So first of all, just a moment on AI. So
you asked a question earlier which is
like is is there like a there's a
traditionalist view of like we need to
go get the jobs that used to exist. We
need to somehow bring them back. And and
you know, and here you get into this
argument of like well, you've got you
know, Chinese laborers working for a
dollar an hour or whatever. Like you're
not going to get those jobs back. It's
just not cost-effective to bring those
jobs back. What we have though is we're
we're at this very specific and
important and fundamental and I think
profound turning point in technology
which is the rise of AI. Um and I would
say that the good news, let's make the
panel optimistic. The good news is the
US is like by far the leader on AI. Now,
by the way, we need to decide whether we
want to be the leader in AI cuz there's
like a lot of forces in the US and in
the US government that are trying to
prevent that from happening.
>> But there's over a thousand laws at the
state level we've been trying to fight
and stop that would screw this up,
right? So there's a lot of people who
are against this.
>> California state state government almost
passed a law to outlaw AI in California
uh six months ago which which is an
indication of how crazy California is.
But yes, that we we we can it's like
everything else. We can have it if we
want it. Uh these are all choices. But
like AI's happening and AI's only
happening in two places. Like it's only
US versus China. There's a whole
competition thing there. But like the US
is like very powerful on this. And then
by the way, US is benefiting a lot cuz
Europe has basically made AI illegal.
Like they're they're regulating
themselves to death. And so the smart AI
people in Europe are moving to the US.
Um and so like the the US really is
ground zero for AI. Um today AI is
software, right? And so when you use AI,
you use ChatGPT. It's an app on your
phone, right? It it doesn't there's no
manufacturing kind of component, you
know, really to you know, people are
building giant data centers. But you
know, it's not a huge generator of like
middle class jobs. Um but there's
another turn on AI that's coming which
is the turn to the turn to embodied
physical AI which is robotics, right? Um
and this is a transition that has
already happened. It's already actually
happened in drones uh where drones went
from being human piloted to actually be
auto you know, autonomously piloted.
They they now fly themselves.
Um it's a transition that's happening in
cars right now, right? Where cars went
from, you know, basically purely
physical products to now basically being
rolling computers that drive themselves.
And if you if you haven't tried this
yet, if go to go to if you ever have the
misfortune of finding yourself in San
Francisco, uh take a Waymo.
>> [snorts]
>> It's amazing to live in a cyberpunk
future reality where you're in literally
a state of the art self-driving car
that's literally out of the Jetsons
driving past people who are dying of
fentanyl overdoses on the sidewalk. I
recommend having the experience at least
once.
But the self-driving car really does
work. Like it's really happening. And
then what's going to happen you've all
probably seen you know Elon has this
Optimus robot that he's building.
These humanoid robots like the
general purpose robotics thing is going
to happen and it's going to happen in
the next decade and it's going to happen
at giant scale. I I think there's a
plausible argument which Elon also
believes that robotics is going to be
the biggest industry in the history of
the planet. It's going to be gigantic.
There's going to be billions, tens of
billions, hundreds of billions of robots
of all shapes, sizes, descriptions
running around doing all kinds of
things. Those robots need to get
designed and built, right? And so my
view is you know we don't try to get the
old manufacturing jobs back. What we
what we should do is lean hard into the
manufacturing jobs of the future, which
is designing and building all of these
new things, right? By the way which
includes drones and cars, right? And
robots.
And you know we shouldn't be building
manufacturing lines that have people
sitting on a rubber mat for 10 hours
screwing screws in by hand, but we
should be building you know what Elon
calls he calls them alien dreadnought
factories, right? Which is like these
super sophisticated factories with like
tons of robotics.
But you can imagine thousands and
thousands of different categories of
industrial production that have to
happen all across the US in order to
make this happen.
And and would cause just like enormous
economic growth throughout the country.
We'd get a huge payoff from all the tech
investment that that the coasts are
making, but we generate many tens or
hundreds of millions of jobs in the
countryside. And then the US would would
lead. It would be you know be the third
industrial or fourth industrial
revolution. We would we would lead in
the development of all these new things.
And then you get to the national
security side of this which is if you
don't do this you're living in a world
of Chinese robots everywhere.
And that has you know very profound
consequences for let's just say yeah
profound consequences.
>> I know you've been reading about and
President Trump's been reading about
from 100 years before which is which is
William McKinley. William McKinley of
course was a staunch protectionist at
first.
Uh and and is this historic historical
example of how to revitalize American
industry relevant today? Is it relevant
to everything we're doing in AI and in
energy? Like, how do you think about
this?
>> Yeah, so I've been doing my best to kind
of spin up in the history of this. And
so there there was something in the 19th
century, uh actually originally created
by Alexander Hamilton uh when he was
Treasury Secretary, called the American
system. Um and and and really what what
they were and this is even, you know,
this is like 80 years before McKinley,
um 90 years before McKinley. Uh you
know, what they were dealing with was
the the the the British Empire, right?
The UK was the preeminent industrial
power. You know, they they were the
first like giant industrial power uh on
on on planet Earth. Um and it had built
this, you know, very commanding
economic, technological, military uh
position. And of course, America, you
know, at the at the time of the
revolution was mostly farmland. Um and
then there was this raging debate
between Hamilton and Jefferson as to
whether you you'd want America to be an
industrial, you know, kind of urban,
financialized power, which is what
Hamilton wanted, or of kind of a rural,
you know, kind of rural farming
aristocratic uh power, which is what
Jefferson wanted. And you know, and
Hamilton won that debate. And so that
led to what was called the American
system. The American system led to, you
know, the creation of the United States
as an industrial power that developed
over the course of the 19th century. But
it really took off at the tail end of
the 19th century in what was called the
what's called the Second Industrial
Revolution. And so between, call it
1870, 1880, and like 1930, basically
everything we view today as the modern
world um was basically invented uh then.
And and like you could almost look
around the room here today, and you
could say basically every piece of
physical infrastructure, everything
around us, you know, literally
everything. The airplane, um you know,
manufacturing techniques to be able to
make um
uh to be able to make uh you know,
modern modern, you know, glassware,
um uh at scale, um uh air conditioning,
um you know, television, radio, uh you
know, auto uh the automobiles we drove
to get here, um you know, it it all was
a uh the the result of the Second
Industrial Revolution. And then the
McKinley presidency was right in the
middle of that of of of that period.
It's really when the whole thing
catalyzed. Um and so, it was really the
point at which America became the the
the world's industrial superpower. And
then, of course, that led to America
being able to become a huge player in
both World War I and World War II, you
know, ultimately, you know, we you know,
you you know, winning winning World War
II and and and becoming the uh you know,
the arsenal of democracy.
They all kind of flowed out of that
period. And and so, and and the and the
the American system is something that
people really don't like to talk about
these days cuz it was heavily
protectionist. Um and the the whole
theory of it was you you don't
industrialize uh just with like a single
technology or a single company. You
industrialize by having an entire
ecosystem, right? You industrialize by
having, you know, not just three car
companies, but by having all the
thousands of parts suppliers that make
all the parts for the cars. Um and so,
the the Brits had done this. They had
run a very uh heavily protectionist uh
program to be able to boot up the first
industrial revolution in the UK. You
know, they then advocated for free trade
when exports became more important. Uh
McKinley was the inheritor of the
American system. And so, he was heavily
protectionist to build up the second
industrial revolution. And then, he
actually switched to become more of an
advocate for free trade because his
American exporters were becoming more
successful. They needed exports. They
needed They needed foreign markets.
Um and then, by the way, um in the 20th
century, basically, every industrial
revolution in every other country that
has worked in the last 100 years
basically follows that model as well.
So, you know, it's it's the same thing
Germany did. It's the same thing Japan
did. It's the same thing Korea did. It's
the same thing China uh is doing right
now. China's basically been running the
American system playbook um in China.
And so, and then, this this goes to
tariff policy. And like, I and and and
and I'm not here to like make an
argument like it's like in favor of
tariffs, but more of like a historical
observation that there has been this
phenomenon over time that if you want to
have the kind of clustering effect that
leads to industrial preeminence, uh it
is not unusual uh to have, you know, a
combination of of of of of tools and you
know, kind of tools and methods used to
do that. You know, it it has been
historically characteristic that tariffs
would be a part of that. But, it's also
historically characteristic that there
there becomes this point when you also
therefore want to go for exports and you
want to go for free trade. Um and and
and McKinley specifically, you know, he
ran for office very much on the idea of
having a high and permanent tariff. Once
he was in office, he switched more to a
model of what he what he called
reciprocity.
Uh right? And and basically it was, you
know, we we start out by tariffing, but
the goal actually is to get everybody to
lower their tariffs. That that you know,
the goal is to get a a a world of much
more free trade and then therefore these
American exporters are able to to
succeed globally. Um and and of course
he was able he was able to use tariffs
as American tariffs as a lever to get
other countries to drop their tariffs.
And so
in that sense, [clears throat] what
we're going through right now in the US
is is sort of very characteristic of
that. And what makes maybe the whole
thing, you know, kind of more poignant
is, you know, there is this kind of
fundamental underlying question, which
is do we want the US to be an industrial
superpower or not? Right? And in the
19th century they were like completely
clear that they wanted that. And in in
our time, you know, I I think we're
we're still ambiguous about
>> Uh I want to shift a little bit ask you
about how Silicon Valley, which, you
know, has really been that one of the
greatest engines of innovation in human
history, how did it become so illiberal,
uh woke we call it, and how do how do we
fix it? How do we break out of that?
>> Yeah, so it's a real you know, it's a
real puzzle. Um I I think there's a very
straightforward explanation for the why
it happened, um which is basically it
has to do with a
one of the one of the fundamental
personality traits of human beings
called openness, trait openness, which
is one of the big five kind of
personality traits.
Uh openness, which is by definition, um
you know, people who are super into
technology and entrepreneurship are, you
know, we're we're highly open people. We
love new ideas and and and we love doing
new things.
Um and then there's just on the
political spectrum it just openness is
is openness is a trait of the left. Um
and uh and then you know, the the on the
other side on the right is
conscientiousness.
Um you know, so structure and order,
which is a a trait on the right. And so
so we we just And by the way, this is
not just tech. This is true in the arts.
Any any any time you have a clustering
of people who are high in openness,
inevitably they're they're on the
they're on the political left and
usually very far uh to the political
left. And so I I think we're we're we're
just a classic example of what happens
when very open people cluster together.
Um, you know, you read these kind of
again, these kind of historical things
in the newspaper or whatever that
Silicon Valley is so, you know, hotbed
of, you know, libertarianism or, you
know, incipient whatever cryptofascism
or whatever. And it's like it's just
like not true. It's like, you know, the
valley is like 99% you know, Democrat
and you can see that in the political
donations and the voting patterns
um, and and and everything else. And so,
it is just it's a it's a classic
left-wing democratic um, environment.
Um, you know, look,
I think people should have the politics
that they want to have if they're voting
for the people they like and they're
getting the results that they want.
>> But I didn't just say like left or
liberal, I said illiberal too. Like why
is this so illiberal? Cuz that wasn't
the case in the '90s, right? Or was it?
>> Well, but you know, it's it's a this is
the para- the paradox of tolerance,
right? Like to right to be to be
maximally to be maximally tolerant, we
can we have to be intolerant of
intolerance. Right? Um, and so, right?
And so, it it it is absolutely morally
required if if if you're going to
maximize tolerance above everything
else, you have to be intolerant towards
the intolerant. And and and and and
that's how liberalism becomes
illiberalism. I I quibble with the
illiberal illiberal quibble with the
illiberalism framing of it because I
actually don't think it's something you
and I might disagree on. I don't think
illiberalism I don't think what's
happening now is a betrayal of
liberalism. I think it's just the full
realization of liberalism. Like I think
it's it's it's what happens when you
take liberalism to the extreme, which is
you declare war on the intolerant,
right? And then and then the fact that
you yourself have become intolerant,
like somehow you never quite like wrap
your head around that, but off you go.
>> I think a lot of us have noticed that
our cities are deeply dysfunctional.
They're they're broken in our society,
right? You have I think we have a
candidate who's like a socialist
Islamist who's in danger of taking over
New York very soon right now. And we
have we have we have all sorts of
problems. People can't afford to have
kids in cities the way they're set up. I
think there's this cost disease for
everything. Like what what went wrong?
What's going on with these cities? And
and what's what's an optimistic take on
it, Mark? How do you how do we fix this?
>> Yeah, so the guy in the guy in New York
is interesting And he's you know kind of
running slight slightly behind Cuomo for
New York mayor, but he's literally
running a platform among other things of
having city-owned grocery stores.
Right. And so and so that you know the
joke is you know he's reinventing
breadlines you know for the 21st century
like you know
really tremendous tremendously
tremendous platform.
Um the Seattle is another case study of
this. Seattle to give you a sense how
crazy things are that that the district
in Seattle the city council district
where the most Microsoft and Amazon
executives lived live elected
city councilwoman who ran a platform of
nationalizing Microsoft and Amazon.
>> [laughter]
>> Um right. And then you know in the Bay
Area I've got you know it's you know I I
live in I live in ground zero of
dysfunction and collapse and I've got
I've got I've got thousands of stories
from the Bay Area.
Yeah. So like I think there's like a
very fundamental political economy thing
happening which which is which which
falls out of this sort of economic
transformation that's happened and this
policy transformation that's happened.
And so the
there's a guy there's a French actually
a French writer geographer who has I
think the best analysis on this. Um
and he's written a great book on it
called Twilight of the Elites. And what
he basically observes is if you look at
all the Western countries through this
process of sort of deindustrialization
and financialization and technology you
know sort of technology knowledge work
you know becoming preeminent at the
height of the economy you sort of have
this pattern where
basically like
in an agricultural economy or in a
manufacturing economy you kind of have
to spread everything out throughout the
country cuz like natural resources
really matter and energy really matters
and physical infrastructure matters and
access to you know trains and all these
things you know really rivers and water
and so forth really matters. But if
you're if you're going to replant your
the sort of commanding heights of your
economy on knowledge work then what
you're going to do is you're going to
slam basically all economic growth into
the cities. Right? Cuz you you which is
what what Silicon Valley has done.
Right? So you want all the software
developers in one place. And and you
want them you know they're going to the
good news is they're going to eat at
high-end restaurants and they're going
to employ a lot of nannies. But like
fundamentally they're going to be
sitting in conference rooms you know
designing software together.
And and you want that to happen in a
single concentrated place.
Um
And so what happens is basically this
the city's become a magnet for basically
two kinds of people. One is sort of the
upper class, which is to say kind of the
you know, highly educated elite in the
knowledge of professions
uh who make enough money where they can
afford to buy, you know, $3 million
houses. Um and then you've basically got
the So you've got basically like super
expensive homes and then you've got
public housing, which is so-called
public housing, right? Council estates
in the UK or section 8 8H or whatever it
is housing here. Um where you've got,
you know, basically public housing. And
so and then and then the public housing
is is is, you know, it's basically it's
it's, you know, it's it's, you know,
people who are very low-paid, kind of on
the you know, low end of the service
economy.
>> [snorts]
>> Um and then and then and then that's the
city. And and then who's squeezed out of
the city is basically, right, the entire
middle class is squeezed out. And just
to give you an example of this in the
Bay Area, if you're a cop or a
firefighter in the Bay Area and you work
in Palo Alto, you can be there 3 hours
both ways.
Uh just to be able to like have a house
and afford a family, right? And and then
of course you look what's happening in
the cities and like it's, you know,
there's there's like lots of dogs,
there's very few kids, right? And it's
like, you know, what on earth is that
about? Well, you you can't buy a house.
You you you can't you can't you can't
have a family. Um you may not even own a
car. Like how are you going to kids
anywhere?
But what happens is from from a
political standpoint is the cities then
are run, right? They're in in in their
local democracies, it's these super
high-end kind of, you know, very
progressive elites and then it's sort of
the clientele underclass, right? Um that
um you know, of of, you know, in the US
it's Hispanic immigrants or African
Americans. You know, in Europe it's
it's, you know, they're you know,
they're they're other kind of uh
clientele underclass that they've
developed in places like Paris. Um and
so you've got this high plus you you've
got this kind of old political kind of
structure of of, you know, sort of top
plus bottom, you know, kind of
configuration. What you've done is you
squeezed the middle and you've ejected
the middle out of the cities into the
countryside, but you've deindustrialized
and so there's no new economic activity
happening in the countryside uh for all
the middle class people to do.
>> so far is not an optimistic panel, Mark.
>> I'm sorry.
>> [laughter]
>> Sorry.
Let me try to make it optimistic. Sorry.
Let me try to make it optimistic.
>> I okay.
>> Let me try to make it optimistic. These
are choices. Like these are choices.
These are policy choices. Like
sometimes the way this stuff gets
painted as that these are just like
impersonal economic forces or technology
forces. And of course, you know, I get
this a lot. It's like you idiot, don't
you understand that like, you know, it's
it's just like, you know, manufacturing
is old fashion. We have this all this
new stuff. We shouldn't need, you know,
what what No, these are choices. Like
there were specific choices made. So,
the classic choice the classic choice
that was made I'll be since I'm at the
Reagan Library I'll be I'll beat up on
Nixon. Um, so Richard Nixon the year I
was born 1971 he proposed something
called Project Independence. Uh, he he
actually visionary visionary president.
He saw the energy crisis coming. Um, and
he said it's a national imperative to
build a thousand new nuclear power
plants
uh, in the civilian nuclear power plants
in the US by the year 2000. Cut the
entire US electric grid over to nuclear.
>> Right.
>> It would by the way it would have gotten
us to electric cars, you know, probably
30 40 years faster.
Um, and would have completely separated
us from strategic involvement in the in
the in the Middle East for for for
energy. Um, so that was the good news.
The bad news is he created the EPA and
the Nuclear Regulatory Commission which
then prevented Project Independence from
happening and of course the no new
nuclear power plants got built.
But But that was a policy choice, right?
And And I'm sure that was not the out
that was not the result that he
intended, but that was a policy choice.
Um, and so you know, And And by the way,
a lot of this is happening right now in
Washington. Like there is an opportunity
to look at everything from nuclear
power, minerals, natural resources. I
was on a call this morning on it natural
resources with very promising progress
being made. Um, you know, rare earth
metals there's in you know,
construction.
We have a project to build a new city
um, in California. Um, we bought, you
know, four times the acreage of of
Manhattan. Uh, in Solano County. Um, you
know, we have the opportunity to build
like entirely new cities in this
country. Um, you know, we like there's
any number of things any number of these
policy decisions that we could make that
would cause a revitalization.
>> So, we're So, we're going to fix better
policies. There's something else going
on though which I think you and I are
both very bullish on and which is which
is these AI wave and AI services, right?
And so I think I think one thing is very
interesting right now is obviously we
all know there's major cost disease in
our society. Healthcare is becoming more
and more unaffordable. We I think people
are talking about our $36 trillion debt.
It's actually 150 trillion more if you
include our health care promises. It's a
much bigger problem. Uh you obviously
education, housing, all these things are
becoming less affordable. But but you
and I have both seen that a lot of these
new technologies are going to turn some
of these services into much more
affordable products. And I think we're
making big bets on that. So tell tell us
about this.
>> The this new regulation on AI that's
coming out will basically completely
prevent anyone from using AI at all in
Europe and just put them way behind. And
you know, everyone hates Brexit in the
UK where I am right now. Well, not
everyone, but a lot of people do. And in
London they hate it overall. Uh but one
of the bright sides may be that they
escape Europe's terrible regulatory
regime on AI. Maybe it maybe frees them
from it ironically and helps out. Like
what are the actual biggest dangers for
countries with AI? Like is Europe going
to be left behind? And and and and and
and how should we think about the race
with China in that regard as well?
>> Yeah, so let me look. Europe seems I
just take them at their word. They seem
determined to They seem determined They
seem enthusiastic about being left
behind. Um they seem to really badly
want that. Um you know, that look they
data Here's one where the where the data
is clear, which is you know, 25 years
ago if you did a list of the 50 biggest
tech companies in the world, like a
pretty significant number of them were
European. It's like I don't know,
probably 15 of the top 50 or something.
You know, today you're you're basically
down to Spotify.
Um which is great. I love Spotify. But
like it's you know, that's not that's
not what the European tech sector used
to be.
Um and so you know, they they have sort
of deliberately disarmed and turned away
from
technology and progress and capitalism
in really profound ways for a long, you
know, for decades now. And and yeah,
look the the EU is designed to do this.
That's kind of the whole point of the
structure is to impose these kinds of
controls and stop, you know, stop stop
these kinds of things.
Um and yeah, they they I mean they they
they're very vocal about it. Like they
they very much want it. I I think they
would be very happy if AI didn't happen
inside the EU. I think they'd be very
happy if social media didn't exist. I
think they'd be quite happy to shut up
the internet. Um and so I I think the
you know, the citizenry of the EU have
an even more kind of pointed version of
the of the question that that that that
that our own citizens have.
You know, I have a lot of friends who
are like, you know, optimistic about
this or that or the French are going to
save it or whatever. And it's like I I
hope that happens, but the overall, you
know, kind of vector there is is is is
in a very clear direction. Um and then
look at look, China's the other China's
the the sort of, you know, the China's
sort of the the dog that didn't bark in
a lot of these conversations. So So what
I find is when I talk to people in
Washington about AI or tech policy
generally, there's two completely
different conversations I can have with
the same people the same person and on
the same day. Um so conversation number
one is a conversation about tech in the
US, right? Um and so tech in the US, US
government, US tech companies, US tech
policy, you know, US society, all the
all the all these issues. And and and
that's a discussion that's characterized
by just enormous antipathy and loathing
from Washington towards American tech
and towards, you know, increasingly just
like flat-out capitalism, market
economy, uh innovation kinds of things.
Um
I can have a completely different
conversation with that same person where
China's on the table. Um and the minute
the China topic's on the table, it's a
completely different conversation, which
is how does the American public sector
or private sector team up to make sure
that we win versus China in this sort of
new cold war cold cold war that's
underway. Um and so, you know, the the
cynical optimist in me would say as long
as China's a big enough threat, it will
be sufficient to kind of glue us
together uh to be able to actually make
some rational decisions
uh as a society. Um you know, may maybe
the presence of an external enemy is
actually helpful to kind of galvanize,
you know, kind of your own side because
to you know, to organize and and and be
smarter.
Um you know, the cynical pessimist,
right, would say China's not enough of a
threat. Um and uh you know, boy, it
would maybe be actually even a little
bit more of a threat.
Um you know, many other people would
find this entire line of conversation
horrifying. I I I would just say
I I I don't actually know how to have a
productive conversation with people in
DC right now if they if they if they
don't have China in their head cuz
otherwise the the the sort of the the
the the the the negative emotion, fear,
whatever, um panic just turns internal
and I I and I think it's very dangerous.
>> A lot of people are very scared right
now with the disruption. I'm very
excited about the higher productivity
cuz I see it being disinflationary. I
see it creating more jobs. How How do we
How do we If assuming we're aligned, how
do we explain this to people? How do we
explain this is actually a really good
thing for the country cuz a lot of
Luddites are going to attack this. They
are going to try to stop it.
>> Yeah, so the the Luddites the the
Luddites are wrong. Neil Neil Luddites
are wrong. They're wrong in the sense of
like the for the reason they've always
been wrong, which is the reality is we
have not in the last 50 years been in an
era of fast productivity growth
technological change. We've been in an
area of slow productivity change, slow
economic growth.
Uh AI if it works the way that actually
people either hope or fear that it's
going to work, it's going to cause
productivity growth to shoot up. That
will naturally accelerate uh economic
growth. Um but look, if if it just stays
as software, then the result of that is
it's going to make the cities much
better off, right? It's going to It's
going to make San Francisco and LA and
and New York much better off and Austin
much better off, but it's still not
going to answer this question of what
happens in the countryside. Uh to answer
the question of what happens in the
countryside,
you need you need to get to the next
thing. You need You need to make AI
hardware as as well as software. And And
you're right back to that same
prescription of basically
reindustrializing the country. And
again, not reindus- not
reindustrializing the country on the
stuff that has already left.
Reindustrializing the country based on
all the things that have to be invented
and built.
Um and it's just like we know I mean
just like I'm just like incredibly
highly confident that there's going to
be these giant industries of you know
computerized everything, AI powered all
kinds of hardware of all kinds of
shapes, sizes, and descriptions. You
know, you're finding a lot of that kind
of thing. I'm finding a lot of that kind
of thing. I mean look, the entire
defense base has to get rebuilt around
this.
>> We're going to redo the manufacturing.
We're going to build roads and tunnels
and everything cheaper. It's going to be
exciting.
>> Yeah. And so so so this this is this is
the thing that that that this is the
thing to do. Like the to me the path
like on this is very clear cuz we we
have to we have to do it we have to do
this because it's necessary for national
security standpoint. We have to do it
because we need it we need the economic
growth. We have to do it cuz we need an
answer for the entire population of the
country, not just the cities.
Um and yeah, and we have to do it
because if we don't do it, China's going
to do it and we don't want to live in
that world.
>> Obviously, the second industrial
revolution from 1870 to 1900 or so, I
think it tripled the living standards in
the US. But what do we learn from prior
leaps in human progress and what lessons
should we apply today and what mistakes
should we avoid? You didn't you're not
too concerned about mistakes, but are
there lessons are there mistakes to
avoid? How how how you thinking about
this based on history?
>> Yeah, so look the big thing is for a
very long time we had a model or an
ethos or an ethic of of industrial
progress. It started with like um you
know basically once upon a time it was
uh to to for the greater glory of God it
was to discover the secrets of the
universe
and nature as a way to to glorify God,
but then that that turned into what
became known as natural philosophy which
turned into science
uh and then the the enlightenment hit
and then science became technology. Um
that became industrialization and then
you know we had about a 250 year run or
something from basically like steam
power through basically the 1970s
um where we said wow this is great like
we can build all these new things, we
can do electric you know we can
electrify, uh we can have you know
lighting everywhere,
um you know we can have you know
antibiotics, we can have you know
computers, we can have cars, you know we
we can we can build all these amazing
things. By the way, we can build
weapons, right? We can build like you
know guns and we can have like you know
increasingly powerful tanks and aircraft
and and so forth um
uh and and we can do all these things
you know good and bad or you can say
maybe civilian and military or you know
sort of pacifistic or or potentially
violent, but we we have we have all
these things that that we can do and and
we should do these things like we should
build these things and and by doing that
you know and and certainly in in
aggregate overwhelmingly we we we can
improve the world and that that created
the civilization that we live in today.
Um you know I think basically things
took a bit of a left turn in the 70s um
with what was uh a concept invented by
the German Greens at the time called uh
the precautionary principle. Um and that
sort of uh you know kind of corresponded
with the rise of a lot of the kind of
defining ethical ideas of our period
um you know which basically basically
advocated a much more cautious stance um
and and that's very kind of an an
evocative idea right now number one
because it just seems so obvious to
people in our modern world that if a
technology is just going to release a
new technology the burden should be on
the technologists to prove that the
technology is not harmful and that's
what you keep hearing about about AI.
Um, you know, the observation that I
make is the German Greens invented that
concept in the 1970s to stop nuclear
power. Um, and I think, you know, the
the consequences of stopping nuclear
power 50 years later, you know, it it's
just was just like a catastrophic
mistake. Um, you know, geopolitically,
it's directly connected to Russia's
invasion of Ukraine. Europe is
effectively funding Russia to invade
Ukraine cuz they keep, you know, buying
oil from them. Which they wouldn't need
to buy if they were running all nuclear.
And then of course, carbon emissions.
Nuclear is carbon, you know, zero
carbon. Um, and of course, the
alternative to nuclear and to carbon,
um, you know, oil is is is coal. And so,
you know, emissions in Europe are are
right now spiking through the roof
because they're bringing all these coal
plants online.
Uh, and so anyway, like I would argue in
the last 50 years we've we've we've
played out this experiment with the
so-called precautionary principle. It
did not work well
uh, for the technology it was invented
for. I don't think it makes a lot of
sense for most of the new technologies.
There's now an attempt to apply that
that that that ethos or that rule for
AI, which I think would be another
catastrophic mistake. And as as we sit
here today, I think it it uh, it's uh,
the future hangs in the balance because
there's a lot of people who are pushing
for that right now. At some point you
have to ask the question of like, okay,
what levels of controls are required to
accomplish these various levels of
goals? And then, um, to what extent are
you ultimately basically destroying
yourself um, in order to uh, in order to
in order to basically save the world.
And and and and and this is where things
get very dystopian in the other
direction very, very fast. Um, because
what various people, and I won't I won't
name names and whatever, but like
there's a there's a general development
of policy recommendations that are
coming. Um, including from from some
very big companies and some very
respected people. And they say things
like, well, you know, advanced chips
ought to be registered. Um, and they say
things like advanced chips ought to have
like a monitoring agent on them so that
the government can tell what people are
doing with them.
>> That's scary.
>> Right. And then they ought to have an
override. The government ought to have a
kill switch. And if the government, you
know, sees what you're doing on your
computer and then they don't like it,
they should be able to turn your
computer off. Right. Um, and then, you
know, the more extreme form of this is
what was in Time magazine, which was a
call for air strikes on rogue AI data
centers. Right. And And by the way, in
complete contravention of international
law, right? Just like, yeah, screw it,
bomb them. Um, and then some people in
this movement are saying we should be
willing to risk nuclear war in order to
prevent or prevent hostile foreign
states from using AI. And so, at this
point, like, you're walking down a path
basically saying, I want a global
totalitarian superstate, omnipresent,
right? Um,
uh, authoritarian dictatorship that's
monitoring every computer and every
person and every chip and every piece of
software. Um, and then I want to destroy
anybody who deviates from that. And
that's how I'm going to save things.
Right? And so, that that that's what's
gotten just like it to for me completely
surreal about this whole discussion. And
And I And I think this can only be
happening cuz they we're in a moral
panic right now. Um, but like the
>> [laughter]
>> the rationality has left the room for
the time being, and I think I think we
should bring it back.
>> No, I I mean, and that's totally fair. I
mean, the the the only excuse for that
for me, if we talk about the Fermi
paradox, the paradox of the fact that
that that there, you know, seem to be
trillions of worlds to support life, and
then we're not seeing life. And one of
the one of the possibilities is that as
we get to this point and whenever a
civilization figures out how to create
thinking machines, it destroys itself
somehow with the power of those
machines. But it's the argument they're
making, which I realize is maybe
somewhat ridiculous, but if people
believe that, of course, it would
justify almost any of the things they're
saying if they believe that. But I guess
they're making a really big leap.
>> Yeah, it's the precautionary principle
on steroids. Um, and it leads to
stagnation and and and violence and
madness and death. Um, it's an
incredibly anti-human philosophy.
Um, you know, it's basically anti
everything that's made Western
civilization great, um, in terms of like
what you do as a consequence of that.
Uh, and then look, and then I would make
one I would make one final claim against
that kind of position, and I do hear
that, and it's it's a completely
unscientific position. By definition,
it's a completely unscientific position
because there is no way at all to prove
or disprove the theory.
Right. And so, it is the worst kind of
ivory tower thought experiment. Um, and
you know, the history If we've learned
one thing from the last 100 years, it's
that ivory tower thought experiments
should not be the way that we determine
the shape of society.
>> Well, I think both of us see how health
care doesn't need to take up 20% of our
GDP, right? It's very obvious to AI to
make these things half a third the cost
for better outcomes. And yet, every one
of our states in this country has very
powerful provider and payer lobbies and
has very powerful cartels to block
competition. And so, if we want to go
build something with AI, we're not
allowed to right now. Like, do you see
the left and the right tech working
together to fight these special
interests to actually fix the regulatory
state cuz it's not like the
administration could just do it, right?
We have It's going to take a lot of
effort. Like, what does that coalition
look like of the of the of the good guys
against this kind of broken regulatory
cartel in all these industries? Is that
Is that something we're going to all
have to fight?
>> Yeah, so the the way I think about this
is there's basically, right, three
components to the American dream. Right,
there's house, you know, have a have a
house, have a house for your kids where
and you know, kind of by definition,
this goes back to the geographic lens,
the house needs to be like near a place
where there's like great job
opportunities, right? So, it can't be
can't be a house in the middle of
nowhere, it has to be a house where
there's going to be great jobs,
um and you know, a safe environment and
so forth. Um number one, number two is
is is health care. You need you need
great health care for you and your
family. And then three is education. You
want to send your kids kids to good
schools. Um if you disaggregate sectors
of the economy over the last 25 years,
what you find is basically everything
that technology touches basically
collapses in price. Um and so,
television sets and you know, computer
games and you know, everything, you
know, anything anything with electronics
in it basically that the the the the
price is productivity growth in those
sectors has advanced very fast, product
prices have have have have collapsed.
Everything is is much cheaper. Um but if
you chart uh housing, education, and
health care, um the prices of those
things are skyrocketing, right? And and
we all we all you you we all see this.
It's just the price of housing is
exploding, the price of education is
exploding, and the the price of health
care. I mean, health care is 20% of the
economy on its way to 50%. Um and and
so, the the the prices are exploding.
And and and so, it's like, okay, what
what makes health care what what makes
those three components of the American
dream have these exploding prices? It's
basically it's two things. Number one is
and these are interrelated, but number
one is technology is not really touching
them very much. The and you could say
the tech industry has done a bad job at
addressing housing and health care and
education or or you could say that, you
know, these industries have been
resistant to it.
But technology hasn't really touched
them that much, which means that they
productivity growth hasn't happened in
them.
And then the other thing is they're
heavily regulated. They're they're
heavily dominated by the government,
they're heavily regulated and then in
all three of those sectors,
the government both restricts supply,
right? And so zoning laws and university
certifications and so forth, right? Mean
that it's very hard to actually build
new of of those three things.
And then the government responds to the
political pressure caused by the rising
prices by subsidizing demand. And of
course, if you subsidize demand and a
fixed supply, you get skyrocketing
prices, right? And so every time there's
a housing crisis, you know, it's a
perpetual housing crisis in California,
the answer always is subsidies for home
buyers, which always has the effect of
driving prices up further and making the
problem worse.
Right. And so so so like that that that
drives, you know, that those are the
microeconomics of
of what's happening.
I don't think there's easy answers on
this. I but I think the acknowledgement
that this is in fact what's happening
and then I think the the people I think
the people have to decide that they do
not want to live this way.
I think like the ordinary the marginal
American voter needs to decide it's just
look it's just not acceptable that if I
can't spend $3 million I can't have a
house that's in a good neighborhood with
a good job attached to it. And it's just
not acceptable
that a four-year college degree is going
to cost you a million dollars, which is
which is what it's on its way to doing.
And it's just not acceptable that like
this health care, you know, monster is
running out of control.
And and so you know,
you know, in many ways like that this is
like the fundamental, you know, policy
political question of our time, you
know, that kind of kind of needs to be
dealt with. Again, these issues become
much easier to deal with in a
high-growth environment, right? And and
in an environment with very rapid
technological development.
Let me give you the good news story on
on health care. Everybody talks about
health care, you can't break these
curves. There's one area of health care
that is incredibly sophisticated,
technologically advanced, and the price
curve looks like Moore's law. The price
curve looks like video games. The price
curve is just a straight line down, and
the quality of the service has gone way
up. And And that's LASIK eye surgery.
Right? And And the reason it's And the
reason that's the case is cuz it's it's
outside the system. It's a discretionary
purchase paid for by the consumer. Um
you know, in the the the LASIK eye
studios are in like strip malls.
Um and they've got like, you know, three
people in like a super advanced, you
know, incredibly advanced like laser eye
machine. Um and you go in there and you
get zapped and you have perfect
eyesight. And it every year it gets
better and cheaper.
Um and so that's like an example of what
happens if you're able to touch these
sectors with technology and if you're
able to get them out from under all the
regulatory and government overhang. And
so that That potential exists, I think,
throughout the system. But But
fundamentally, we have to decide we want
it. And And fundamentally, the the
population of the country has to decide
that that it wants
>> Um my name is John Kidenda.
Um one of my favorite parts of the the
AI articles where you say something to
the effect of um the AI's are not going
to want to kill you because
they're not going to want.
And my question is, given that, do you
think that uh the whole conversation
around AI alignment is
is is net negative because it kind of
tends to push people towards
anthropomorphizing
at the technology and
um
creates this this sense that it's going
to have these wants that are similar to
the ones of a of a human being.
>> Um
the whole field of AI alignment and
safety starts with the assumption that
there's a problem. Right? And you get
you also see this in the academic world
cuz like the origin point of a lot of
these ideas at Oxford, I think it's
called the Center for Existential Risk.
Right? And And notice they didn't say
the they didn't say the Center for
Existential like, you know, benefits and
risk.
They [laughter] just said the Center for
Existential Risk. And so, you know,
these things are wired from the start.
You know, I have like nightmares that
there had been like a center of internet
risk in like 1993, which there wasn't.
Um but had there been like would we have
ever been allowed to even do what we
did? Like I don't know. You know,
imagine a center for microchip risk in
1960. Probably also a bad idea.
A center for automobile risk in 1930
would have been a bad idea. So so so
they're just they're sort of inherently
inclined to the negative. They're
they're sort of critical. You know, the
in thing in academia is critical
everything and they're sort of this is
sort of critical technology studies.
And so they're inclined to the negative.
And then the other thing that just
happens is there's a tip off.
There's a tip off for what's happening,
which is actually in the term AI
alignment. Which is the AI alignment is
not the original term that was actually
used for this whole field of things. The
original term was AI risk.
And or AI safety, right? And it was it
was literally it started with people who
were like literally worried about things
like killer robots and and so forth. So
it was like it was it was that real like
like actual risk to life. The the the
flip of the vocabulary was very telling.
So when it went from AI risk to AI
alignment, which was about 10 years ago,
it was very telling what was actually
happening there, which and and and the
reason is because with risk it's just
like, okay, risk, safe, unsafe. That's
like, okay, very clear. Alignment, it's
like, okay, alignment with what? Right?
And then what what they'll tell you is
that it's alignment with human values,
right? And then it's like, well,
alignment with whose human values?
Right? And basically what happened is
there were a set of people who basically
took what had been a literally life or
death existential question and turned it
into a morality and ethics question. And
of course why did they do that? You
know, in part is because they had
legitimate concerns about the future of
morals and ethics, but in part because
they had envisioned themselves as being
able to answer those questions. And so
and this is why in my essay I I started
with, you know, sort of risk number one,
which was existential risk, but then I
went to risk number two, which is sort
of societal risk, right? The risk of the
big fall societal harms. The harm word
is another one of these tip offs when
you know you're you're dealing with this
kind of thing. And and for that second
risk, expressed as AI alignment, what
you're dealing with fundamentally is
social engineering. And when you're
dealing with social engineering, you're
necessarily dealing with politics. And
so that that's the sort of collapse
that's happened kind of inside the AI
risk field. By the way, the original AI
risk people who are worried about the
killer robots, they actually are very
upset that this shift happens and they
think that their movement has been
hijacked by basically social activists,
which it has been.
Um and they now have proposed that the
the the name of the field should be
changed to something even more vividly
clear about what they're concerned
about. And the leading candidate is AI
not kill everyoneism.
>> [laughter]
>> Has not taken off yet, but you know,
give it time.