poverty,

the one million people who take their own life every year,

the 20 million people that attempt it,

and the many, many more millions who suffer quietly

in ways that numbers can never know.

I’m troubled by the cruelty and pain of war.

Today, my heart goes out to the people of Ukraine.

My grandfather spilled his blood on this land,

held the line as a machine gunner

against the Nazi invasion, surviving impossible odds.

I am nothing without him.

His blood runs in my blood.

My words are useless here.

I send my love.

It’s all I have.

I hope to travel to Russia and Ukraine soon.

I will speak to citizens and leaders,

including Vladimir Putin.

As I’ve said in the past, I don’t care about access,

fame, money, or power, and I’m afraid of nothing.

But I am who I am, and my goal in conversation

is to understand the human being before me,

no matter who they are, no matter their position.

And I do believe the line between good and evil

runs through the heart of every man.

So this is it.

This is our world.

It is full of hate, violence, and destruction.

But it is also full of love, beauty,

and the insatiable desire to help each other.

The people who run the social networks

that show this world, that show us to ourselves,

have the greatest of responsibilities.

In a time of war, pandemic, atrocity,

we turn to social networks to share real human insights

and experiences, to organize protests and celebrations,

to learn and to challenge our understanding of the world,

of our history and of our future,

and above all, to be reminded of our common humanity.

When the social networks fail,

they have the power to cause immense suffering.

And when they succeed,

they have the power to lessen that suffering.

This is hard.

It’s a responsibility, perhaps,

almost unlike any other in history.

This podcast conversation attempts to understand the man

and the company who take this responsibility on,

where they fail and where they hope to succeed.

Mark Zuckerberg’s feet are often held to the fire,

as they should be, and this actually gives me hope.

The power of innovation and engineering,

coupled with the freedom of speech

in the form of its highest ideal,

I believe can solve any problem in the world.

But that’s just it, both are necessary,

the engineer and the critic.

I believe that criticism is essential, but cynicism is not.

And I worry that in our public discourse,

cynicism too easily masquerades as wisdom, as truth,

becomes viral and takes over,

and worse, suffocates the dreams of young minds

who want to build solutions to the problems of the world.

We need to inspire those young minds.

At least for me, they give me hope.

And one small way I’m trying to contribute

is to have honest conversations like these

that don’t just ride the viral wave of cynicism,

but seek to understand the failures

and successes of the past, the problems before us,

and the possible solutions

in this very complicated world of ours.

I’m sure I will fail often,

and I count on the critic to point it out when I do.

But I ask for one thing,

and that is to fuel the fire of optimism,

especially in those who dream to build solutions,

because without that, we don’t have a chance

on this too fragile, tiny planet of ours.

This is the Lex Friedman podcast.

To support it, please check out our sponsors

in the description.

And now, dear friends, here’s Mark Zuckerberg.

Can you circle all the traffic lights, please?

You actually did it.

That is very impressive performance.

Okay, now we can initiate the interview procedure.

Is it possible that this conversation is happening

inside a metaverse created by you,

by Meta many years from now,

and we’re doing a memory replay experience?

I don’t know the answer to that.

Then I’d be some computer construct

and not the person who created that Meta company.

But that would truly be Meta.

Right, so this could be somebody else

using the Mark Zuckerberg avatar

who can do the Mark and the Lex conversation replay

from four decades ago when Meta, it was first sort of.

I mean, it’s not gonna be four decades

before we have photorealistic avatars like this.

So I think we’re much closer to that.

Well, that’s something you talk about

is how passionate you are about the idea

of the avatar representing who you are in the metaverse.

So I do these podcasts in person.

You know, I’m a stickler for that,

because there’s a magic to the in person conversation.

How long do you think it’ll be before

you can have the same kind of magic in the metaverse,

the same kind of intimacy in the chemistry,

whatever the heck is there when we’re talking in person?

How difficult is it?

How long before we have it in the metaverse?

Well, I think this is like the key question, right?

Because the thing that’s different about virtual

and hopefully augmented reality

compared to all other forms of digital platforms before

is this feeling of presence, right?

The feeling that you’re right,

that you’re in an experience

and that you’re there with other people or in another place.

And that’s just different from all of the other screens

that we have today, right?

Phones, TVs, all the stuff.

They’re trying to, in some cases, deliver experiences

that feel high fidelity,

but at no point do you actually feel like you’re in it, right?

At some level, your content is trying to sort of convince you

that this is a realistic thing that’s happening,

but all of the kind of subtle signals are telling you,

no, you’re looking at a screen.

So the question about how you develop these systems is like,

what are all of the things that make the physical world

all the different cues?

So I think on visual presence and spatial audio,

we’re making reasonable progress.

Spatial audio makes a huge deal.

I don’t know if you’ve tried this experience,

workrooms that we launched where you have meetings.

And I basically made a rule for all of the top,

you know, management folks at the company

that they need to be doing standing meetings

in workrooms already, right?

I feel like we got to dog food this,

you know, this is how people are gonna work in the future.

So we have to adopt this now.

And there were already a lot of things

that I think feel significantly better

than like typical Zoom meetings,

even though the avatars are a lot lower fidelity.

You know, the idea that you have spatial audio,

you’re around a table in VR with people.

If someone’s talking from over there,

it sounds like it’s talking from over there.

You can see, you know, the arm gestures

and stuff feel more natural.

You can have side conversations,

which is something that you can’t really do in Zoom.

I mean, I guess you can text someone over,

like out of band,

but if you’re actually sitting around a table with people,

you know, you can lean over

and whisper to the person next to you

and like have a conversation that you can’t,

you know, that you can’t really do

with in just video communication.

So I think it’s interesting in what ways

some of these things already feel more real

than a lot of the technology that we have,

even when the visual fidelity isn’t quite there,

but I think it’ll get there over the next few years.

Now, I mean, you were asking about comparing that

to the true physical world,

not Zoom or something like that.

And there, I mean, I think you have feelings

of like temperature, you know, olfactory,

obviously touch, right, we’re working on haptic gloves,

you know, the sense that you wanna be able to,

you know, put your hands down

and feel some pressure from the table.

You know, all of these things

I think are gonna be really critical

to be able to keep up this illusion

that you’re in a world

and that you’re fully present in this world.

But I don’t know,

I think we’re gonna have a lot of these building blocks

within, you know, the next 10 years or so.

And even before that, I think it’s amazing

how much you’re just gonna be able to build with software

that sort of masks some of these things.

I realize I’m going long,

but I was told we have a few hours here.

So it’s a…

We’re here for five to six hours.

Yeah, so I mean, it’s, look,

I mean, that’s on the shorter end

of the congressional testimonies I’ve done.

But it’s, but, you know, one of the things

that we found with hand presence, right?

So the earliest VR, you just have the headset

and then, and that was cool, you could look around,

you feel like you’re in a place,

but you don’t feel like you’re really able to interact with it

until you have hands.

And then there was this big question

where once you got hands,

what’s the right way to represent them?

And initially, all of our assumptions was, okay,

when I look down and see my hands in the physical world,

I see an arm and it’s gonna be super weird

if you see, you know, just your hand.

But it turned out to not be the case

because there’s this issue with your arms,

which is like, what’s your elbow angle?

And if the elbow angle that we’re kind of interpolating

based on where your hand is and where your headset is

actually isn’t accurate,

it creates this very uncomfortable feeling

where it’s like, oh, like my arm is actually out like this,

but it’s like showing it in here.

And that actually broke the feeling of presence a lot more.

Whereas it turns out that if you just show the hands

and you don’t show the arms,

it actually is fine for people.

So I think that there’s a bunch

of these interesting psychological cues

where it’ll be more about getting the right details right.

And I think a lot of that will be possible

even over a few year period or a five year period.

And we won’t need like every single thing to be solved

to deliver this like full sense of presence.

Yeah, it’s a fascinating psychology question

of what is the essence

that makes in person conversation special?

It’s like emojis are able to convey emotion really well,

even though they’re obviously not photorealistic.

And so in that same way, Jessica, you’re saying,

just showing the hands is able

to create a comfortable expression with your hands.

So I wonder what that is.

People in the world wars used to write letters

and you can fall in love with just writing letters.

You don’t need to see each other in person.

You can convey emotion.

You can be depth of experience with just words.

So that’s, I think, a fascinating place

to explore psychology of like,

how do you find that intimacy?

Yeah, and the way that I come to all of this stuff is,

I basically studied psychology and computer science.

So all of the work that I do

is sort of at the intersection of those things.

I think most of the other big tech companies

are building technology for you to interact with.

What I care about is building technology

to help people interact with each other.

So I think it’s a somewhat different approach

than most of the other tech entrepreneurs

and big companies come at this from.

And a lot of the lessons

in terms of how I think about designing products

come from some just basic elements of psychology, right?

In terms of our brains,

you can compare it to the brains of other animals.

We’re very wired to specific things, facial expressions.

I mean, we’re very visual, right?

So compared to other animals,

I mean, that’s clearly the main sense

that most people have.

But there’s a whole part of your brain

that’s just kind of focused on reading facial cues.

So when we’re designing the next version of Quest

or the VR headset, a big focus for us is face tracking

and basically eye tracking so you can make eye contact,

which again, isn’t really something

that you can do over a video conference.

It’s sort of amazing how far video conferencing

has gotten without the ability to make eye contact, right?

It’s sort of a bizarre thing if you think about it.

You’re looking at someone’s face,

sometimes for an hour when you’re in a meeting

and you looking at their eyes to them

doesn’t look like you’re looking at their eyes.

You’re always looking past each other, I guess.

I guess you’re right.

You’re not sending that signal.

Well, you’re trying to.

Right, you’re trying to.

A lot of times, or at least I find myself,

I’m trying to look into the other person’s eyes.

But they don’t feel like you’re looking to their eyes.

So then the question is,

all right, am I supposed to look at the camera

so that way you can have a sensation

that I’m looking at you?

I think that that’s an interesting question.

And then with VR today,

even without eye tracking

and knowing what your eyes are actually looking at,

you can fake it reasonably well, right?

So you can look at where the head pose is.

And if it looks like I’m kind of looking

in your general direction,

then you can sort of assume

that maybe there’s some eye contact intended

and you can do it in a way where it’s like,

okay, maybe it’s not a fixated stare,

but it’s somewhat natural.

But once you have actual eye tracking,

you can do it for real.

And I think that that’s really important stuff.

So when I think about Meta’s contribution to this field,

I have to say it’s not clear to me

that any of the other companies

that are focused on the Metaverse

or on virtual and augmented reality

are gonna prioritize putting these features in the hardware

because like everything, they’re trade offs, right?

I mean, it adds some weight to the device.

Maybe it adds some thickness.

You could totally see another company taking the approach

of let’s just make the lightest and thinnest thing possible.

But I want us to design the most human thing possible

that creates the richest sense of presence

and cause so much of human emotion and expression

comes from these like micro movements.

If I like move my eyebrow millimeter,

you will notice and that like means something.

So the fact that we’re losing these signals

and a lot of communication I think is a loss.

So it’s not like, okay, there’s one feature

and you add this, then it all of a sudden

is gonna feel like we have real presence.

You can sort of look at how the human brain works

and how we express and kind of read emotions

and you can just build a roadmap of that,

of just what are the most important things

to try to unlock over a five to 10 year period

and just try to make the experience

more and more human and social.

When do you think would be a moment,

like a singularity moment for the Metaverse

where there’s a lot of ways to ask this question,

but people will have many or most

of their meaningful experiences

in the Metaverse versus the real world.

And actually it’s interesting to think about

the fact that a lot of people are having

the most important moments of their life

happen in the digital sphere,

especially not during COVID,

like even falling in love or meeting friends

or getting excited about stuff

that is happening on the 2D digital plane.

When do you think the Metaverse

will provide those experiences for a large number,

like a majority of the population?

Yeah, I think it’s a really good question.

There was someone, I read this piece

that framed this as a lot of people think

that the Metaverse is about a place,

but one definition of this is it’s about a time

when basically immersive digital worlds

become the primary way that we live our lives

and spend our time.

I think that that’s a reasonable construct.

And from that perspective,

I think you also just wanna look at this as a continuation

because it’s not like, okay,

we are building digital worlds,

but we don’t have that today.

I think you and I probably already live

a very large part of our life in digital worlds.

They’re just not 3D immersive virtual reality,

but I do a lot of meetings over video

or I spend a lot of time writing things over email

or WhatsApp or whatever.

So what is it gonna take to get there

for kind of the immersive presence version of this,

which I think is what you’re asking.

And for that, I think that there’s just a bunch

of different use cases.

And I think when you’re building technology,

I think a lot of it is just you’re managing this duality

where on the one hand,

you wanna build these elegant things that can scale

and have billions of people use them

and get value from them.

And then on the other hand,

you’re fighting this kind of ground game

where there are just a lot of different use cases

and people do different things

and you wanna be able to unlock them.

So the first ones that we basically went after

were gaming with Quest and social experiences.

And it goes back to when we started working

on virtual reality.

My theory at the time was basically

people thought about it as gaming,

but if you look at all computing platforms up to that point,

gaming is a huge part, it was a huge part of PCs,

it was a huge part of mobile,

but it was also very decentralized.

There wasn’t, for the most part,

one or two gaming companies.

There were a lot of gaming companies

and gaming is somewhat hits based.

I mean, we’re getting some games that have more longevity,

but in general, there were a lot of different games

out there.

But on PC and on mobile,

the companies that focused on communication

and social interaction,

there tended to be a smaller number of those

and that ended up being just as important of a thing

as all of the games that you did combined.

I think productivity is another area.

That’s obviously something

that we’ve historically been less focused on,

but I think it’s gonna be really important for us.

With workroom, do you mean productivity

in the collaborative aspect?

Yeah, I think that there’s a workroom’s aspect of this,

like a meeting aspect,

and then I think that there’s like a Word, Excel,

productivity, either you’re working or coding

or knowledge work as opposed to just meetings.

So you can kind of go through all these different use cases.

Gaming, I think we’re well on our way.

Social, I think we’re just the kind of preeminent company

that focuses on this.

And I think that that’s already on Quest becoming the,

if you look at the list of what are the top apps,

social apps are already number one, two, three.

So that’s kind of becoming a critical thing, but I don’t know.

I would imagine for someone like you,

it’ll be until we get a lot of the work things dialed in.

When this is just like much more adopted

and clearly better than Zoom for VC,

when if you’re doing your coding or your writing

or whatever it is in VR,

which it’s not that far off to imagine that

because pretty soon you’re just gonna be able

to have a screen that’s bigger than,

it’ll be your ideal setup and you can bring it with you

and put it on anywhere

and have your kind of ideal workstation.

So I think that there are a few things to work out on that,

but I don’t think that that’s more than five years off.

And then you’ll get a bunch of other things

that like aren’t even possible

or you don’t even think about using a phone

or PC for today, like fitness, right?

So, I mean, I know you’re, we were talking before

about how you’re into running

and like I’m really into a lot of things

around fitness as well,

different things in different places.

I got really into hydrofoiling recently

and surfing and I used to fence competitively.

I like run.

So, and you were saying that you were thinking

about trying different martial arts

and I tried to trick you and convince you

into doing Brazilian Jiu Jitsu.

Or you actually mentioned that that was one

you’re curious about and I don’t know.

Is that a trick?

Yeah, I don’t know.

We’re in the metaverse now.

Yeah, no, I took that seriously.

I thought that that was a real suggestion.

That would be an amazing chance

if we ever step on the mat together

and just like roll around.

I’ll show you some moves.

Well, give me a year to train and then we can do it.

You know, you’ve seen Rocky IV

where the Russian faces off the American.

I’m the Russian in this picture.

And then you’re the Rocky, the underdog

that gets to win in the end.

The idea of me as Rocky and like fighting is…

If he dies, he dies.

Sorry, I just had to.

I mean.

Anyway, yeah.

But I mean, a lot of aspects of fitness.

You know, I don’t know if you’ve tried supernatural

on Quest or…

So first of all, can I just comment on the fact

every time I played around with Quest 2,

I just, I get giddy every time I step into virtual reality.

So you mentioned productivity and all those kinds of things.

That’s definitely something I’m excited about,

but really I just love the possibilities

of stepping into that world.

Maybe it’s the introvert in me,

but it just feels like the most convenient way

to travel into worlds,

into worlds that are similar to the real world

or totally different.

It’s like Alice in Wonderland.

Just try out crazy stuff.

The possibilities are endless.

And I just, I personally am just love,

get excited for stepping in those virtual worlds.

So I’m a huge fan.

In terms of the productivity as a programmer,

I spend most of my day programming.

That’s really interesting also,

but then you have to develop the right IDEs.

You have to develop, like there has to be a threshold

where a large amount of the programming community

moves there, but the collaborative aspects

that are possible in terms of meetings,

in terms of when two coders are working together,

I mean, the possibilities there are super, super exciting.

I think that in building this, we sort of need to balance.

There are gonna be some new things

that you just couldn’t do before.

And those are gonna be the amazing experiences.

So teleporting to any place, right?

Whether it’s a real place or something that people made.

And I mean, some of the experiences

around how we can build stuff in new ways,

where a lot of the stuff that,

when I’m coding stuff, it’s like, all right,

you code it and then you build it

and then you see it afterwards.

But increasingly it’s gonna be possible to,

you’re in a world and you’re building the world

as you are in it and kind of manipulating it.

One of the things that we showed at our Inside the Lab

for recent artificial intelligence progress

is this Builder Bot program,

where now you can just talk to it and say,

hey, okay, I’m in this world,

like put some trees over there and it’ll do that.

And like, all right, put some bottles of water

on our picnic blanket and it’ll do that

and you’re in the world.

And I think there are gonna be new paradigms for coding.

So yeah, there are gonna be some things

that I think are just pretty amazing,

especially the first few times that you do them,

but that you’re like, whoa,

like I’ve never had an experience like this.

But most of your life, I would imagine,

is not doing things that are amazing for the first time.

A lot of this in terms of,

I mean, just answering your question from before around,

what is it gonna take

before you’re spending most of your time in this?

Well, first of all, let me just say it as an aside,

the goal isn’t to have people spend a lot more time

in computing.

It’s to make it so that. I’m asking for myself.

Yeah, it’s to make it. When will I spend all my time in?

Yeah, it’s to make computing more natural.

But I think you will spend most of your computing time

in this when it does the things

that you use computing for somewhat better.

So maybe having your perfect workstation

is a 5% improvement on your coding productivity.

Maybe it’s not like a completely new thing.

But I mean, look, if I could increase the productivity

of every engineer at Meta by 5%,

we’d buy those devices for everyone.

And I imagine a lot of other companies would too.

And that’s how you start getting to the scale

that I think makes this rival

some of the bigger computing platforms that exist today.

Let me ask you about identity.

We talked about the avatar.

How do you see identity in the Metaverse?

Should the avatar be tied to your identity

or can I be anything in the Metaverse?

Like, can I be whatever the heck I want?

Can I even be a troll?

So there’s exciting freeing possibilities

and there’s the darker possibilities too.

Yeah, I mean, I think that there’s gonna be a range, right?

So we’re working on, for expression and avatars,

on one end of the spectrum are kind of expressive

and cartoonish avatars.

And then on the other end of the spectrum

are photorealistic avatars.

And I just think the reality is

that there are gonna be different use cases

for different things.

And I guess there’s another axis.

So if you’re going from photorealistic to expressive,

there’s also like representing you directly

versus like some fantasy identity.

And I think that there are gonna be things

on all ends of that spectrum too, right?

So you’ll want photo, like in some experience,

you might wanna be like a photorealistic dragon, right?

Or if I’m playing Onward,

or just this military simulator game,

I think getting to be more photorealistic as a soldier

in that could enhance the experience.

There are times when I’m hanging out with friends

where I want them to know it’s me.

So a kind of cartoonish or expressive version of me is good.

But there are also experiences like,

VRChat does this well today,

where a lot of the experience is kind of dressing up

and wearing a fantastical avatar

that’s almost like a meme or is humorous.

So you come into an experience

and it’s almost like you have like a built in icebreaker

because like you see people and you’re just like,

all right, I’m cracking up at what you’re wearing

because that’s funny.

And it’s just like, where’d you get that?

Or, oh, you made that?

That’s, it’s awesome.

Whereas, okay, if you’re going into a work meeting,

maybe a photorealistic version of your real self

is gonna be the most appropriate thing for that.

So I think the reality is there aren’t going to be,

it’s not just gonna be one thing.

You know, my own sense of kind of how you wanna

express identity online has sort of evolved over time.

And that, you know, early days in Facebook,

I thought, okay, people are gonna have one identity.

And now I think that’s clearly not gonna be the case.

I think you’re gonna have all these different things

and there’s utility in being able to do different things.

So some of the technical challenges

that I’m really interested in around it

are how do you build the software

to allow people to seamlessly go between them?

So say, so you could view them

as just completely discrete points on a spectrum,

but let’s talk about the metaverse economy for a second.

Let’s say I buy a digital shirt

for my photorealistic avatar, which by the way,

I think at the time where we’re spending a lot of time

in the metaverse doing a lot of our work meetings

in the metaverse and et cetera,

I would imagine that the economy around virtual clothing

as an example is going to be quite as big.

Why wouldn’t I spend almost as much money

in investing in my appearance or expression

for my photorealistic avatar for meetings

as I would for whatever I’m gonna wear in my video chat.

But the question is, okay, so you,

let’s say you buy some shirt

for your photorealistic avatar.

Wouldn’t it be cool if there was a way

to basically translate that into a more expressive thing

for your kind of cartoonish or expressive avatar?

And there are multiple ways to do that.

You can view them as two discrete points and okay,

maybe if a designer sells one thing,

then it actually comes in a pack and there’s two

and you can use either one on that,

but I actually think this stuff might exist more

as a spectrum in the future.

And that’s what I do think the direction

on some of the AI advances that is happening

to be able to, especially stuff around like style transfer,

being able to take a piece of art or express something

and say, okay, paint me this photo in the style of Gauguin

or whoever it is that you’re interested in.

Take this shirt and put it in the style

of what I’ve designed for my expressive avatar.

I think that’s gonna be pretty compelling.

And so the fashion, you might be buying like a generator,

like a closet that generates a style.

And then like with the GANs,

you’ll be able to infinitely generate outfits

thereby making it, so the reason I wear the same thing

all the time is I don’t like choice.

You’ve talked about the same thing,

but now you don’t even have to choose.

Your closet generates your outfit for you every time.

So you have to live with the outfit it generates.

I mean, you could do that, although,

no, I think that that’s, I think some people will,

but I think like, I think there’s going to be a huge aspect

of just people doing creative commerce here.

So I think that there is going to be a big market

around people designing digital clothing.

But the question is, if you’re designing digital clothing,

do you need to design, if you’re the designer,

do you need to make it for each kind of specific discrete

point along a spectrum, or are you just designing it

for kind of a photo realistic case or an expressive case,

or can you design one

and have it translate across these things?

If I buy a style from a designer who I care about,

and now I’m a dragon, is there a way to morph that

so it goes on the dragon in a way that makes sense?

And that I think is an interesting AI problem

because you’re probably not going to make it

so that designers have to go design for all those things.

But the more useful the digital content is that you buy

in a lot of uses, in a lot of use cases,

the more that economy will just explode.

And that’s a lot of what all of the,

we were joking about NFTs before,

but I think a lot of the promise here is that

if the digital goods that you buy are not just tied

to one platform or one use case,

they end up being more valuable,

which means that people are more willing

and more likely to invest in them,

and that just spurs the whole economy.

But the question is, that’s a fascinating positive aspect,

but the potential negative aspect is that

you can have people concealing their identity

in order to troll or even not people, bots.

So how do you know in the metaverse

that you’re talking to a real human or an AI

or a well intentioned human?

Is that something you think about,

something you’re concerned about?

Well, let’s break that down into a few different cases.

I mean, because knowing that you’re talking to someone

who has good intentions is something that I think

is not even solved in pretty much anywhere.

But I mean, if you’re talking to someone who’s a dragon,

I think it’s pretty clear that they’re not representing

themselves as a person.

I think probably the most pernicious thing

that you want to solve for is,

I think probably one of the scariest ones is

how do you make sure that someone isn’t impersonating you?

So, okay, you’re in a future version of this conversation,

and we have photorealistic avatars,

and we’re doing this in work rooms

or whatever the future version of that is,

and someone walks in who looks like me.

How do you know that that’s me?

And one of the things that we’re thinking about

is it’s still a pretty big AI project

to be able to generate photorealistic avatars

that basically can like,

they work like these codecs of you, right?

So you kind of have a map from your headset

and whatever sensors of what your body’s actually doing,

and it takes the model and it kind of displays it in VR.

But there’s a question, which is,

should there be some sort of biometric security

so that when I put on my VR headset

or I’m going to go use that avatar,

I need to first prove that I am that?

And I think you probably are gonna want something like that.

So as we’re developing these technologies,

we’re also thinking about the security for things like that

because people aren’t gonna wanna be impersonated.

That’s a huge security issue.

Then you just get the question

of people hiding behind fake accounts

to do malicious things,

which is not gonna be unique to the metaverse,

although certainly in a environment

where it’s more immersive

and you have more of a sense of presence,

it could be more painful.

But this is obviously something

that we’ve just dealt with for years

in social media and the internet more broadly.

And there, I think there have been a bunch of tactics

that I think we’ve just evolved to,

we’ve built up these different AI systems

to basically get a sense of,

is this account behaving in the way that a person would?

And it turns out,

so in all of the work that we’ve done around,

we call it community integrity

and it’s basically like policing harmful content

and trying to figure out where to draw the line.

And there are all these like really hard

and philosophical questions around like,

where do you draw the line on some of this stuff?

And the thing that I’ve kind of found the most effective

is as much as possible trying to figure out

who are the inauthentic accounts

or where are the accounts that are behaving

in an overall harmful way at the account level,

rather than trying to get into like policing

what they’re saying, right?

Which I think the metaverse is gonna be even harder

because the metaverse I think will have more properties of,

it’s almost more like a phone call, right?

Or it’s not like I post a piece of content

and is that piece of content good or bad?

So I think more of this stuff will have to be done

at the level of the account.

But this is the area where,

between the kind of counter intelligence teams

that we built up inside the company

and like years of building just different AI systems

to basically detect what is a real account and what isn’t.

I’m not saying we’re perfect,

but like this is an area where I just think

we are like years ahead of basically anyone else

in the industry in terms of having built those capabilities.

And I think that that just is gonna be incredibly important

for this next wave of things.

And like you said, on a technical level,

on a philosophical level,

it’s an incredibly difficult problem to solve.

By the way, I would probably like to open source my avatar

so there could be like millions of Lexis walking around

just like an army.

Like Agent Smith?

Agent Smith, yeah, exactly.

So the Unity ML folks built a copy of me

and they sent it to me.

So there’s a person running around

and I’ve just been doing reinforcement learning on it.

I was gonna release it

because just to have sort of like thousands of Lexis

doing reinforcement.

So they fall over naturally,

they have to learn how to like walk around and stuff.

So I love that idea,

this tension between biometric security,

you want to have one identity,

but then certain avatars, you might have to have many.

I don’t know which is better security,

sort of flooding the world with Lexis

and thereby achieving security

or really being protective of your identity.

I have to ask you a security question actually.

Well, how does flooding the world with Lexis help me know

in our conversation that I’m talking to the real Lex?

I completely destroy the trust

in all my relationships then, right?

If I flood,

cause then it’s, yeah, that.

I think that one’s not gonna work that well for you.

It’s not gonna work that well for the original copy.

It probably fits some things.

Like if you’re a public figure

and you’re trying to have a bunch of,

if you’re trying to show up

in a bunch of different places in the future,

you’ll be able to do that in the metaverse.

So that kind of replication I think will be useful.

But I do think that you’re gonna want a notion of like,

I am talking to the real one.

Yeah.

Yeah, especially if the fake ones start outperforming you

and all your private relationships

and then you’re left behind.

I mean, that’s a serious concern I have with clones.

Again, the things I think about.

Okay, so I recently got, I use QNAP NAS storage.

So just storage for video and stuff.

And I recently got hacked.

This is the first time for me with ransomware.

It’s not me personally, it’s all QNAP devices.

So the question that people have

is about security in general.

Because I was doing a lot of the right things

in terms of security and nevertheless,

ransomware basically disabled my device.

Is that something you think about?

What are the different steps you could take

to protect people’s data on the security front?

I think that there’s different solutions for,

and strategies where it makes sense to have stuff

kind of put behind a fortress, right?

So the centralized model versus the decentralizing.

Then I think both have strengths and weaknesses.

So I think anyone who says, okay,

just decentralize everything, that’ll make it more secure.

I think that that’s tough because,

I mean, the advantage of something like encryption

is that we run the largest encrypted service

in the world with WhatsApp.

And we’re one of the first to roll out

a multi platform encryption service.

And that’s something that I think was a big advance

for the industry.

And one of the promises that we can basically make

because of that, our company doesn’t see

when you’re sending an encrypted message

and to an encrypted message,

what the content is of what you’re sharing.

So that way, if someone hacks Meta servers,

they’re not gonna be able to access the WhatsApp message

that you’re sending to your friend.

And that I think matters a lot to people

because obviously if someone is able to compromise

a company’s servers and that company has hundreds

of millions or billions of people,

then that ends up being a very big deal.

The flip side of that is, okay,

all the content is on your phone.

Are you following security best practices on your phone?

If you lose your phone, all your content is gone.

So that’s an issue.

Maybe you go back up your content from WhatsApp

or some other service in an iCloud or something,

but then you’re just at Apple’s whims about,

are they gonna go turn over the data to some government

or are they gonna get hacked?

So a lot of the time it is useful to have data

in a centralized place too because then you can train

systems that can just do much better personalization.

I think that in a lot of cases, centralized systems

can offer, especially if you’re a serious company,

you’re running the state of the art stuff

and you have red teams attacking your own stuff

and you’re putting out bounty programs

and trying to attract some of the best hackers in the world

to go break into your stuff all the time.

So any system is gonna have security issues,

but I think the best way forward is to basically try

to be as aggressive and open about hardening

the systems as possible, not trying to kind of hide

and pretend that there aren’t gonna be issues,

which I think is over time why a lot of open source systems

have gotten relatively more secure is because they’re open

and it’s not, rather than pretending that there aren’t

gonna be issues, just people surface them quicker.

So I think you want to adopt that approach as a company

and just constantly be hardening yourself.

Trying to stay one step ahead of the attackers.

It’s an inherently adversarial space.

I think it’s an interesting security is interesting

because of the different kind of threats

that we’ve managed over the last five years,

there are ones where basically the adversaries

keep on getting better and better.

So trying to kind of interfere with security

is certainly one area of this.

If you have nation states that are trying

to interfere in elections or something,

they’re kind of evolving their tactics.

Whereas on the other hand, I don’t want to be too simplistic

about it, but if someone is saying something hateful,

people usually aren’t getting smarter and smarter

about how they say hateful things.

So maybe there’s some element of that,

but it’s a very small dynamic compared

to how advanced attackers and some of these other places

get over time.

I believe most people are good,

so they actually get better over time

and not being less hateful

because they realize it’s not fun being hateful.

That’s at least the belief I have.

But first, bathroom break.

Sure, okay.

So we’ll come back to AI,

but let me ask some difficult questions now.

Social Dilemma is a popular documentary

that raised concerns about the effects

of social media on society.

You responded with a point by point rebuttal titled,

What the Social Dilemma Gets Wrong.

People should read that.

I would say the key point they make

is because social media is funded by ads,

algorithms want to maximize attention and engagement

and an effective way to do so is to get people angry

at each other, increase division and so on.

Can you steel man their criticisms and arguments

that they make in the documentary

as a way to understand the concern

and as a way to respond to it?

Well, yeah, I think that’s a good conversation to have.

I don’t happen to agree with the conclusions

and I think that they make a few assumptions

that are just very big jumps

that I don’t think are reasonable to make.

But I understand overall why people would be concerned

that our business model and ads in general,

we do make more money

as people use the service more in general, right?

So as a kind of basic assumption, okay,

do we have an incentive for people to build a service

that people use more?

Yes, on a lot of levels.

I mean, we think what we’re doing is good.

So we think that if people are finding it useful,

they’ll use it more.

Or if you just look at it as this sort of,

if the only thing we cared about is money,

which is not for anyone who knows me,

but okay, we’re a company.

So let’s say you just kind of simplified it down to that,

then would we want people to use the services more?

Yes, and then you get to the second question,

which is does kind of getting people agitated

make them more likely to use the services more?

And I think from looking at other media in the world,

especially TV, and there’s the old news adage,

if it bleeds, it leads.

Like I think that this is,

there are a bunch of reasons why someone might think

that that kind of provocative content

would be the most engaging.

Now, what I’ve always found is two things.

One is that what grabs someone’s attention in the near term

is not necessarily something

that they’re going to appreciate having seen

or going to be the best over the long term.

So I think what a lot of people get wrong

is that I’m not building this company

to make the most money or get people to spend the most time

on this in the next quarter or the next year.

I’ve been doing this for 17 years at this point,

and I’m still relatively young,

and I have a lot more that I wanna do

over the coming decades.

So I think that it’s too simplistic to say,

hey, this might increase time in the near term,

therefore, it’s what you’re gonna do.

Because I actually think a deeper look

at kind of what my incentives are,

the incentives of a company

that are focused on the long term,

is to basically do what people

are gonna find valuable over time,

not what is gonna draw people’s attention today.

The other thing that I’d say is that,

I think a lot of times people look at this

from the perspective of media

or kind of information or civic discourse,

but one other way of looking at this is just that,

okay, I’m a product designer, right?

Our company, we build products,

and a big part of building a product

is not just the function and utility

of what you’re delivering,

but the feeling of how it feels, right?

And we spend a lot of time talking about virtual reality

and how the kind of key aspect of that experience

is the feeling of presence, which it’s a visceral thing.

It’s not just about the utility that you’re delivering,

it’s about like the sensation.

And similarly, I care a lot about how people feel

when they use our products,

and I don’t want to build products that make people angry.

I mean, that’s like not, I think,

what we’re here on this earth to do,

is to build something that people spend a bunch of time doing

and it just kind of makes them angrier at other people.

I mean, I think that that’s not good.

That’s not what I think would be

sort of a good use of our time

or a good contribution to the world.

So, okay, it’s like people, they tell us

on a per content basis, does this thing,

do I like it?

Do I love it?

Does it make me angry?

Does it make me sad?

And based on that, we choose to basically show content

that makes people angry less,

because of course, if you’re designing a product

and you want people to be able to connect

and feel good over a long period of time,

then that’s naturally what you’re gonna do.

So, I don’t know, I think overall,

I understand at a high level,

if you’re not thinking too deeply about it,

why that argument might be appealing.

But I just think if you actually look

at what our real incentives are,

not just like if we were trying to optimize

for the next week,

but like as people working on this,

like why are we here?

And I think it’s pretty clear

that that’s not actually how you would wanna

design the system.

I guess one other thing that I’d say is that,

while we’re focused on the ads business model,

I do think it’s important to note that a lot

of these issues are not unique to ads.

I mean, so take like a subscription news business model,

for example, I think that has just as many

potential pitfalls.

Maybe if someone’s paying for a subscription,

you don’t get paid per piece of content that they look at,

but say for example, I think like a bunch

of the partisanship that we see could potentially

be made worse by you have these kind of partisan

news organizations that basically sell subscriptions

and they’re only gonna get people on one side

to basically subscribe to them.

So their incentive is not to print content

or produce content that’s kind of centrist

or down the line either.

I bet that what a lot of them find is that

if they produce stuff that’s kind of more polarizing

or more partisan, then that is what gets

the more subscribers.

So I think that this stuff is all,

there’s no perfect business model.

Everything has pitfalls.

The thing that I think is great about advertising

is it makes it so the consumer service is free,

which if you believe that everyone should have a voice

and everyone should be able to connect,

then that’s a great thing, as opposed to building

a luxury service that not everyone can afford.

But look, every business model, you have to be careful

about how you’re implementing what you’re doing.

You responded to a few things there.

You spoke to the fact that there is a narrative

of malevolence, like you’re leaning into them,

making people angry just because it makes more money

in the short term, that kind of thing.

So you responded to that.

But there’s also kind of reality of human nature.

Just like you spoke about, there’s fights,

arguments we get in and we don’t like ourselves afterwards,

but we got into them anyway.

So our longterm growth is, I believe for most of us,

has to do with learning, challenging yourself,

improving, being kind to each other,

finding a community of people that you connect with

on a real human level, all that kind of stuff.

But it does seem when you look at social media

that a lot of fights break out,

a lot of arguments break out,

a lot of viral content ends up being sort of outrage

in one direction or the other.

And so it’s easy from that to infer the narrative

that social media companies are letting

this outrage become viral.

And so they’re increasing the division in the world.

I mean, perhaps you can comment on that

or further, how can you be,

how can you push back on this narrative?

How can you be transparent about this battle?

Because I think it’s not just motivation or financials,

it’s a technical problem too,

which is how do you improve longterm wellbeing

of human beings?

I think that going through some of the design decisions

would be a good conversation.

But first, I actually think,

I think you acknowledged that,

that narrative is somewhat anecdotal.

And I think it’s worth grounding this conversation

in the actual research that has been done on this,

which by and large finds that social media

is not a large driver of polarization, right?

And, I mean, there’s been a number of economists

and social scientists and folks who have studied this.

In a lot of polarization, it varies around the world.

If social media is basically in every country,

Facebook’s in pretty much every country

except for China and maybe North Korea.

And you see different trends in different places

where in a lot of countries polarization is declining,

in some it’s flat, in the US it’s risen sharply.

So the question is, what are the unique phenomenon

in the different places?

And I think for the people who are trying to say,

hey, social media is the thing that’s doing this.

I think that that clearly doesn’t hold up

because social media is a phenomenon

that is pretty much equivalent

in all of these different countries.

And you have researchers like this economist at Stanford,

Matthew Genskow, who has just written at length about this.

And it’s a bunch of books by political scientists,

Ezra Klein and folks, why we’re polarized,

basically goes through this decades long analysis in the US.

Before I was born, basically talking about

some of the forces in kind of partisan politics

and Fox News and different things

that predate the internet in a lot of ways

that I think are likely larger contributors.

So to the contrary on this,

not only is it pretty clear that social media

is not a major contributor,

but most of the academic studies that I’ve seen

actually show that social media use

is correlated with lower polarization.

And Genskow, the same person who just did the study

that I cited about longitudinal polarization

across different countries,

also did a study that basically showed

that if you looked after the 2016 election in the US,

the voters who were the most polarized

were actually the ones who were not on the internet.

So, and there have been recent other studies,

I think in Europe and around the world,

basically showing that as people stop using social media,

they tend to get more polarized.

Then there’s a deeper analysis around,

okay, well, polarization actually isn’t even one thing.

Cause you know, having different opinions on something

isn’t, I don’t think that that’s by itself bad.

What people who study this say is most problematic

is what they call affective polarization,

which is basically are you,

do you have negative feelings towards people

of another group?

And the way that a lot of scholars study this

is they basically ask a group,

would you let your kids marry someone of group X?

Whatever the groups are that you’re worried

that someone might have negative feelings towards.

And in general, use of social media

has corresponded to decreases

in that kind of affective polarization.

So I just wanna, I think we should talk

through the design decisions and how we handle

the kind of specific pieces of content,

but overall, I think it’s just worth grounding

that discussion in the research that’s existed

that I think overwhelmingly shows

that the mainstream narrative around this

is just not right.

But the narrative does take hold

and it’s compelling to a lot of people.

There’s another question I’d like to ask you on this.

I was looking at various polls and saw that you’re

one of the most disliked tech leaders today,

54% unfavorable rating.

Elon Musk is 23%.

It’s basically everybody has a very high unfavorable rating

that are tech leaders.

Maybe you can help me understand that.

Why do you think so many people dislike you?

Some even hate you.

And how do you regain their trust and support?

Given everything you just said,

why are you losing the battle

in explaining to people what actual impact

social media has on society?

Well, I’m curious if that’s a US survey or world.

It is US, yeah.

So I think that there’s a few dynamics.

One is that our brand

has been somewhat uniquely challenged in the US

compared to other places.

It’s not that there are.

I mean, other countries, we have issues too,

but I think in the US, there was this dynamic where

if you look at like the next sentiment

of kind of coverage or attitude towards us,

before 2016, I think that there were probably

very few months, if any, where it was negative.

And since 2016, I think that there probably

been very few months, if any, then it’s been positive.

Politics.

But I think it’s a specific thing.

And this is very different from other places.

So I think in a lot of other countries in the world,

the sentiment towards meta and our services

is extremely positive.

In the US, we have more challenges.

And I think compared to other companies,

you can look at certain industries,

I think if you look at it from like a partisan perspective,

not from like a political perspective,

but just kind of culturally,

it’s like there are people who are probably

more left of center and there are people

who are more right of center,

and there’s kind of blue America and red America.

There are certain industries that I think

maybe one half of the country has a more positive view

towards than another.

And I think we’re in a,

one of the positions that we’re in that I think

is really challenging is that because of a lot

of the content decisions that we’ve basically

had to arbitrate, and because we’re not a partisan company,

we’re not a Democrat company or a Republican company,

we’re trying to make the best decisions we can

to help people connect and help people have as much voice

as they can while having some rules

because we’re running a community.

The net effect of that is that we’re kind of constantly

making decisions that piss off people in both camps.

And the effect that I’ve sort of seen is that

when we make a decision that is,

that’s a controversial one that’s gonna upset,

say about half the country,

those decisions are all negative sum,

from a brand perspective, because it’s not like,

if we make that decision in one way

and say half the country is happy

about that particular decision that we make,

they tend to not say, oh, sweet, meta got that one right.

They’re just like, ah, you didn’t mess that one up.

But their opinion doesn’t tend to go up by that much.

Whereas the people who kind of are on the other side of it

are like, God, how could you mess that up?

How could you possibly think that that piece of content

is okay and should be up and should not be censored?

Or, and so I think the, whereas if you leave it up

and, you know, it’s, or if you take it down,

the people who thought it should be taken down or,

you know, it’s like, all right, fine, great.

You didn’t mess that one up.

So our internal assessment of,

and the kind of analytics on our brand

are basically anytime one of these big controversial things

comes up in society,

our brand goes down with half of the country.

And then like, if you,

and then if you just kind of extrapolate that out,

it’s just been very challenging for us to try to navigate

what is a polarizing country in a principled way,

where we’re not trying to kind of hew to one side

or the other, we’re trying to do

what we think is the right thing.

But that’s what I think is the right thing

for us to do though.

So, I mean, that’s what we’ll try to keep doing.

Just as a human being, how does it feel though,

when you’re giving so much of your day to day life

to try to heal division, to try to do good in the world,

as we’ve talked about, that so many people in the US,

the place you call home have a negative view

of you as a leader, as a human being

and the company you love?

Well, I mean, it’s not great,

but I mean, look, if I wanted people to think positively

about me as a person,

I don’t know, I’m not sure if you go build a company.

I mean, it’s like.

Or a social media company.

It seems exceptionally difficult to do

with a social media company.

Yeah, so, I mean, I don’t know,

there is a dynamic where a lot of the other people

running these companies, internet companies,

have sort of stepped back and they just do things

that are sort of, I don’t know, less controversial.

And some of it may be that they just get tired over time.

But, you know, it’s, so I don’t know.

I think that, you know, running a company is hard,

building something at scale is hard.

You only really do it for a long period of time

if you really care about what you’re doing.

And yeah, so, I mean, it’s not great, but like,

but look, I think that at some level,

whether 25% of people dislike you

or 75% of people dislike you,

your experience as a public figure is gonna be

that there’s a lot of people who dislike you, right?

So, I actually am not sure how different it is.

You know, certainly, you know,

the country’s gotten more polarized

and we in particular have gotten, you know,

more controversial over the last five or years or so.

But, I don’t know, I kind of think like as a public figure

and leader of one of these enterprises.

Comes with the job.

Yeah, part of what you do is like,

and look, the answer can’t just be ignore it, right?

Because like a huge part of the job

is like you need to be getting feedback

and internalizing feedback on how you can do better.

But I think increasingly what you need to do

is be able to figure out, you know,

who are the kind of good faith critics

who are criticizing you because

they’re trying to help you do a better job

rather than tear you down.

And those are the people I just think you have to cherish

and like, and listen very closely

to the things that they’re saying,

because, you know, I think it’s just as dangerous

to tune out everyone who says anything negative

and just listen to the people who are kind of positive

and support you, you know,

as it would be psychologically to pay attention

trying to make people who are never gonna like you like you.

So I think that that’s just kind of a dance

that people have to do.

But I mean, I, you know,

so you kind of develop more of a feel for like,

who actually is trying to accomplish

the same types of things in the world

and who has different ideas about how to do that

and how can I learn from those people?

And like, yeah, we get stuff wrong.

And when the people whose opinions I respect

call me out on getting stuff wrong,

that hurts and makes me wanna do better.

But I think at this point, I’m pretty tuned to just,

all right, if someone, if I know they’re,

they’re kind of like operating in bad faith

and they’re not really trying to help,

then, you know, I don’t know, it’s not, it’s, it doesn’t,

you know, I think over time,

it just doesn’t bother you that much.

But you are surrounded by people that believe in the mission

that love you.

Are there friends or colleagues in your inner circle

you trust that call you out on your bullshit

whenever your thinking may be misguided

as it is for leaders at times?

I think we have a famously open company culture

where we sort of encourage that kind of dissent internally,

which is, you know, why there’s so much material

internally that can leak out

with people sort of disagreeing

is because that’s sort of the culture.

You know, our management team, I think it’s a lot of people,

you know, there are some newer folks who come in,

there are some folks who’ve kind of been there for a while,

but there’s a very high level of trust.

And I would say it is a relatively confrontational

group of people.

And my friends and family, I think, will push me on this.

But look, it’s not just,

but I think you need some diversity, right?

It can’t just be, you know,

people who are your friends and family.

It’s also, you know, I mean, there are journalists

or analysts or, you know,

peer executives at other companies

or, you know, other people who sort of are insightful

about thinking about the world,

you know, certain politicians

or people kind of in that sphere

who I just think have like very insightful perspectives

who even if they would,

they come at the world from a different perspective,

which is sort of what makes the perspective so valuable.

But, you know, I think fundamentally

we’re trying to get to the same place

in terms of, you know, helping people connect more,

helping the whole world function better,

not just, you know, one place or another.

And I don’t know, I mean,

those are the people whose opinions really matter to me.

And I just, it’s, you know,

that’s how I learn on a day to day basis.

People are constantly sending me comments on stuff

or links to things they found interesting.

And I don’t know, it’s kind of constantly evolving

this model of the world

and kind of what we should be aspiring to be.

You’ve talked about, you have a famously open culture

which comes with the criticism

and the painful experiences.

So let me ask you another difficult question.

Frances Haugen, the Facebook whistleblower,

leaked the internal Instagram research

into teenagers and wellbeing.

Her claim is that Instagram is choosing profit

over wellbeing of teenage girls.

So Instagram is quote, toxic for them.

Your response titled,

what our research really says about teen wellbeing

and Instagram says, no, Instagram research shows

that 11 of 12 wellbeing issues,

teenage girls who said they struggle

with those difficult issues also said

that Instagram made them better rather than worse.

Again, can you steal man and defend the point

and Frances Haugen’s characterization of the study

and then help me understand the positive

and negative effects of Instagram

and Facebook on young people?

So there are certainly questions around teen mental health

that are really important.

It’s hard to, as a parent, it’s like hard to imagine

any set of questions that are sort of more important.

I mean, I guess maybe other aspects of physical health

or wellbeing are probably come to that level,

but like, these are really important questions, right?

Which is why we dedicate teams to studying them.

I don’t think the internet or social media are unique

in having these questions.

I mean, I think people and there’ve been sort of magazines

with promoting certain body types for women

and kids for decades,

but we really care about this stuff.

So we wanted to study it.

And of course, we didn’t expect

that everything was gonna be positive all the time.

So, I mean, the reason why you study this stuff

is to try to improve and get better.

So, I mean, look, the place where I disagree

with the characterization first,

I thought some of the reporting and coverage of it

just took the whole thing out of proportion

and that it focused on, as you said,

I think there were like 20 metrics in there

and on 18 or 19, the effect of using Instagram

was neutral or positive on the teen’s wellbeing.

And there was one area where I think it showed

that we needed to improve

and we took some steps to try to do that

after doing the research.

But I think having the coverage just focus on that one

without focusing on the,

I mean, I think an accurate characterization

would have been that kids using Instagram

or not kids, teens is generally positive

for their mental health.

But of course, that was not the narrative that came out.

So I think it’s hard to,

that’s not a kind of logical thing to straw man,

but I sort of disagree or steel man,

but I sort of disagree with that overall characterization.

I think anyone sort of looking at this objectively would,

but then, I mean, there is this sort of intent critique

that I think you were getting at before,

which says, it assumes some sort of malevolence, right?

It’s like, which it’s really hard for me

to really wrap my head around this

because as far as I know,

it’s not clear that any of the other tech companies

are doing this kind of research.

So why the narrative should form that we did research

because we were studying an issue

because we wanted to understand it to improve

and took steps after that to try to improve it,

that your interpretation of that would be

that we did the research

and tried to sweep it under the rug.

It just, it sort of is like, I don’t know,

it’s beyond credibility to me

that like that’s the accurate description of the actions

that we’ve taken compared to the others in the industry.

So I don’t know, that’s kind of, that’s my view on it.

These are really important issues

and there’s a lot of stuff

that I think we’re gonna be working on

related to teen mental health for a long time,

including trying to understand this better.

And I would encourage everyone else

in the industry to do this too.

Yeah, I would love there to be open conversations

and a lot of great research being released internally

and then also externally.

It doesn’t make me feel good

to see press obviously get way more clicks

when they say negative things about social media.

Objectively speaking, I can just tell

that there’s hunger to say negative things

about social media.

And I don’t understand how that’s supposed to lead

to an open conversation about the positives

and the negatives, the concerns about social media,

especially when you’re doing that kind of research.

I mean, I don’t know what to do with that,

but let me ask you as a father,

there’s a weight heavy on you

that people get bullied on social networks.

So people get bullied in their private life.

But now because so much of our life is in the digital world,

the bullying moves from the physical world

to the digital world.

So you’re now creating a platform

on which bullying happens.

And some of that bullying can lead to damage

to mental health.

And some of that bullying can lead to depression,

even suicide.

There’s a weight heavy on you

that people have committed suicide

or will commit suicide based on the bullying

that happens on social media.

Yeah, I mean, there’s a set of harms

that we basically track and build systems to fight against.

And bullying and self harm are,

these are some of the biggest things

that we are most focused on.

For bullying, like you say, it’s gonna be,

while this predates the internet,

then it’s probably impossible to get rid of all of it.

You wanna give people tools to fight it

and you wanna fight it yourself.

And you also wanna make sure that people have the tools

to get help when they need it.

So I think this isn’t like a question of,

can you get rid of all bullying?

I mean, it’s like, all right, I mean, I have two daughters

and they fight and push each other around and stuff too.

And the question is just,

how do you handle that situation?

And there’s a handful of things that I think you can do.

We talked a little bit before around some of the AI tools

that you can build to identify

when something harmful is happening.

It’s actually, it’s very hard in bullying

because a lot of bullying is very context specific.

It’s not like you’re trying to fit a formula of like,

if like looking at the different harms,

someone promoting a terrorist group is like,

probably one of the simpler things to generally find

because things promoting that group are gonna look

at a certain way or feel a certain way.

Bullying could just be, you know,

someone making some subtle comment about someone’s appearance

that’s idiosyncratic to them.

And it could look at just like humor.

So humor to one person can be destructive

to another human being, yeah.

So with bullying, I think there are certain things

that you can find through AI systems,

but I think it is increasingly important

to just give people more agency themselves.

So we’ve done things like making it

so people can turn off comments

or take a break from hearing from a specific person

without having to signal at all

that they’re gonna stop following them

or kind of make some stand that,

okay, I’m not friends with you anymore.

I’m not following you.

I just like, I just don’t wanna hear about this,

but I also don’t wanna signal at all publicly

that or to them that there’s been an issue.

And then you get to some of the more extreme cases

like you’re talking about

where someone is thinking about self harm or suicide.

And there we’ve found that that is a place

where AI can identify a lot

as well as people flagging things.

If people are expressing something

that is potentially they’re thinking of hurting themselves,

those are cues that you can build systems

and hundreds of languages around the world

to be able to identify that.

And one of the things that I’m actually quite proud of

is we’ve built these systems

that I think are clearly leading at this point

that not only identify that,

but then connect with local first responders

and have been able to save, I think at this point,

it’s in thousands of cases,

be able to get first responders to people

through these systems who really need them

because of specific plumbing that we’ve done

between the AI work and being able to communicate

with local first responder organizations.

We’re rolling that out in more places around the world.

And I think the team that worked on that

just did awesome stuff.

So I think that that’s a long way of saying,

yeah, I mean, this is a heavy topic

and you want to attack it in a bunch of different ways

and also kind of understand that some of nature

is for people to do this to each other,

which is unfortunate,

but you can give people tools and build things that help.

It’s still one hell of a burden though.

A platform that allows people

to fall in love with each other

is also by nature going to be a platform

that allows people to hurt each other.

And when you’re managing such a platform, it’s difficult.

And I think you spoke to it,

but the psychology of that, of being a leader in that space,

of creating technology that’s playing in this space,

like you mentioned, psychology is really damn difficult.

And I mean, the burden of that is just great.

I just wanted to hear you speak to that point.

I have to ask about the thing you’ve brought up a few times,

which is making controversial decisions.

Let’s talk about free speech and censorship.

So there are two groups of people pressuring Meta on this.

One group is upset that Facebook, the social network,

allows misinformation in quotes to be spread on the platform.

The other group are concerned that Facebook censors speech

by calling it misinformation.

So you’re getting it from both sides.

You, in 2019, October at Georgetown University,

eloquently defended the importance of free speech,

but then COVID came and the 2020 election came.

Do you worry that outside pressures

from advertisers, politicians, the public,

have forced Meta to damage the ideal of free speech

that you spoke highly of?

Just to say some obvious things upfront,

I don’t think pressure from advertisers

or politicians directly in any way

affects how we think about this.

I think these are just hard topics.

So let me just take you through our evolution

from kind of the beginning of the company

to where we are now.

You don’t build a company like this

unless you believe that people expressing themselves

is a good thing, right?

So that’s sort of the foundational thing.

You can kind of think about our company as a formula

where we think giving people voice

and helping people connect creates opportunity, right?

So those are the two things that we’re always focused on

are sort of helping people connect.

We talked about that a lot,

but also giving people voice

and ability to express themselves.

Then by the way, most of the time

when people express themselves,

that’s not like politically controversial content.

It’s like expressing something about their identity

that’s more related to the avatar conversation

we had earlier in terms of expressing some facet,

but that’s what’s important to people on a day to day basis.

And sometimes when people feel strongly enough

about something, it kind of becomes a political topic.

That’s sort of always been a thing that we’ve focused on.

There’s always been the question of safety in this,

which if you’re building a community,

I think you have to focus on safety.

We’ve had these community standards from early on,

and there are about 20 different kinds of harm

that we track and try to fight actively.

We’ve talked about some of them already.

So it includes things like bullying and harassment.

It includes things like terrorism or promoting terrorism,

inciting violence, intellectual property theft.

And in general, I think call it about 18 out of 20 of those.

There’s not really a particularly polarized definition

of that.

I think you’re not really gonna find many people

in the country or in the world

who are trying to say we should be

fighting terrorist content less.

I think the content where there are a couple of areas

where I think that this has gotten more controversial

recently, which I’ll talk about.

And you’re right, the misinformation is basically is up there.

And I think sometimes the definition of hate speech

is up there too.

But I think in general, most of the content

that I think we’re working on for safety

is not actually, people don’t kind of have these questions.

So it’s sort of this subset.

But if you go back to the beginning of the company,

this was sort of pre deep learning days.

And therefore, it was me and my roommate Dustin join me.

And if someone posted something bad,

it was the AI technology did not exist yet

to be able to go basically look at all the content.

And we were a small enough outfit

that no one would expect that we could review it all.

Even if someone reported it to us,

we basically did our best, right?

It’s like someone would report it

and we try to look at stuff and deal with stuff.

And for call it the first seven or eight years

of the company, we weren’t that big of a company.

For a lot of that period, we weren’t even really profitable.

The AI didn’t really exist to be able to do

the kind of moderation that we do today.

And then at some point in kind of the middle

of the last decade, that started to flip.

And we got to the point where we were sort of a larger

and more profitable company.

And the AI was starting to come online

to be able to proactively detect

some of the simpler forms of this.

So things like pornography,

you could train an image classifier

to identify what a nipple was,

or you can fight against terrorist content.

You still could.

There’s actually papers on this, it’s great.

Oh, of course there are.

Technical papers.

Of course there are.

Those are relatively easier things to train AI to do

than for example, understand the nuances

of what is inciting violence

in a hundred languages around the world

and not have the false positives of like,

okay, are you posting about this thing

that might be inciting violence

because you’re actually trying to denounce it?

In which case we probably shouldn’t take that down.

Where if you’re trying to denounce something

that’s inciting violence in some kind of dialect

in a corner of India, as opposed to,

okay, actually you’re posting this thing

because you’re trying to incite violence.

Okay, building an AI that can basically get

to that level of nuance and all the languages

that we serve is something that I think

is only really becoming possible now,

not towards the middle of the last decade.

But there’s been this evolution,

and I think what happened,

people sort of woke up after 2016

and a lot of people are like,

okay, the country is a lot more polarized

and there’s a lot more stuff here than we realized.

Why weren’t these internet companies on top of this?

And I think at that point it was reasonable feedback

that some of this technology had started becoming possible.

And at that point, I really did feel like

we needed to make a substantially larger investment.

We’d already worked on this stuff a lot,

on AI and on these integrity problems,

but that we should basically invest,

have a thousand or more engineers

basically work on building these AI systems

to be able to go and proactively identify the stuff

across all these different areas.

Okay, so we went and did that.

Now we’ve built the tools to be able to do that.

And now I think it’s actually a much more complicated

set of philosophical rather than technical questions,

which is the exact policies, which are okay.

Now, the way that we basically hold ourselves accountable

is we issue these transparency reports every quarter

and the metric that we track is for each of these

20 types of harmful content.

How much of that content are we taking down

before someone even has to report it to us?

So how effective is our AI at doing this?

But that basically creates this big question,

which is okay, now we need to really be careful

about how proactive we set the AI

and where the exact policy lines are

around what we’re taking down.

It’s certainly at a point now where I felt like

at the beginning of that journey

of building those AI systems, there was a lot of push.

There’s saying, okay, you’ve got to do more.

There’s clearly a lot more bad content

that people aren’t reporting or that you’re not getting to

and you need to get more effective at that.

And I was pretty sympathetic to that.

But then I think at some point along the way,

there started to be almost equal issues on both sides

of, okay, actually you’re kind of taking down

too much stuff, right?

Or some of the stuff is borderline

and it wasn’t really bothering anyone

and they didn’t report it.

So is that really an issue that you need to take down?

Whereas we still have the critique on the other side too

where a lot of people think we’re not doing enough.

So it’s become, as we built the technical capacity,

I think it becomes more philosophically interesting almost

where you wanna be on the line.

And I just think you don’t want one person

making those decisions.

So we’ve also tried to innovate

in terms of building out this independent oversight board,

which has people who are dedicated to free expression

but from around the world who people can appeal cases to.

So a lot of the most controversial cases basically go to them

and they make the final binding decision

on how we should handle that.

And then of course, their decisions,

we then try to figure out what the principles are

behind those and encode them into the algorithms.

And how are those people chosen, which, you know,

you’re outsourcing a difficult decision.

Yeah, the initial people,

we chose a handful of chairs for the group

and we basically chose the people

for a commitment to free expression

and like a broad understanding of human rights

and the trade offs around free expression.

So they fundamentally people

who are gonna lean towards free expression.

Towards freedom of speech.

Okay, so there’s also this idea of fact checkers.

So jumping around to the misinformation questions,

especially during COVID,

which is an exceptionally speaking of polarization.

Can I speak to the COVID thing?

I mean, I think one of the hardest set of questions

around free expression,

because you asked about Georgetown

has my stance fundamentally changed?

And the answer to that is no, my stance has not changed.

It is fundamentally the same as when I was talking

at Georgetown from a philosophical perspective.

The challenge with free speech is that everyone agrees

that there is a line where if you’re actually

about to do physical harm to people

that there should be restrictions.

So, I mean, there’s the famous Supreme Court

historical example of like,

you can’t yell fire in a crowded theater.

The thing that everyone disagrees on

is what is the definition of real harm?

Where I think some people think,

okay, this should only be a very literal,

I mean, take it back to the bullying conversation

we were just having, where is it just harm

if the person is about to hurt themselves

because they’ve been bullied so hard?

Or is it actually harm like as they’re being bullied?

And kind of at what point in the spectrum is that?

And that’s the part that there’s not agreement on.

But I think what people agree on pretty broadly

is that when there is an acute threat

that it does make sense from a societal perspective

to tolerate less speech.

That could be potentially harmful in that acute situation.

So I think where COVID got very difficult is,

I don’t think anyone expected this to be going on for years.

But if you’d kind of asked now a priori,

would a global pandemic where a lot of people are dying

and catching this, is that an emergency

that where you’d kind of consider it

that it’s problematic to basically yell fire

in a crowded theater?

I think that that probably passes that test.

So I think that it’s a very tricky situation,

but I think the fundamental commitment

to free expression is there.

And that’s what I believe.

And again, I don’t think you start this company

unless you care about people being able

to express themselves as much as possible.

But I think that that’s the question,

is how do you define what the harm is

and how acute that is?

And what are the institutions that define that harm?

A lot of the criticism is that the CDC, the WHO,

the institutions we’ve come to trust as a civilization

to give the line of what is and isn’t harm

in terms of health policy have failed in many ways,

in small ways and in big ways, depending on who you ask.

And then the perspective of meta and Facebook is like,

well, where the hell do I get the information

of what is and isn’t misinformation?

So it’s a really difficult place to be in,

but it’s great to hear that you’re leaning

towards freedom of speech on this aspect.

And again, I think this actually calls to the fact

that we need to reform institutions

that help keep an open mind

of what is and isn’t misinformation.

And misinformation has been used to bully on the internet.

I mean, I just have, I’m friends with Joe Rogan

and he is called as a,

I remember hanging out with him in Vegas

and somebody yelled, stop spreading misinformation.

I mean, and there’s a lot of people that follow him

that believe he’s not spreading misinformation.

Like you can’t just not acknowledge the fact

that there’s a large number of people

that have a different definition of misinformation.

And that’s such a tough place to be.

Like who do you listen to?

Do you listen to quote unquote experts who gets,

as a person who has a PhD, I gotta say,

I mean, I’m not sure I know what defines an expert,

especially in a new,

in a totally new pandemic or a new catastrophic event,

especially when politics is involved

and especially when the news are,

the media involved that can propagate

sort of outrageous narratives

and thereby make a lot of money.

Like what the hell?

Where’s the source of truth?

And then everybody turns to Facebook.

It’s like, please tell me what the source of truth is.

Well, I mean, well, how would you handle this

if you were in my position?

Is very, very, very, very difficult.

I would say,

I would more speak about how difficult the choices are

and be transparent about like,

what the hell do you do with this?

Like here, you got exactly,

ask the exact question you just asked me,

but to the broader public, like, okay, yeah,

you guys tell me what to do.

So like crowdsource it.

And then the other aspect is when you spoke really eloquently

about the fact that there’s this going back and forth

and now there’s a feeling like you’re censoring

a little bit too much.

So I would lean, I would try to be ahead of that feeling.

I would now lean towards freedom of speech and say,

we’re not the ones that are going to define misinformation.

Let it be a public debate, let the idea stand.

And I actually place, this idea of misinformation,

I place the responsibility

on the poor communication skills of scientists.

They should be in the battlefield of ideas

and everybody who is spreading information

against the vaccine, they should not be censored.

They should be talked with and you should show the data,

you should have open discussion

as opposed to rolling your eyes and saying,

I’m the expert, I know what I’m talking about.

No, you need to convince people, it’s a battle of ideas.

So that’s the whole point of freedom of speech.

It’s the way to defeat bad ideas

is with good ideas, with speech.

So like the responsibility here falls

on the poor communication skills of scientists.

Thanks to social media, scientists are not communicators.

They have the power to communicate.

Some of the best stuff I’ve seen about COVID

from doctors is on social media.

It’s a way to learn to respond really quickly,

to go faster than the peer review process.

And so they just need to get way better

at that communication.

And also by better, I don’t mean just convincing,

I also mean speak with humility,

don’t talk down to people, all those kinds of things.

And as a platform, I would say,

I would step back a little bit.

Not all the way, of course,

because there’s a lot of stuff that can cause real harm

as we’ve talked about,

but you lean more towards freedom of speech

because then people from a brand perspective

wouldn’t be blaming you for the other ills of society,

which there are many.

The institutions have flaws, the political divide,

obviously politicians have flaws, that’s news.

The media has flaws that they’re all trying to work with.

And because of the central place of Facebook in the world,

all of those flaws somehow kind of propagate to Facebook.

And you’re sitting there as Plato, the philosopher,

have to answer to some of the most difficult questions

asking, being asked of human civilization.

So I don’t know, maybe this is an American answer though,

to lean towards freedom of speech.

I don’t know if that applies globally.

So yeah, I don’t know.

But transparency and saying, I think as a technologist,

one of the things I sense about Facebook and meta

when people talk about this company

is they don’t necessarily understand

fully how difficult the problem is.

You talked about AI has to catch

a bunch of harmful stuff really quickly.

Just the sea of data you have to deal with.

It’s a really difficult problem.

So like any of the critics,

if you just hand them the helm for a week,

let’s see how well you can do.

Like that, to me, that’s definitely something

that would wake people up to how difficult this problem is

if there’s more transparency

of saying how difficult this problem is.

Let me ask you about, on the AI front,

just because you mentioned language and my ineloquence.

Translation is something I wanted to ask you about.

And first, just to give a shout out to the supercomputer.

You’ve recently announced the AI research supercluster, RSC.

Obviously, I’m somebody who loves the GPUs.

It currently has 6,000 GPUs.

NVIDIA DGX A100 is the systems that have

in total 6,000 GPUs.

And it will eventually, maybe this year,

maybe soon, will have 16,000 GPUs.

So it can do a bunch of different kinds

of machine learning applications.

There’s a cool thing on the distributed storage aspect

and all that kind of stuff.

So one of the applications that I think is super exciting

is translation, real time translation.

I mentioned to you that having a conversation,

I speak Russian fluently,

I speak English somewhat fluently,

and having a conversation with Vladimir Putin,

say, as a use case.

Me, as a user, coming to you as a use case.

We both speak each other’s language.

I speak Russian, he speaks English.

How can we have that communication go well

with the help of AI?

I think it’s such a beautiful and a powerful application

of AI to connect the world,

that bridge the gap, not necessarily between me and Putin,

but people that don’t have that shared language.

Can you just speak about your vision with translation?

Because I think that’s a really exciting application.

If you’re trying to help people connect

all around the world,

a lot of content is produced in one language

and people in all these other places are interested in it.

So being able to translate that

just unlocks a lot of value on a day to day basis.

I mean, so the kind of AI around translation is interesting

because it’s gone through a bunch of iterations.

But the basic state of the art

is that you don’t wanna go through

different kind of intermediate symbolic

representations of language or something like that.

You basically wanna be able to map the concepts

and basically go directly from one language to another.

And you just can train bigger and bigger models

in order to be able to do that.

And that’s where the research supercluster comes in

is basically a lot of the trend in machine learning

is just you’re building bigger and bigger models

and you just need a lot of computation to train them.

So it’s not that like the translation would run

on the supercomputer, the training of the model,

which could have billions or trillions of examples

of just basically that.

You’re training models on this supercluster

in days or weeks that might take a much longer period of time

on a smaller cluster.

So it just wouldn’t be practical for most teams to do.

But the translation work,

we’re basically getting from being able to go

between about a hundred languages seamlessly today

to being able to go to about 300 languages in the near term.

So from any language to any other language.

Yeah.

And part of the issue when you get closer to more languages

is some of these get to be pretty,

not very popular languages, right?

Where there isn’t that much content in them.

So you end up having less data

and you need to kind of use a model that you’ve built up

around other examples.

And this is one of the big questions around AI

is like how generalizable can things be?

And that I think is one of the things

that’s just kind of exciting here

from a technical perspective.

But capturing, we talked about this with the metaverse,

capturing the magic of human to human interaction.

So me and Putin, okay.

Again, this is therapy session.

I mean, it’s a tough example

because you actually both speak Russian and English.

No, but that’s.

But in the future.

I see it as a touring test of a kind

because we would both like to have an AI that improves

because I don’t speak Russian that well.

He doesn’t speak English that well.

Yeah.

It would be nice to outperform our abilities

and it sets a really nice bar

because I think AI can really help in translation

for people that don’t speak the language at all,

but to actually capture the magic of the chemistry,

the translation, which would make the metaverse

super immersive.

I mean, that’s exciting.

You remove the barrier of language, period.

Yeah, so when people think about translation,

I think a lot of that is they’re thinking about text to text,

but speech to speech, I think is a whole nother thing.

And I mean, one of the big lessons on that,

which I was referring to before is I think early models,

it’s like, all right, they take speech,

they translate it to text,

translate the text to another language

and then kind of output that as speech in that language.

And you don’t wanna do that.

You just wanna be able to go directly from speech

in one language to speech in another language

and build up the models to do that.

And I mean, I think one of the,

there have been,

when you look at the progress in machine learning,

there have been big advances in the techniques,

some of the advances in self supervised learning,

which I know you talked to Jan about

and he’s like one of the leading thinkers in this area.

I just think that that stuff is really exciting,

but then you couple that with the ability

to just throw larger and larger amounts of compute

at training these models.

And you can just do a lot of things

that were harder to do before.

But we’re asking more of our systems too, right?

So if you think about the applications

that we’re gonna need for the metaverse,

or think about it, okay,

so let’s talk about AR here for a second.

You’re gonna have these glasses,

they’re gonna look hopefully

like a normal ish looking pair of glasses,

but they’re gonna be able to put holograms in the world

and intermix virtual and physical objects in your scene.

And one of the things that’s gonna be unique about this

compared to every other computing device

that you’ve had before,

is that this is gonna be the first computing device

that has all the same signals

about what’s going on around you that you have.

Right, so your phone,

you can have it take a photo or a video,

but I mean, these glasses are gonna,

whenever you activate them,

they’re gonna be able to see what you see

from your perspective,

they’re gonna be able to hear what you hear

because the microphones and all that

are gonna be right around where your ears are.

So you’re gonna want an AI assistant,

that’s a new kind of AI assistant

that can basically help you process the world

from this first person perspective

or from the perspective that you have.

And the utility of that is gonna be huge,

but the kinds of AI models that we’re gonna need

are going to be just,

I don’t know, there’s a lot that we’re gonna need

to basically make advances in.

But I mean, but that’s why I think these concepts

of the metaverse and the advances in AI

are so fundamentally interlinked

that I mean, they’re kind of enabling each other.

Yeah, like the world builder is a really cool idea.

Like you can be like a Bob Ross,

like I’m gonna put a little tree right here.

Yeah.

I need a little tree, it’s missing a little tree.

And then, but at scale,

like enriching your experience in all kinds of ways.

You mentioned the assistant too,

that’s really interesting how you can have AI assistants

helping you out on different levels

of sort of intimacy of communication.

It could be just like scheduling

or it could be like almost like therapy.

Clearly I need some.

So let me ask you,

you’re one of the most successful people ever.

You’ve built an incredible company

that has a lot of impact.

What advice do you have for young people today?

How to live a life they can be proud of?

How to build something that can have a big positive impact

on the world?

Well, let’s break that down.

Cause I think you proud of, have a big positive impact.

Well, you’re actually listening.

And how to live your life

are actually three different things that I think,

I mean, they could line up,

but, and also like what age of people are you talking to?

Cause I mean, I can like.

High school and college.

So you don’t really know what you’re doing,

but your dream big.

And you really have a chance to do something unprecedented.

Yeah.

So I guess just to.

Also for people my age.

Okay, so let’s maybe start with the kind of most

philosophical and abstract version of this.

Every night when I put my daughters to bed,

we go through this thing and like,

they call it the good night things.

Cause we’re basically what we talk about at night.

And I just, I go through them.

Sounds like a good show.

The good night things.

Yeah.

Priscilla’s always asking, she’s like,

can I get good night things?

Like, I don’t know.

You go to bed too early.

But it’s,

but I basically go through with Max and Augie,

what are the things that are most important in life?

Right.

That I just, it’s like, what do I want them to remember

and just have like really ingrained in them as they grow up?

And it’s health, right?

Making sure that you take care of yourself

and keep yourself in good shape,

loving friends and family, right?

Because having the relationships,

the family and making time for friends,

I think is perhaps one of the most important things.

And then the third is maybe a little more amorphous,

but it is something that you’re excited about for the future.

And when I’m talking to a four year old,

often I’ll ask her what she’s excited about

for tomorrow or the week ahead.

But I think for most people, it’s really hard.

I mean, the world is a heavy place.

And I think like the way that we navigate it

is that we have things that we’re looking forward to.

So whether it is building AR glasses for the future

or being able to celebrate my 10 year wedding anniversary

with my wife that’s coming up,

it’s like, I think people,

you know, you have things that you’re looking forward to.

Or for the girls, it’s often I want to see mom

in the morning, right?

It’s just, but it’s like that’s a really critical thing.

And then the last thing is I ask them every day,

what did you do today to help someone?

Because I just think that that’s a really critical thing

is like, it’s easy to kind of get caught up in yourself

and kind of stuff that’s really far down the road,

but like, did you do something just concrete today

to help someone?

And, you know, it can just be as simple as, okay, yeah,

I helped set the table for lunch, right?

Or, you know, this other kid in our school

was having a hard time with something

and I like helped explain it to him.

But in that those are, that’s sort of like,

if you were to boil down my overall life philosophy

into what I try to impart to my kids,

those are the things that I think are really important.

So, okay, so let’s say college.

So if you’re a graduate in college,

probably more practical advice, I’m always very focused

on people.

And I think the most important decision

you’re probably gonna make if you’re in college

is who you surround yourself with,

because you become like the people

you surround yourself with.

And I sort of have this hiring heuristic at Metta,

which is that I will only hire someone to work for me

if I could see myself working for them.

Not necessarily that I want them to run the company

because I like my job, but in an alternate universe,

if it was their company and I was looking

to go work somewhere, would I be happy to work for them?

And I think that that’s a helpful heuristic

to help balance, you know,

when you’re building something like this,

there’s a lot of pressure to, you know,

you wanna build out your team,

because there’s a lot of stuff that you need to get done.

And everyone always says, don’t compromise on quality,

but there’s this question of, okay,

well, how do you know that someone is good enough?

And I think my answer is, I would want someone

to be on my team if I would work for them.

But I think it’s actually a pretty similar answer

to like, if you were choosing friends or a partner

or something like that.

So when you’re kind of in college,

trying to figure out what your circle is gonna be,

trying to figure out, you know,

you’re evaluating data,

your circle is gonna be trying to figure out, you know,

you’re evaluating different job opportunities.

Who are the people, even if they’re gonna be peers

in what you’re doing,

who are the people who in an alternate university,

you would wanna work for them,

because you think you’re gonna learn a lot from them,

because they know, because they are kind of values aligned

on the things that you care about,

and they’re gonna like, and they’re gonna push you,

but also they know different things

and have different experiences

that are kind of more of what you wanna become like

over time.

But I don’t know, I think probably people are too,

in general, objective focused,

and maybe not focused enough on the connections

and the people who they’re basically building relationships

with.

I don’t know what it says about me,

but my place in Austin now has seven legged robots.

So I’m surrounded myself by robots,

which is probably something I should look into.

What kind of world would you like to see your daughters

grow up in, even after you’re gone?

Well, I think one of the promises of all the stuff

that is getting built now is that it can be a world

where more people can just live out their imagination.

One of my favorite quotes,

I think it was attributed to Picasso,

it’s that all children are artists,

and the challenge is how do you remain one

when you grow up?

And if you have kids, this is pretty clear,

I mean, they just have wonderful imaginations.

And part of what I think is gonna be great

about the creator economy and the metaverse

and all this stuff is this notion around

that a lot more people in the future

are gonna get to work doing creative stuff

than what I think today we would just consider

traditional labor or service.

And I think that that’s awesome.

And that’s a lot of what people are here to do

is collaborate together, work together,

think of things that you wanna build and go do it.

And I don’t know, one of the things

that I just think is striking,

so I teach my daughters some basic coding with Scratch.

I mean, they’re still obviously really young,

but I think of coding as building,

where it’s like when I’m coding,

I’m building something that I want to exist.

But my youngest daughter, she’s very musical

and pretty artistic and she thinks about coding as art.

She calls it code art, not the code,

but the output of what she is making.

It’s like, she’s just very interesting visually

in what she can kind of output and how it can move around.

And do we need to fix that?

Are we good?

What happened?

Do we have to clap, Alexa?

Yeah, so I was just talking about Augie and her code art,

but I mean, to me, this is like a beautiful thing, right?

The notion that like for me,

coding was this functional thing and I enjoyed it.

And it like helped build something utilitarian,

but that for the next generation of people,

it will be even more an expression

of their kind of imagination and artistic sense

for what they want to exist.

So I don’t know if that happens,

if we can help bring about this world

where a lot more people can,

that that’s like their existence going forward

is being able to basically create

and live out all these different kinds of art.

I just think that that’s like a beautiful

and wonderful thing and will be very freeing for humanity

to spend more of our time on the things that matter to us.

Yeah, allow more and more people to express their art

in the full meaning of that word.

That’s a beautiful vision.

We mentioned that you are mortal.

Are you afraid of death?

Do you think about your mortality?

And are you afraid of it?

You didn’t sign up for this on a podcast, did you?

No, I mean, it’s an interesting question.

I mean, I’m definitely aware of it.

I do a fair amount of like extreme sport type stuff.

So like, so I’m definitely aware of it.

And you’re flirting with it a bit.

I train hard.

I mean, so it’s like, if I’m gonna go out

in like a 15 foot wave.

Go out big.

Well, then it’s like, all right,

I’ll make sure we have the right safety gear

and like make sure that I’m like used to that spot

and all that stuff.

But like, but you know, I mean, you.

The risk is still there.

You take some head blows along the way.

Yes, but definitely aware of it.

Definitely would like to stay safe.

I have a lot of stuff that I want to build and want to.

Does it freak you out that it’s finite though?

That there’s a deadline when it’s all over

and that there’ll be a time when your daughters are around

and you’re gone?

I don’t know.

That doesn’t freak me out.

I think, I don’t know.

Constraints are helpful.

Yeah.

Yeah, the finiteness makes ice cream

taste more delicious somehow.

The fact that it’s gonna be over.

There’s something about that with the metaverse too.

You want, we talked about this identity earlier,

like having just one, like NFTs.

There’s something powerful about the constraint

of finiteness or uniqueness.

That this moment is singular in history.

But I mean, a lot of,

as you go through different waves of technology,

I think a lot of what is interesting is

what becomes in practice infinite

or kind of there can be many, many of a thing

and then what ends up still being constrained.

So the metaverse should hopefully allow

a very large number or maybe in practice,

hopefully close to an infinite amount of expression

and worlds, but we’ll still only have

a finite amount of time.

Yes.

I think living longer I think is good.

And obviously all of my, our philanthropic work is,

it’s not focused on longevity,

but it is focused on trying to achieve

what I think is a possible goal in this century,

which is to be able to cure, prevent

or manage all diseases.

So I certainly think people kind of getting sick

and dying is a bad thing because,

and I’m dedicating almost all of my capital

towards advancing research in that area to push on that,

which I mean, we could do a whole,

another one of these podcasts about that

because that’s a fascinating topic.

I mean, this is with your wife Priscilla Chan,

you formed the Chan Zuckerberg Initiative,

gave away 99% or pledged to give away 99%

of Facebook non meta shares.

I mean, like you said, we could talk forever

about all the exciting things you’re working on there,

including the sort of moonshot of eradicating disease

by the mid century marker.

I don’t actually know if you’re gonna ever eradicate it,

but I think you can get to a point where you

can either cure things that happened, right?

So people get diseases, but you can cure them.

Prevent is probably closest to eradication

or just be able to manage as sort of like ongoing things

that are not gonna ruin your life.

And I think that that’s possible.

I think saying that there’s gonna be no disease at all

probably is not possible within the next several decades.

Basic thing is increase the quality of life

and maybe keep the finiteness

because it makes everything taste more delicious.

Maybe that’s just being a romantic 20th century human.

Maybe, but I mean, but it was an intentional decision

to not focus on our philanthropy on like explicitly

on longevity or living forever.

Yes.

If at the moment of your death, and by the way,

I like that the lights went out

when we started talking about death.

You get to meet God.

It does make it a lot more dramatic.

It does.

I should get closer to the mic.

At the moment of your death, you get to meet God

and you get to ask one question.

What question would you like to ask?

Or maybe a whole conversation.

I don’t know.

It’s up to you.

It’s more dramatic when it’s just one question.

Well, if it’s only one question and I died,

I would just wanna know that Priscilla and my family,

like if they were gonna be okay.

That might depend on the circumstances of my death.

But I think that in most circumstances that I can think of,

that’s probably the main thing that I would care about.

Yeah, I think God will hear that question and be like,

all right, fine, you get in.

That’s the right question to ask.

Is it?

I don’t know.

The humility and selfishness.

All right, you’re in.

I mean, but well, maybe.

They’re gonna be fine.

Don’t worry, you’re in.

Okay, but I mean, one of the things that I think

I struggle with at least is on the one hand,

that’s probably the thing that’s closest to me

and maybe the most common human experience.

But I don’t know, one of the things that I just struggle with

in terms of running this large enterprise is like,

should the thing that I care more about

be that responsibility?

And I think it’s shifted over time.

I mean, like before I really had a family

that was like the only thing I cared about.

And at this point, I mean, I care deeply about it,

but yeah, I think that that’s not as obvious of a question.

Yeah, we humans are weird.

You get this ability to impact millions of lives

and it’s definitely something, billions of lives,

it’s something you care about,

but the weird humans that are closest to us,

those are the ones that mean the most.

And I suppose that’s the dream of the metaverse

is to connect, form small groups like that

where you can have those intimate relationships.

Let me ask you the big, ridiculous.

Well, and to be able to be close,

not just based on who you happen to be next to.

I think that’s what the internet is already doing

is allowing you to spend more of your time

not physically proximate.

I mean, I always think when you think about the metaverse,

people ask this question about the real world.

It’s like the virtual world versus the real world.

And it’s like, no, the real world is a combination

of the virtual world and the physical world.

But I think over time, as we get more technology,

the physical world is becoming less of a percent

of the real world.

And I think that that opens up a lot of opportunities

for people, because you can work in different places.

You can stay more close to, stay closer to people

who are in different places.

So I think that’s good.

Removing barriers of geography

and then barriers of language.

That’s a beautiful vision.

Big, ridiculous question.

What do you think is the meaning of life?

I think that, well, there are probably a couple

of different ways that I would go at this.

But I think it gets back to this last question

that we talked about, about the duality

between you have the people around you

who you care the most about,

and then there’s like this bigger thing

that maybe you’re building.

And I think that in my own life, I mean,

I sort of think about this tension,

but I mean, it’s like, I started this whole company

and my life’s work is around human connection.

So I think it’s intellectually probably the thing

that I go to first is just that human connection

is the meaning.

And I mean, I think that it’s a thing

that our society probably systematically undervalues.

I mean, I just remember when I was growing up

and in school, it’s like, do your homework

and then go play with your friends after.

And it’s like, no, well, what if playing

with your friends is the point?

That sounds like an argument your daughter would make.

Well, I mean, I don’t know, I just think it’s interesting.

Homework doesn’t even matter, man.

Well, I think it’s interesting because it’s,

and people, I think people tend to think

about that stuff as wasting time,

or that’s like what you do in the free time that you have.

But like, what if that’s actually the point?

So that’s one.

But here’s maybe a different way of counting out this,

which is maybe more like religious in nature.

I mean, I always like,

there’s a rabbi who I’ve studied with

who kind of gave me this,

we were talking through Genesis and the Bible and the Torah

and they’re basically walking through,

it’s like, okay, you go through the seven days of creation

and it’s basically, it’s like,

why does the Bible start there?

Right, it’s like it could have started anywhere,

right, in terms of like how to live.

But basically it starts with talking about

how God created people in his, her image.

But the Bible starts by talking about

how God created everything.

So I actually think that there’s like a compelling argument

that I think I’ve always just found meaningful

and inspiring that a lot of the point

of what sort of religion has been telling us

that we should do is to create and build things.

So these things are not necessarily at odds.

I mean, I think like, I mean, that’s,

and I think probably to some degree

you’d expect me to say something like this

because I’ve dedicated my life to creating things

that help people connect.

So, I mean, that’s sort of the fusion of,

I mean, getting back to what we talked about earlier,

it’s, I mean, what I studied in school

or psychology and computer science, right?

So it’s, I mean, these are like the two themes

that I care about, but I don’t know for me,

that’s kind of what I think about, that’s what matters.

To create and to love, which is the ultimate form

of connection.

I think this is one hell of an amazing replay experience

in the metaverse.

So whoever is using our avatars years from now,

I hope you had fun and thank you for talking today.

Thank you.

Thanks for listening to this conversation

with Mark Zuckerberg.

To support this podcast, please check out our sponsors

in the description.

And now, let me leave you with the end of the poem, If,

by Roger Kipling.

If you can talk with crowds and keep your virtue,

or walk with kings, nor lose the common touch,

if neither foes nor loving friends can hurt you,

if all men count with you, but none too much.

If you can fill the unforgiving minute

with 60 seconds worth of distance run,

yours is the earth and everything that’s in it.

And which is more, you’ll be a man, my son.

Thank you for listening and hope to see you next time.