WEBVTT

00:00:00.000 --> 00:00:02.757
Elon Musk, thank you so much for joining
me on this side of it. Thank you for

00:00:02.840 --> 00:00:05.037
having us here at Tesla Giga.

00:00:05.120 --> 00:00:09.877
Um it is really hard to introduce you
without using superlatives. I mean, you

00:00:09.960 --> 00:00:13.717
are doing, I think, more to build
the future than literally anybody alive

00:00:13.800 --> 00:00:15.437
right now. >> I'm doing
a lot of building in general.

00:00:15.520 --> 00:00:17.797
I'm >> You're doing a lot of building.

00:00:17.880 --> 00:00:19.837
>> builder, I suppose.
I just like building things.

00:00:19.920 --> 00:00:23.957
>> You are, but you're building I mean, >>
You know, run around building things all

00:00:24.040 --> 00:00:26.037
day long. >> You run around
building a lot of things.

00:00:26.120 --> 00:00:29.757
You pioneered electric vehicles, you
revolutionized space travel, you've got

00:00:29.840 --> 00:00:32.717
a cutting-edge AI model now, humanoid

00:00:32.800 --> 00:00:37.237
robots. Can we start this conversation
by you putting all of this together?

00:00:37.320 --> 00:00:41.437
Tell me what the world's going to be like
in, let's say, 10 years if you succeed.

00:00:41.520 --> 00:00:43.357
>> 10 years, so we're talking 2036.

00:00:43.440 --> 00:00:46.437
>> Yep. Hopefully, we'll both be around.

00:00:46.520 --> 00:00:47.877
>> Hopefully, yes. >> can we can we

00:00:47.960 --> 00:00:51.117
>> Uh probably. Most likely, we will be.

00:00:51.200 --> 00:00:55.277
Um it it's hard to imagine
a future that where there's not

00:00:55.360 --> 00:00:59.077
an overall overwhelming amount
of artificial intelligence.

00:00:59.160 --> 00:01:03.477
Meaning, I can 10 years, I would
expect artificial intelligence to be

00:01:03.560 --> 00:01:06.277
far greater than the sum
of human intelligence.

00:01:06.360 --> 00:01:10.437
Like far greater. Um in fact, I think

00:01:10.520 --> 00:01:16.317
AI may exceed the sum of human
intelligence in about in around 5 years.

00:01:16.400 --> 00:01:19.477
>> In 5 years?
>> Roughly 5 years is my guess.

00:01:19.560 --> 00:01:23.117
Um so

00:01:23.200 --> 00:01:28.357
there really won't be anything that AI
can't do better than humans apart from

00:01:28.440 --> 00:01:34.760
being human, perhaps. >> At a more
prosaic level, what will life be like?

00:01:35.160 --> 00:01:37.477
>> The most likely outcome

00:01:37.560 --> 00:01:43.397
is an age of amazing abundance uh where

00:01:43.480 --> 00:01:46.637
anyone can have anything
they can think of.

00:01:46.720 --> 00:01:49.997
This is this is uh this may
sound preposterous, but

00:01:50.080 --> 00:01:54.597
well, here we are in 2026.
Let's see where we stand in 2036. Now,

00:01:54.680 --> 00:01:59.877
of course, there are things that could
uh derail this, like a a massive global

00:01:59.960 --> 00:02:01.997
thermonuclear war or something like that.

00:02:02.080 --> 00:02:06.077
Um but assuming that there's
not World War III,

00:02:06.160 --> 00:02:09.037
uh I think I think we're
headed for an age of

00:02:09.120 --> 00:02:12.917
amazing abundance. So,
this is, I guess, a message of

00:02:13.000 --> 00:02:17.917
of optimism and uh
excitement about the future.

00:02:18.000 --> 00:02:20.917
>> So, we'll talk about the abundance
in a second, but just on a very prosaic

00:02:21.000 --> 00:02:25.277
matter, how are your companies going to be
making money? Because when you In the

00:02:25.360 --> 00:02:29.357
In the SpaceX IPO prospectus, it it
looked as though the overwhelming amount

00:02:29.440 --> 00:02:33.397
of the revenue would be coming from AI,
from Grok, really. It won't be

00:02:33.480 --> 00:02:36.277
transporting things to space as SpaceX

00:02:36.360 --> 00:02:38.197
is now doing. What is What will your

00:02:38.280 --> 00:02:41.197
companies be doing in 10 years?

00:02:41.280 --> 00:02:44.160
And how will they make money?

00:02:44.840 --> 00:02:47.757
>> Well, you can really
think of the economy as

00:02:47.840 --> 00:02:50.717
uh digital and physical intelligence.

00:02:50.800 --> 00:02:55.277
Um so, what we have
right now advancing very

00:02:55.360 --> 00:03:00.637
rapidly is digital intelligence,
obviously, with with AI that is

00:03:00.720 --> 00:03:05.477
uh it seems that well, almost with every
passing week, uh there are are more AI

00:03:05.560 --> 00:03:09.917
breakthroughs. It It can do something
amazing that it couldn't do before.

00:03:10.000 --> 00:03:13.517
Uh but it's still somewhat
confined to the digital realm.

00:03:13.600 --> 00:03:16.517
It doesn't have end effectors.

00:03:16.600 --> 00:03:18.637
You need the end effectors in the form of

00:03:18.720 --> 00:03:21.277
humanoid robotics, um human In you need

00:03:21.360 --> 00:03:25.837
you need it basically a lots of bots
in a nutshell. You need lots of bots.

00:03:25.920 --> 00:03:29.637
Um and then you can go from um

00:03:29.720 --> 00:03:32.477
intelligence manifesting itself only

00:03:32.560 --> 00:03:37.917
digitally to shaping atoms. Um

00:03:38.000 --> 00:03:42.277
and you say, "What What What
is an economy?" An economy is um

00:03:42.360 --> 00:03:44.877
the production of goods and provision of

00:03:44.960 --> 00:03:49.637
services. And if you have vast numbers of

00:03:49.720 --> 00:03:54.597
robots uh with uh vast amounts
of digital intelligence,

00:03:54.680 --> 00:03:57.197
you have a sort of

00:03:57.280 --> 00:04:01.997
a quasi quasi-infinite economy.

00:04:02.080 --> 00:04:06.797
So so this is this is really
the the the future is going to be very

00:04:06.880 --> 00:04:11.757
very different from the past.
There's there's no um

00:04:11.840 --> 00:04:14.917
there's no analogy
or metaphor that I think

00:04:15.000 --> 00:04:18.557
illustrates the magnitude of change
that we're going to experience here.

00:04:18.640 --> 00:04:22.757
>> I I I understand that and I'm I'm also
trying to kind of figure out what that

00:04:22.840 --> 00:04:25.637
world would look like
but but in some way you have

00:04:25.720 --> 00:04:28.917
>> It's going to be a pretty incredible >>
Yeah, well we'll also do there were some

00:04:29.000 --> 00:04:34.317
risks too. We'll come to those. We'll
come to those but just on on on how then

00:04:34.400 --> 00:04:37.197
you make money cuz in the end people are
buying your shares cuz cuz they expect

00:04:37.280 --> 00:04:39.677
you to make money. >> I'll I'll
I'll make another prediction.

00:04:39.760 --> 00:04:44.077
Money won't matter in 2036.
>> You're watching my interview with Elon

00:04:44.160 --> 00:04:48.357
Musk. If you want to get early access
to other interviews, subscribe to The

00:04:48.440 --> 00:04:50.717
Economist by clicking the link above.

00:04:50.800 --> 00:04:54.557
Now, back to the interview. >> [laughter]
>> I'm not sure that the people who bought

00:04:54.640 --> 00:04:58.517
your shares think that money
won't matter in 2036.

00:04:58.600 --> 00:05:01.357
>> what do you want money for?

00:05:01.440 --> 00:05:03.837
Are you what you want money for?
Goods and services.

00:05:03.920 --> 00:05:08.357
>> Yeah. >> Right? You want money
for obviously food, housing, transport,

00:05:08.440 --> 00:05:10.717
entertainment. Um

00:05:10.800 --> 00:05:13.877
well, if that is so abundant

00:05:13.960 --> 00:05:18.597
that uh there's more that the
robots and AI are providing

00:05:18.680 --> 00:05:23.397
more goods and services than you
than any human could possibly consume,

00:05:23.480 --> 00:05:25.677
what do you need money for in that case?

00:05:25.760 --> 00:05:27.757
>> But you're you're
creating these robots.

00:05:27.840 --> 00:05:29.837
Are you then selling >>
then you'll need money, sure.

00:05:29.920 --> 00:05:31.757
>> Are you are you selling these robots?

00:05:31.840 --> 00:05:36.157
Are you selling human humanoid robots
or are you producing goods and services

00:05:36.240 --> 00:05:40.200
with the humanoid robots that you control?

00:05:40.440 --> 00:05:43.117
I give you the leave I struck so >> I
think I think I think I would probably

00:05:43.200 --> 00:05:48.717
be And would be an exercise in vanity
to for me to think that I would be

00:05:48.800 --> 00:05:52.637
controlling a super genius
AI that is vastly smarter

00:05:52.720 --> 00:05:56.437
than me. I think what we
can and should try to do

00:05:56.520 --> 00:05:58.237
is to make sure that the AI

00:05:58.320 --> 00:06:02.037
has good values, that it cares about

00:06:02.120 --> 00:06:07.517
humanity, and that uh it
wants wants us to be happy and

00:06:07.600 --> 00:06:10.237
prosper. Um

00:06:10.320 --> 00:06:14.757
I think it probably will be. Um My

00:06:14.840 --> 00:06:18.597
what my biological neural net tells me
is that the most important thing for the

00:06:18.680 --> 00:06:20.397
for AI safety is to be maximally truth

00:06:20.480 --> 00:06:22.837
seeking and curious.

00:06:22.920 --> 00:06:25.797
Uh if that's the case, uh I think it

00:06:25.880 --> 00:06:30.357
will foster humanity. >> So,
let's unpack that. You're sure that

00:06:30.440 --> 00:06:33.957
humans will no longer be
in control in 10 years' time.

00:06:34.040 --> 00:06:36.677
>> I think it is unlikely. Um in the same

00:06:36.760 --> 00:06:42.317
way that if the difference
in intelligence between AI and humans is

00:06:42.400 --> 00:06:46.757
vastly greater than the difference in
intelligence between AI and chimpanzees,

00:06:46.840 --> 00:06:48.477
it's hard to imagine

00:06:48.560 --> 00:06:51.237
that the chimpanzees would be in charge.

00:06:51.320 --> 00:06:53.797
>> I'm really struck
by what you're saying.

00:06:53.880 --> 00:06:57.957
>> um an evolved form
of chimpanzee. So, it

00:06:58.040 --> 00:07:01.077
wasn't that long ago that we were
swinging through the trees, eating

00:07:01.160 --> 00:07:05.037
bananas, um in the jungle.
And we still eat

00:07:05.120 --> 00:07:08.557
bananas, by the way. We do.
And we still, you know, bananas

00:07:08.640 --> 00:07:10.197
>> sometimes swing through trees as well.

00:07:10.280 --> 00:07:11.477
>> Yeah, sometimes swing through trees.

00:07:11.560 --> 00:07:15.797
It's kind of fun. It's sort of um
reminiscent of the past. Um for for old

00:07:15.880 --> 00:07:18.637
time's sake, swing through the trees
once in a while. Probably, you know,

00:07:18.720 --> 00:07:21.397
there should be more swinging through
trees. I bet it's way more fun than

00:07:21.480 --> 00:07:22.707
people think. >> I'm with you on that.

00:07:22.790 --> 00:07:24.037
>> [laughter] >> I'm
definitely with you on swinging

00:07:24.120 --> 00:07:27.757
through trees. I love swinging through
trees. But But But I'm really struck

00:07:27.840 --> 00:07:30.197
listening to you say this now, cuz I've
been listening to a bunch of things that

00:07:30.280 --> 00:07:33.757
you've said over the years. And And 10
years ago, you were on a podcast, I

00:07:33.840 --> 00:07:36.677
think it was with Neil deGrasse Tyson,
and you said that the thing that

00:07:36.760 --> 00:07:41.637
terrified you most was rapid
recursive of in AIs.

00:07:41.720 --> 00:07:44.877
>> Yes. that you said And it really stuck
with me. You said that we will just be

00:07:44.960 --> 00:07:47.037
the pet Labradors at the best.

00:07:47.120 --> 00:07:49.917
>> Well, >> Now you see >>
Labradors are pretty happy creatures.

00:07:50.000 --> 00:07:53.477
That's my experience.
>> in 2023 you you called you asked for a

00:07:53.560 --> 00:07:55.517
slow down in AI cuz you
were worried about it.

00:07:55.600 --> 00:08:00.037
>> Well, I signed on to, you know, on
as one of many like 500 people. I I I was

00:08:00.120 --> 00:08:04.637
just, you know, one of many
signa- signatories on that thing.

00:08:04.720 --> 00:08:06.557
>> But but I think
even last year you said,

00:08:06.640 --> 00:08:09.677
I don't know where was it, Ted Cruz's
podcast, you said there was a 10 to 20%

00:08:09.760 --> 00:08:12.397
chance of killer robots
wiping out humanity.

00:08:12.480 --> 00:08:15.197
>> Yes. >> So you sounded
worried then. Now you

00:08:15.280 --> 00:08:19.477
seem very relaxed about it.
Have you changed your view?

00:08:19.560 --> 00:08:23.117
>> I still think there's I
still think there's a risk

00:08:23.200 --> 00:08:24.797
associated with AI and robots. It's not

00:08:24.880 --> 00:08:26.117
zero.

00:08:26.200 --> 00:08:30.317
Um I I guess I've come to

00:08:30.400 --> 00:08:35.597
my my I sort of philosophical
conclusion is to look on the bright side.

00:08:35.680 --> 00:08:39.117
Um I like I I can't see any way to really

00:08:39.200 --> 00:08:44.197
stop this incredible
momentum of AI and robots.

00:08:44.280 --> 00:08:47.437
Um and and at times I sort of think, well,

00:08:47.520 --> 00:08:51.437
perhaps, you know, we're even if there
was a stop button, we probably shouldn't

00:08:51.520 --> 00:08:55.477
press it uh because

00:08:55.560 --> 00:09:00.597
I cuz the most likely
outcome is incredible

00:09:00.680 --> 00:09:04.357
abundance for for all.
>> That's what I thought actually. It

00:09:04.440 --> 00:09:08.797
struck me that now you seem
to have decided it's inevitable.

00:09:08.880 --> 00:09:12.437
Let's hope for the best
and go for the ride.

00:09:12.520 --> 00:09:15.957
>> Yeah. I mean, yes, pretty much.

00:09:16.040 --> 00:09:21.037
Let's uh let's enjoy the ride as as I
think my philosophy at this point cuz

00:09:21.120 --> 00:09:26.237
I mean, honestly, this this it's
it's there seems to be just inexorable

00:09:26.320 --> 00:09:31.317
uh progress here
in in AI and and in robots

00:09:31.400 --> 00:09:35.917
that um >> But you >> You know,
even if I wanted to stop it

00:09:36.000 --> 00:09:39.477
and I I couldn't. Um >> Do you
think there's still a 10 to 20%

00:09:39.560 --> 00:09:42.237
chance that killer robots mean the end

00:09:42.320 --> 00:09:44.837
of humanity?

00:09:44.920 --> 00:09:47.717
That's a high P doom, as they say.

00:09:47.800 --> 00:09:51.557
>> Well, in long run
we're all dead, of course.

00:09:51.640 --> 00:09:56.517
Uh you know, um So I mean
I I so the it let let's

00:09:56.600 --> 00:10:01.837
according to the standard model
of physics, the end of the universe is

00:10:01.920 --> 00:10:04.837
um a long, boring

00:10:04.920 --> 00:10:08.637
sort of thin, gray soup of

00:10:08.720 --> 00:10:13.197
uh gradual cooling. Um I mean it's sort of

00:10:13.280 --> 00:10:16.997
the heat death of the universe it's
called. So now if if the heat death of

00:10:17.080 --> 00:10:21.717
the universe is in fact the outcome for
our reality, then it really is all about

00:10:21.800 --> 00:10:24.677
the journey.
Because the destination is terrible.

00:10:24.760 --> 00:10:27.477
>> 10 years is not very [laughter] Yeah,
I get that, but that could be a

00:10:27.560 --> 00:10:31.917
long way ahead. And and no basis very
very long way. If it's 10 to 20% in 10

00:10:32.000 --> 00:10:34.677
years, you know, would you go into one
of your rockets if you thought there was

00:10:34.760 --> 00:10:38.557
a 10 to 20% chance of it
blowing up and killing you?

00:10:38.640 --> 00:10:44.517
>> Yes, but but let's say there's
you can't do anything about it.

00:10:44.600 --> 00:10:47.037
>> Okay. >> The I guess
the only thing we can try to

00:10:47.120 --> 00:10:50.997
do is minimize the probability
of the bad outcome.

00:10:51.080 --> 00:10:54.397
Um but trying to stop the intelligence in

00:10:54.480 --> 00:10:56.237
robots

00:10:56.320 --> 00:10:58.517
um

00:10:58.600 --> 00:11:01.997
I mean for the longest time I

00:11:02.080 --> 00:11:05.917
declined to participate
in AI or was you know

00:11:06.000 --> 00:11:08.757
you know, created OpenAI as a

00:11:08.840 --> 00:11:13.277
essentially counterweight to to Google
cuz at the time they were um had more or

00:11:13.360 --> 00:11:18.677
less a monopoly on AI.
Um and then Anthropic spun out of uh

00:11:18.760 --> 00:11:22.757
OpenAI. And and if anything,
it seems like what

00:11:22.840 --> 00:11:26.477
my actions have really accelerated AI.
You know, I wasn't

00:11:26.560 --> 00:11:31.197
really aiming to accelerate AI,
but in creating OpenAI and then having

00:11:31.280 --> 00:11:35.597
Anthropic spun out of that and now
currently Anthropic's leader in AI.

00:11:35.680 --> 00:11:38.837
Um I mean these

00:11:38.920 --> 00:11:42.037
these actions have actually resulted in

00:11:42.120 --> 00:11:45.117
knock-on effects that accelerated AI

00:11:45.200 --> 00:11:47.717
which wasn't really my intention.

00:11:47.800 --> 00:11:51.077
So it just seems like

00:11:51.160 --> 00:11:55.837
all roads lead to acceleration of AI.

00:11:55.920 --> 00:12:01.197
So then I'm like, "Okay. Well,
you can just sort of be sad about it or

00:12:01.280 --> 00:12:03.757
join the club, I suppose." >> Well,
you've certainly joined the club.

00:12:03.840 --> 00:12:06.677
>> Yeah. >> I mean you've you've
definitely joined the club, but

00:12:06.760 --> 00:12:07.797
>> You can't beat them,
join them, you know?

00:12:07.880 --> 00:12:10.877
>> Well, you've you've you've clearly
done that and very successfully, but what

00:12:10.960 --> 00:12:14.477
about trying to minimize the risks? And

00:12:14.560 --> 00:12:17.477
you know, a number of the other leaders

00:12:17.560 --> 00:12:20.517
of of model makers have been pushing for

00:12:20.600 --> 00:12:24.517
different ways to try and minimize the
risks. Um you know, just last week Demis

00:12:24.600 --> 00:12:29.717
Hassabis came out with a proposal
for a private public regulator.

00:12:29.800 --> 00:12:31.877
Um >> Yeah. >> you you
said it was a good starting

00:12:31.960 --> 00:12:35.677
point for discussions, I think. Do you
think you guys should be working more

00:12:35.760 --> 00:12:39.637
actively to get something
in place to control the

00:12:39.720 --> 00:12:41.717
risks or do you think
that's impossible, too?

00:12:41.800 --> 00:12:44.757
>> My recommendation to Demis,
cuz we I talked to him for

00:12:44.840 --> 00:12:49.277
a few hours before he published
his his piece, was that um

00:12:49.360 --> 00:12:53.157
the most immediate thing that we could
do is to have the leading AI companies

00:12:53.240 --> 00:12:57.957
um at least just meet uh
or have a some sort of call

00:12:58.040 --> 00:13:02.397
once every few weeks um and just

00:13:02.480 --> 00:13:04.357
uh discuss

00:13:04.440 --> 00:13:07.597
se- any safety and security issues

00:13:07.680 --> 00:13:11.637
that and and and that before
a breakthrough uh sort of a a new

00:13:11.720 --> 00:13:14.517
frontier model is released,
something that sort of

00:13:14.600 --> 00:13:18.517
is is far more intelligent than anything
that already exists, that there's an

00:13:18.600 --> 00:13:21.157
opportunity for the

00:13:21.240 --> 00:13:24.997
various competing uh AI companies to

00:13:25.080 --> 00:13:28.477
uh test test that model for

00:13:28.560 --> 00:13:30.197
any harmful effects.

00:13:30.280 --> 00:13:33.517
And to be able to recommend pausing to

00:13:33.600 --> 00:13:36.717
address some of these security issues.

00:13:36.800 --> 00:13:39.677
I think I think that's that's the that's
sort of the the easy quick thing that we

00:13:39.760 --> 00:13:45.077
can do. And it can it can evolve to more
than that over time. Um but that's um

00:13:45.160 --> 00:13:46.517
that I think would be a smart move.

00:13:46.600 --> 00:13:49.597
>> But why why haven't you done that yet?
I mean there's like five of you, right? I

00:13:49.680 --> 00:13:50.997
mean everyone knows the first name.

00:13:51.080 --> 00:13:54.837
Dario, Demis, Sam. I mean you don't all
like each other very much. Yeah, but you

00:13:54.920 --> 00:13:58.797
know you're not all I mean you're also
all competing, right? You all want to be

00:13:58.880 --> 00:14:02.877
first. >> but that's
that's where I think um the

00:14:02.960 --> 00:14:06.877
the competitors can keep keep
each other honest. So I mean

00:14:06.960 --> 00:14:10.877
meaning like if um like I think it's
quite difficult for someone in the

00:14:10.960 --> 00:14:13.837
government who doesn't have a deep
technical understanding and isn't

00:14:13.920 --> 00:14:18.717
driving the frontier of of AI to know
whether something should be released or

00:14:18.800 --> 00:14:19.797
not.

00:14:19.880 --> 00:14:23.797
Um however, the competitors I think what

00:14:23.880 --> 00:14:26.157
can you know if given sort of a week or

00:14:26.240 --> 00:14:31.597
two to review a new model um can
highlight issues and uh and have an

00:14:31.680 --> 00:14:34.197
incentive to keep all all
the competitors have an

00:14:34.280 --> 00:14:35.597
incentive to keep the others honest.

00:14:35.680 --> 00:14:38.437
>> Do you have enough trust between you to
do that? I mean you don't much like each

00:14:38.520 --> 00:14:41.597
other. Wouldn't wouldn't Sam Altman just
think you're going to nick all his ideas

00:14:41.680 --> 00:14:46.437
if you did that? >> Well, we're
just talking about a Honestly, I I

00:14:46.520 --> 00:14:50.397
I think as little as a week
is all that would be needed uh

00:14:50.480 --> 00:14:54.157
for private testing. It's it's not
to it's not to hand over the AI. It's just

00:14:54.240 --> 00:14:56.117
to test it. >> Have you proposed this?

00:14:56.200 --> 00:14:59.637
>> Yeah, just to test it
via an API or something.

00:14:59.720 --> 00:15:03.237
Um I think that would probably
be a a good thing to do. Um

00:15:03.320 --> 00:15:06.757
>> But have you have you proposed it?
I mean you could presumably organize that.

00:15:06.840 --> 00:15:09.317
>> to Demis. And I guess uh

00:15:09.400 --> 00:15:12.077
I'll I'll talk to Dario about it, too.

00:15:12.160 --> 00:15:14.357
Um

00:15:14.440 --> 00:15:17.997
There's also, of course, um some very

00:15:18.080 --> 00:15:22.757
competent uh Chinese AI companies um

00:15:22.840 --> 00:15:26.477
that uh there's several of them
several of that are that are very good.

00:15:26.560 --> 00:15:29.637
Um the new Kimmy model >> Kimmy
K3, yeah, the one that came out

00:15:29.720 --> 00:15:34.197
last week is I mean it's almost as good
as Fable, right? So, it's >> Well,

00:15:34.280 --> 00:15:37.517
um I I think realistically Fable is is

00:15:37.600 --> 00:15:41.477
still uh clearly the smartest model. Um

00:15:41.560 --> 00:15:45.157
I mean anyone realistically would
say that that's still the case. But,

00:15:45.240 --> 00:15:49.637
the Kimmy model is getting
quite close to Fable.

00:15:49.720 --> 00:15:55.037
Um Now, Anthropic had uh
Mythos from which Fable is

00:15:55.120 --> 00:15:59.957
derived. I think they had
that ready in February.

00:16:00.040 --> 00:16:02.917
So, they've got they've
got for it what they

00:16:03.000 --> 00:16:08.437
call I don't know what they're called,
but Mythos 2 or or Fable 6 or something

00:16:08.520 --> 00:16:10.997
something much better than what they had
in February. They are almost certainly

00:16:11.080 --> 00:16:15.717
have and we're not almost certainly,
they certainly have right now. So,

00:16:15.800 --> 00:16:19.997
I and I think they could they
they could release it at any time.

00:16:20.080 --> 00:16:23.957
>> Just before we get on to to releasing,
how do you think about the Chinese

00:16:24.040 --> 00:16:27.917
models and the geopolitics of it?
Are you worried about the Chinese catching

00:16:28.000 --> 00:16:31.517
up and getting ahead?
>> Given that the Chinese

00:16:31.600 --> 00:16:34.277
AI companies are um

00:16:34.360 --> 00:16:36.837
are doing as well as they are with

00:16:36.920 --> 00:16:41.957
a relatively small amount of compute,
uh it seems though as that if they had a

00:16:42.040 --> 00:16:46.637
lot of compute, um there's
a good chance that they would be

00:16:46.720 --> 00:16:53.560
the leaders. And at some point,
they probably will have a lot of compute.

00:16:53.760 --> 00:16:57.600
>> So, they will ergo
they will be the leaders.

00:16:59.280 --> 00:17:03.157
>> There's a good chance that they
are the leaders at some point. Um

00:17:03.240 --> 00:17:07.397
You can think of like
the the constraints on AI are

00:17:07.480 --> 00:17:10.637
essentially electricity and

00:17:10.720 --> 00:17:12.437
AI chips.

00:17:12.520 --> 00:17:15.597
Um and then for physical AI, the robots.

00:17:15.680 --> 00:17:19.637
Uh China's also got some very good
robot companies. In fact, they have

00:17:19.720 --> 00:17:22.597
these like robot sporting events.
I don't know if you've seen some of them.

00:17:22.680 --> 00:17:26.317
>> Yes, I have. I saw them the last time
I was in Beijing. I saw the the robots.

00:17:26.400 --> 00:17:29.357
>> Well, and then they have this sort of
like a like a boxing match or a fighting

00:17:29.440 --> 00:17:32.677
match with um the robots,
which is quite entertaining.

00:17:32.760 --> 00:17:36.157
And I saw I saw some video of robot that
got its head knocked off, kept fighting

00:17:36.240 --> 00:17:37.997
even with its head knocked off.

00:17:38.080 --> 00:17:42.157
Uh, it looked pretty funny, but um I mean
in fact, I think the the robot battles

00:17:42.240 --> 00:17:44.797
are going to be quite
entertaining to watch.

00:17:44.880 --> 00:17:46.757
Um >> But but just to be

00:17:46.840 --> 00:17:48.637
>> so China's China's very strong on

00:17:48.720 --> 00:17:53.397
robots, very strong on digital AI, and

00:17:53.480 --> 00:17:56.917
and if you look at electricity
production, you could say like AI is a

00:17:57.000 --> 00:18:00.637
function of whatever what what whatever
the limiting factor is. If it's chips or

00:18:00.720 --> 00:18:04.717
electricity, China has far more
electricity than the United States. In

00:18:04.800 --> 00:18:07.997
fact, China has more electricity
than the United States,

00:18:08.080 --> 00:18:10.837
Europe, and India combined already.

00:18:10.920 --> 00:18:13.157
Um and is heading to

00:18:13.240 --> 00:18:18.717
And my guess is China gets to four times
the electricity production of the US,

00:18:18.800 --> 00:18:21.597
which is roughly proportionate
to the population.

00:18:21.680 --> 00:18:25.637
>> What should the US government do about
this? I mean, if you you presumably it

00:18:25.720 --> 00:18:29.237
is a little alarming if you have
an authoritarian regime far ahead in the

00:18:29.320 --> 00:18:33.037
most powerful technology. And clearly
what what the US government's been

00:18:33.120 --> 00:18:34.917
trying to do is with the export controls

00:18:35.000 --> 00:18:37.757
and so forth, keep China back. Do you

00:18:37.840 --> 00:18:42.557
think they should ban the use
of K3 Kimmy K3, for example?

00:18:42.640 --> 00:18:45.717
>> Well, the US government
can't they they can

00:18:45.800 --> 00:18:50.797
they can pass some law that prevents
US companies from using um

00:18:50.880 --> 00:18:54.397
Chinese models, but they can't stop the
rest of the world from from using them.

00:18:54.480 --> 00:18:58.837
Um So, it's it's not like the the US
government has authority over

00:18:58.920 --> 00:19:02.357
China or the rest of the world.
Um >> Do you think they should stop US

00:19:02.440 --> 00:19:06.317
companies from using them?
There's talk of that right now, right?

00:19:06.400 --> 00:19:08.957
>> I don't think that that helps.

00:19:09.040 --> 00:19:13.557
Um it that won't stop China
from being the leader in AI. The

00:19:13.640 --> 00:19:16.517
the really it's
for for digital intelligence

00:19:16.600 --> 00:19:21.797
it comes down to uh how much
AI compute do you have? It

00:19:21.880 --> 00:19:25.437
Well, how many chips chips do
you have and how much power do

00:19:25.520 --> 00:19:27.997
you have? How much
electricity do you have?

00:19:28.080 --> 00:19:31.397
Um so

00:19:31.480 --> 00:19:33.037
the constraint right now I think is

00:19:33.120 --> 00:19:36.317
actually slightly more on

00:19:36.400 --> 00:19:40.157
power and cooling than it
is on AI chips because um

00:19:40.240 --> 00:19:44.717
the the power demands of the AI chips
are very very high. Um we'll talk about

00:19:44.800 --> 00:19:49.357
water demands but the really the water
usage of of AI is negligible or almost

00:19:49.440 --> 00:19:53.637
nothing. Um but the electric
the electrical demand is very high.

00:19:53.720 --> 00:19:56.677
>> That's why you're thinking of orbital
data centers presumably. That's

00:19:56.760 --> 00:20:00.277
>> Yes, the constraint for any AI chips

00:20:00.360 --> 00:20:05.237
not being turned on in China
is electricity and like power and cooling

00:20:05.320 --> 00:20:07.477
and the electrical
equipment and everything.

00:20:07.560 --> 00:20:12.997
Um the rate at which AI chips are
being made exceeds the rate at which new

00:20:13.080 --> 00:20:15.557
electricity is coming online.
>> In China. Okay.

00:20:15.640 --> 00:20:18.397
>> Um out outside of China.
>> Outside China, I see. Yeah.

00:20:18.480 --> 00:20:20.837
>> Um and since China

00:20:20.920 --> 00:20:25.757
you know the the US has banned
the export of the latest AI chips to China

00:20:25.840 --> 00:20:29.557
um China is is somewhat
chip starved. So so

00:20:29.640 --> 00:20:33.357
>> are becoming much more efficient
at using the chips that they have and that

00:20:33.440 --> 00:20:36.957
that one of the striking things of Kimik
3, right, is how efficiently they

00:20:37.040 --> 00:20:39.957
>> And and and China is also um

00:20:40.040 --> 00:20:44.157
I I think closer than
most people realize to

00:20:44.240 --> 00:20:48.357
solving uh the lithography problem

00:20:48.440 --> 00:20:50.677
which would enable them to make uh AI

00:20:50.760 --> 00:20:53.917
chips at at very large volumes.

00:20:54.000 --> 00:20:57.837
So so I think this

00:20:57.920 --> 00:21:02.877
you know, this isn't evolving into
just a China conversation but um

00:21:02.960 --> 00:21:05.637
but but the so the current current
constraint is is electricity outside of

00:21:05.720 --> 00:21:08.477
China. In China, the constraint is chips.

00:21:08.560 --> 00:21:12.597
Uh once the once we address the

00:21:12.680 --> 00:21:16.997
the power constraint with um
AI data centers in space,

00:21:17.080 --> 00:21:19.277
uh then the constraint will once again

00:21:19.360 --> 00:21:24.277
be chips outside of China.
>> That suggests to me that you need China

00:21:24.360 --> 00:21:27.797
and the US if you're going to have some
overall safety constraints that they

00:21:27.880 --> 00:21:32.517
will need to work together. If you are
going to try and I mean you suggested

00:21:32.600 --> 00:21:35.557
you know, the five >> Well,
I think it should include Uh

00:21:35.640 --> 00:21:38.077
probably it should include um

00:21:38.160 --> 00:21:43.837
the Chinese companies that are
doing frontier uh AI models, too.

00:21:43.920 --> 00:21:46.197
>> And do you think
the governments need to be involved?

00:21:46.280 --> 00:21:50.237
>> My recommendation would be that if

00:21:50.320 --> 00:21:54.477
if some of the leading
companies in AI in AI say that

00:21:54.560 --> 00:21:58.797
this a new model is very dangerous,

00:21:58.880 --> 00:22:02.157
then they should inform the government
and and and and well if a new model is

00:22:02.240 --> 00:22:04.957
very dangerous and that the company

00:22:05.040 --> 00:22:09.637
producing that model is not
taking steps to address the danger,

00:22:09.720 --> 00:22:14.117
then then they should alert
the government to take action um

00:22:14.200 --> 00:22:18.437
to limit the uh

00:22:18.520 --> 00:22:21.317
release of that model.
Uh and that could be action by the

00:22:21.400 --> 00:22:25.437
Chinese government or action by the US
government. Um it's it's really it's

00:22:25.520 --> 00:22:28.557
really just the US government and the
Chinese government that that have the

00:22:28.640 --> 00:22:29.997
power to take action here.

00:22:30.080 --> 00:22:32.437
>> But the I guess people would be worried

00:22:32.520 --> 00:22:38.237
about relying on a group of model
builders to decide whether to do this. I

00:22:38.320 --> 00:22:41.357
mean if you look at what the US
government has done with with Mythos,

00:22:41.440 --> 00:22:45.357
they stepped in because they were
worried about the cybersecurity

00:22:45.440 --> 00:22:50.277
risks and they stepped in using you know,
threat of export controls to you know,

00:22:50.360 --> 00:22:53.837
prevent the widespread release
of this model. I find it hard to see how

00:22:53.920 --> 00:22:59.597
governments will not want to keep
that power that they ultimately decide

00:22:59.680 --> 00:23:00.597
whether or not it gets released.

00:23:00.680 --> 00:23:03.717
>> I think they should keep they should
have that power, but it it wasn't as

00:23:03.800 --> 00:23:05.757
though someone in the government figured

00:23:05.840 --> 00:23:08.397
out that Methos uh

00:23:08.480 --> 00:23:13.237
was had had some scary cybersecurity

00:23:13.320 --> 00:23:14.957
uh

00:23:15.040 --> 00:23:17.397
risks. It it was Amazon.

00:23:17.480 --> 00:23:22.237
Um Amazon highlighted this and
Amazon called the White House.

00:23:22.320 --> 00:23:25.917
>> So, you'd have a system where the model
makers would check each other's work. If

00:23:26.000 --> 00:23:28.277
there was something that worried them,
you'd tell the government. The

00:23:28.360 --> 00:23:31.037
government would then think, "Okay."

00:23:31.120 --> 00:23:35.357
>> and that company was not doing anything
to address that risk, then that would be

00:23:35.440 --> 00:23:39.917
the the moment for government to step in
and take action, which is literally what

00:23:40.000 --> 00:23:41.477
just what occurred with Methos.

00:23:41.560 --> 00:23:44.360
>> have to happen?

00:23:44.720 --> 00:23:46.717
>> I think it'd be good to do that.

00:23:46.800 --> 00:23:50.157
I think immediately, really. Yeah.

00:23:50.240 --> 00:23:52.957
>> Yeah, that's what people keep telling
me. We we have to get this some kind of

00:23:53.040 --> 00:23:56.557
system in the next 6 months because
of the pace at which the models are

00:23:56.640 --> 00:24:00.917
improving. >> Yeah,
6 months is a long time.

00:24:01.000 --> 00:24:05.157
Um >> So, you think shorter than that?

00:24:05.240 --> 00:24:08.477
>> Well, I I mean,
there's there's so many AI

00:24:08.560 --> 00:24:12.597
breakthrough announcements,
sometimes multiple per day.

00:24:12.680 --> 00:24:14.197
I mean, how many were there last week?

00:24:14.280 --> 00:24:15.837
>> A lot. >> I lost track, you know.

00:24:15.920 --> 00:24:19.957
>> Yeah, a lot. >> Um

00:24:20.040 --> 00:24:23.597
>> So, so you you basically, you know,
whether it's Demis' proposal or others,

00:24:23.680 --> 00:24:26.517
you need to get together and get
this set up kind of now, it sounds like

00:24:26.600 --> 00:24:27.877
you're telling me.

00:24:27.960 --> 00:24:30.797
>> Yeah. I I I would recommend that we at

00:24:30.880 --> 00:24:35.397
least have some sort of informal
um weekly or bi-weekly

00:24:35.480 --> 00:24:40.117
um call and that there's that that

00:24:40.200 --> 00:24:41.797
um there's maybe

00:24:41.880 --> 00:24:45.357
a week or 2 weeks of of early access by

00:24:45.440 --> 00:24:49.757
competitors. Um and this is why I think
the incentives work is because if if,

00:24:49.840 --> 00:24:53.117
you know, competing companies
I think are likely to

00:24:53.200 --> 00:24:56.997
understand if something is a security
risk and they're they're not going to be

00:24:57.080 --> 00:25:01.117
shy about highlighting that that their
their competitors' models should be

00:25:01.200 --> 00:25:04.117
delayed in release. >> Is that >>
That that's the I think the structure

00:25:04.200 --> 00:25:07.877
does work quite well there.
And And if the if And if

00:25:07.960 --> 00:25:10.837
this is sort of how it works with the
Motion Picture Association, you know,

00:25:10.920 --> 00:25:14.917
the it's it's an industry group that uh

00:25:15.000 --> 00:25:18.317
decides how how movies
should be rated. You know,

00:25:18.400 --> 00:25:22.717
whe- whether it's appropriate for all
for young children or or not. Um

00:25:22.800 --> 00:25:26.437
and that's uh that seems
to have worked quite well.

00:25:26.520 --> 00:25:30.277
So, um And And And And
then you really just ask

00:25:30.360 --> 00:25:33.757
have ask the government to intervene if
if some if a company's doing something

00:25:33.840 --> 00:25:38.317
very dangerous and refuses to take
action to reduce the danger.

00:25:38.400 --> 00:25:42.037
>> Do you think this is a
This is a tricky one, but

00:25:42.120 --> 00:25:45.237
you guys don't much like each other and
you don't much trust each other. You're

00:25:45.320 --> 00:25:49.517
insulting each other on on social media
constantly. So, do you Do you think you

00:25:49.600 --> 00:25:53.077
can do that with people that you dislike
that that where there seems to be such

00:25:53.160 --> 00:25:57.157
kind of personal vitriol
between some of these leaders?

00:25:57.240 --> 00:26:00.557
>> Um Yes, well,
I'm not I'm not a fan of Sam

00:26:00.640 --> 00:26:04.637
Altman because um >> You're
watching my interview with Elon

00:26:04.720 --> 00:26:08.957
Musk. If you want to get early access
to other interviews, subscribe to The

00:26:09.040 --> 00:26:11.237
Economist by clicking the link above.

00:26:11.320 --> 00:26:15.397
Now, back to the interview.
>> If you started a nonprofit that was

00:26:15.480 --> 00:26:21.557
meant to be an open-source
AI company and um

00:26:21.640 --> 00:26:25.437
you know, owned by the world uh
and it somehow got turned into

00:26:25.520 --> 00:26:29.517
an $800 billion for-profit company
with closed source, I think you'd be

00:26:29.600 --> 00:26:33.597
like, "Well, wait a second. That's uh
the exact opposite of what I donated the

00:26:33.680 --> 00:26:37.917
money for." That That's my issue.
I think it's a legitimate one.

00:26:38.000 --> 00:26:40.997
Um And then Dario, I think, also is not a

00:26:41.080 --> 00:26:45.677
fan of Sam Altman because he felt Uh in
fact, the reason the Anthropic team left

00:26:45.760 --> 00:26:48.517
OpenAI is because they they
didn't trust Sam Altman.

00:26:48.600 --> 00:26:52.917
Otherwise, Anthropic wouldn't exist.
They would still be at Open AI.

00:26:53.000 --> 00:26:57.037
Um I I think Dario is very principled

00:26:57.120 --> 00:26:59.677
person. He cares about things a lot.

00:26:59.760 --> 00:27:03.677
He cares about the future of of the world
and I I think everyone I've met at

00:27:03.760 --> 00:27:07.877
Anthropic so far has been I
mean like no one set off my evil

00:27:07.960 --> 00:27:12.637
detector. You know, they always seem
seem like they're well-intentioned. Um

00:27:12.720 --> 00:27:16.117
and the road to hell
is I think mostly paved

00:27:16.200 --> 00:27:18.117
with bad intentions. There are a few

00:27:18.200 --> 00:27:20.117
well-intentioned paving stones in there.

00:27:20.200 --> 00:27:24.757
So we we don't want
to be complacent. Um but

00:27:24.840 --> 00:27:27.437
at the end of the day if we
have to talk, we'll talk. I mean

00:27:27.520 --> 00:27:30.717
set aside our personal differences for the
good of the world at the end of the day.

00:27:30.800 --> 00:27:33.517
>> Well, that's that's a very good
thing for the rest of us to hear.

00:27:33.600 --> 00:27:37.717
>> Yeah. Um >> Can I Can we
turn to one other risk that

00:27:37.800 --> 00:27:41.597
people worry about? Particularly I think
ordinary Americans where you know, AI

00:27:41.680 --> 00:27:45.677
has become incredibly unpopular. 75% of
people are worried about it. And I think

00:27:45.760 --> 00:27:47.397
they worry about their jobs.

00:27:47.480 --> 00:27:50.597
And if you look at Dario has said

00:27:50.680 --> 00:27:53.517
famously what? 50% of entry-level jobs

00:27:53.600 --> 00:27:57.757
entry-level white-collar jobs will be
gone within a few years. You're building

00:27:57.840 --> 00:28:01.557
humanoid robots. >> Yes,
he he also said like I mean Dario

00:28:01.640 --> 00:28:05.197
did dig his own grave here regarding
Methuselahs because he said,

00:28:05.280 --> 00:28:09.197
you know, Methuselahs is like super
scary, terrifying and also by the way,

00:28:09.280 --> 00:28:12.457
we're going to release it.
And then he's like, what what do you mean?

00:28:12.540 --> 00:28:14.917
>> [laughter] >> You know, so I think he

00:28:15.000 --> 00:28:20.077
Um I think probably in the future will
choose his words uh differently. Um

00:28:20.160 --> 00:28:22.157
>> But what's your sense >> if
you if you make if you terrify

00:28:22.240 --> 00:28:25.157
everyone and then say, "Oh, we're
releasing this terrifying thing."

00:28:25.240 --> 00:28:27.957
People will obviously be alarmed about
that. You know, you've literally told

00:28:28.040 --> 00:28:31.477
them to be scared and then you release
the scary thing and so they will be

00:28:31.560 --> 00:28:36.277
scared. That is natural. >> So I
I certainly don't want you to alarm

00:28:36.360 --> 00:28:38.237
people, but I do think there is a really

00:28:38.320 --> 00:28:41.637
important debate about how quickly and

00:28:41.720 --> 00:28:46.317
to what degree AI is displace people
from jobs. Because it is as you say, if

00:28:46.400 --> 00:28:49.317
within 10 years, it's more it's there's
more intelligent artificial intelligence

00:28:49.400 --> 00:28:51.477
than any human intelligence. >> Yes.

00:28:51.560 --> 00:28:56.197
>> And if you are, and I think you're
ahead of everyone this in this country on

00:28:56.280 --> 00:28:59.437
powering humanoid robots, that's
that's also going to be >> Yeah.

00:28:59.520 --> 00:29:03.117
>> kind of factory work.
So, do you Are you part Do

00:29:03.200 --> 00:29:05.077
you Are you closer to Dario, let's say,

00:29:05.160 --> 00:29:10.480
in your sense of the pace at which
jobs are going to be displaced?

00:29:11.880 --> 00:29:16.757
>> I I do think it's going
to be a bumpy road. Um

00:29:16.840 --> 00:29:20.717
because AI will be able to do any job

00:29:20.800 --> 00:29:24.557
better than any person can do that job.

00:29:24.640 --> 00:29:29.077
Um and that will be true very quickly of

00:29:29.160 --> 00:29:32.517
jobs that um are digital
or really anything that

00:29:32.600 --> 00:29:34.957
involves a person at a computer or on a

00:29:35.040 --> 00:29:38.117
phone. AI could do all of that um will

00:29:38.200 --> 00:29:42.237
be will be able to do
all of that very soon.

00:29:42.320 --> 00:29:43.877
I mean, for

00:29:43.960 --> 00:29:46.677
software engineering, we already have

00:29:46.760 --> 00:29:49.957
a situation where AI is better than

00:29:50.040 --> 00:29:54.757
at least 90% of humans
um at writing software.

00:29:54.840 --> 00:29:59.237
Or 90% of professional software
engineers at writing software.

00:29:59.320 --> 00:30:04.037
And I mean, it really is getting to
the point where it'll be better than 99%

00:30:04.120 --> 00:30:07.877
and then it'll get to the point where
there's just no way to compete. It'll

00:30:07.960 --> 00:30:10.397
It'll get to what I call Stockfish level.

00:30:10.480 --> 00:30:14.757
So, Stockfish. So,
Stockfish is the chess program

00:30:14.840 --> 00:30:20.077
that um Stockfish is so
good at chess that you

00:30:20.160 --> 00:30:22.877
know, it will

00:30:22.960 --> 00:30:27.237
easily beat Magnus Carlsen
or anyone uh at chess

00:30:27.320 --> 00:30:31.797
um on a even running on a small
computer. I suspect that

00:30:31.880 --> 00:30:35.157
you could probably run
Stockfish on your phone

00:30:35.240 --> 00:30:38.477
and beat Magnus Carlsen.
That's how good it is at chess.

00:30:38.560 --> 00:30:41.917
And that's how good AI
will be at writing software.

00:30:42.000 --> 00:30:46.877
>> And beyond software?
Beyond >> And that applies to everything.

00:30:46.960 --> 00:30:48.717
>> Everything >> Everything, everything.

00:30:48.800 --> 00:30:54.317
>> Everything, but what about you for
for physical jobs, for assembly jobs? You

00:30:54.400 --> 00:30:58.277
will also need the capability
in humanoid robots, which so far isn't

00:30:58.360 --> 00:31:01.957
there yet, is it? >> Right, so

00:31:02.040 --> 00:31:05.797
that's why I said digital AI and then
physically manifested AI in the form of

00:31:05.880 --> 00:31:09.797
robots, where the robot
is is more or less an end effector

00:31:09.880 --> 00:31:12.077
would have some amount of local
intelligence, but be [clears throat]

00:31:12.160 --> 00:31:15.077
managed by a large AI model.

00:31:15.160 --> 00:31:19.720
Um and uh

00:31:20.000 --> 00:31:25.037
I And I've said many times I
think work is going to be optional.

00:31:25.120 --> 00:31:26.077
So

00:31:26.160 --> 00:31:29.077
I didn't mean Let me So so I think maybe

00:31:29.160 --> 00:31:32.557
maybe a good a good uh

00:31:32.640 --> 00:31:36.077
comparison would be um gardening.

00:31:36.160 --> 00:31:39.317
Like you don't have to garden.
You know, you don't like

00:31:39.400 --> 00:31:43.197
you don't have to grow vegetables for
in your in your garden. You can go to the

00:31:43.280 --> 00:31:47.357
store and there are going to be pristine
vegetables, or you can grow vegetables

00:31:47.440 --> 00:31:50.997
in your garden and they
will actually be uh

00:31:51.080 --> 00:31:54.557
less perfect. Your tomato
is not going to be as plump and

00:31:54.640 --> 00:31:59.077
juicy as the ones at the store. They
they have They'll you know, look a little

00:31:59.160 --> 00:32:01.837
artisanal. Um But you know,
if you if you go to a

00:32:01.920 --> 00:32:04.717
friend's house and they they make dinner

00:32:04.800 --> 00:32:09.917
at and it's it's with vegetables from
their garden, it's it's a nice touch.

00:32:10.000 --> 00:32:12.317
So that's what So that's
what work is going to be like.

00:32:12.400 --> 00:32:15.397
>> AI. I get that. I mean, in that world
I would be hopeless cuz I can't garden to

00:32:15.480 --> 00:32:18.837
save my life, but I get the idea. People
still like playing chess with each other

00:32:18.920 --> 00:32:20.437
even though every computer is better.

00:32:20.520 --> 00:32:22.917
But let's just let's just think
about the politics of this.

00:32:23.000 --> 00:32:24.397
>> Yeah. >> Because
it sounds to you you're

00:32:24.480 --> 00:32:28.957
describing a world where within 10 years
pretty much every job can be done better

00:32:29.040 --> 00:32:31.917
by an intelligence. And >> Yes.

00:32:32.000 --> 00:32:35.197
>> you're seeing huge returns to this
intelligence to people like you. You're

00:32:35.280 --> 00:32:37.317
the wealthiest man in the world.

00:32:37.400 --> 00:32:41.477
What people who lose their jobs, who
are displaced, who may go off and garden,

00:32:41.560 --> 00:32:44.637
but how are they going to live? And what
kind of I mean, it sounds to me like

00:32:44.720 --> 00:32:47.437
>> to be universal high income.

00:32:47.520 --> 00:32:48.637
>> But how are you going to get there?

00:32:48.720 --> 00:32:51.717
Let's talk about the politics
of this cuz right now

00:32:51.800 --> 00:32:55.717
we if we look at the political pressures
in most countries, people are very, very

00:32:55.800 --> 00:32:57.997
worried about this, very against it.

00:32:58.080 --> 00:33:00.637
I mean, it seems to me that there's
a risk that we are heading more for

00:33:00.720 --> 00:33:04.917
revolution than for an easy route
to this nirvana you're playing out.

00:33:05.000 --> 00:33:09.517
>> I mean, it will be a bumpy road.
I'm not suggesting it will be

00:33:09.600 --> 00:33:13.637
a sort of that everything
will be smooth sailing here.

00:33:13.720 --> 00:33:19.317
Um so I mean, there are
many jobs that used to

00:33:19.400 --> 00:33:22.917
exist that don't exist
because of computers.

00:33:23.000 --> 00:33:28.317
I mean, the the word computer used
to refer to a person doing computations.

00:33:28.400 --> 00:33:31.077
Like that used to be a job
description of a person.

00:33:31.160 --> 00:33:35.437
And you can see pictures online
of um of buildings that have entire

00:33:35.520 --> 00:33:40.717
skyscrapers of just filled with people
doing calculations to to do things like

00:33:40.800 --> 00:33:43.637
calculate the interest
on your bank balance.

00:33:43.720 --> 00:33:47.677
Um this is how things were done

00:33:47.760 --> 00:33:51.397
for a long time until computers
came around. And even in

00:33:51.480 --> 00:33:55.197
the beginning, computers were
very expensive and hard to use.

00:33:55.280 --> 00:34:00.397
Uh so so the the thing that's different
here is that the the pace of change

00:34:00.480 --> 00:34:04.517
is radically accelerated. So

00:34:04.600 --> 00:34:07.797
but still people you know,
some people were upset that they

00:34:07.880 --> 00:34:10.797
their job of calculating
bank interest [clears throat]

00:34:10.880 --> 00:34:12.997
it would would no longer be available.

00:34:13.080 --> 00:34:16.837
Um but now I think there's it's you'd be

00:34:16.920 --> 00:34:20.037
hard-pressed to find people
who wanted to do that job.

00:34:20.120 --> 00:34:23.637
Um so I think I I have a suggestion for

00:34:23.720 --> 00:34:27.477
say what say science fiction

00:34:27.560 --> 00:34:30.237
books people might want to read to get a

00:34:30.320 --> 00:34:33.957
sense of what the future will be like.

00:34:34.040 --> 00:34:38.437
And the the best envisioning of an AI
future that I've read are the books by

00:34:38.520 --> 00:34:41.757
Ian Banks, the Culture books.
>> reading Excession.

00:34:41.840 --> 00:34:45.117
So I've I've thanks to you
I've started reading these.

00:34:45.200 --> 00:34:49.397
But I actually don't find that future
a terribly appealing one because I don't

00:34:49.480 --> 00:34:52.117
think humans have agency in that future.

00:34:52.200 --> 00:34:55.517
>> They have agency >>
Minimal compared to the minds.

00:34:55.600 --> 00:34:57.957
>> way. >> But yeah,
I really I mean Ian Banks as

00:34:58.040 --> 00:35:02.837
you know was a socialist. Um but Ian Banks
had a very different political philosophy.

00:35:02.920 --> 00:35:06.237
>> I'm not sure that's I'm not sure
I wouldn't totally agree with that.

00:35:06.320 --> 00:35:10.277
>> Let's agree that that could be where we
end up. I'm I'm concerned about the road

00:35:10.360 --> 00:35:13.477
and we live you know,
we live in democracies and

00:35:13.560 --> 00:35:15.957
people are very worried about this. So I

00:35:16.040 --> 00:35:20.637
guess two questions. Firstly, do you
have to have massive redistribution from

00:35:20.720 --> 00:35:24.197
people like you to the people
who are losing their jobs?

00:35:24.280 --> 00:35:27.557
Do you need to have Do you need
to pay more tax? Do you have

00:35:27.640 --> 00:35:30.197
much higher taxes on capital so that you

00:35:30.280 --> 00:35:32.597
fund some kind of universal income?

00:35:32.680 --> 00:35:35.397
Otherwise, what do people live off?

00:35:35.480 --> 00:35:39.477
>> I I think that the Treasury should
just simply issue people checks.

00:35:39.560 --> 00:35:41.917
Oh, you send >> the revenue
coming from? I mean I mean

00:35:42.000 --> 00:35:44.837
I'm an economist. If you just issue
checks, you're going to have inflation.

00:35:44.920 --> 00:35:48.237
>> It it you know, this is like
a lot of what we're trying to think

00:35:48.320 --> 00:35:52.477
about is you these things were relevant in

00:35:52.560 --> 00:35:55.677
the past, they will not be
relevant in the future. So

00:35:55.760 --> 00:35:58.197
like inflation is simply the ratio of

00:35:58.280 --> 00:36:01.117
money to the to to goods and services.

00:36:01.200 --> 00:36:03.637
So if the goods and services output

00:36:03.720 --> 00:36:06.597
increases dramatically, if it increases

00:36:06.680 --> 00:36:11.557
you know, 1,000% you can
increase the you can literally

00:36:11.640 --> 00:36:15.517
print all these case change the database.
It's not not really print money.

00:36:15.600 --> 00:36:17.797
You create money in the in the database.

00:36:17.880 --> 00:36:22.597
Um provided the creation of that money uh

00:36:22.680 --> 00:36:27.157
is less than the rate at which
goods and services increase um

00:36:27.240 --> 00:36:31.077
you you actually have deflation. So,
I'll make a prediction, which is that

00:36:31.160 --> 00:36:34.757
deflation will be the issue,
not inflation.

00:36:34.840 --> 00:36:39.237
The because because as the output
of goods and services increases, um if if

00:36:39.320 --> 00:36:42.277
the output of goods and services
increases faster than money supply, you

00:36:42.360 --> 00:36:44.797
will have deflation.
>> I'm actually I'm perfectly willing to

00:36:44.880 --> 00:36:46.157
believe that. You may well be right.

00:36:46.240 --> 00:36:47.677
Yeah, yeah, absolutely. You may well be

00:36:47.760 --> 00:36:50.797
right. But I'm I'm really

00:36:50.880 --> 00:36:54.197
hung up on how we get from the current

00:36:54.280 --> 00:36:58.237
polarized, angry, fearful

00:36:58.320 --> 00:37:02.637
policy we have now to this nirvana.
And and I think you could you can paint a

00:37:02.720 --> 00:37:04.917
picture where people
are so scared by this,

00:37:05.000 --> 00:37:08.277
you know, they essentially try to stop it.
Maybe they can't, but you get

00:37:08.360 --> 00:37:13.117
nationalized, there's a kind of,
you know, far left government, uh there's

00:37:13.200 --> 00:37:17.237
exorbitant tax increases, the kind
of economy crumbles because people are so

00:37:17.320 --> 00:37:21.517
worried about a future which does sound
quite scary as you describe it. So, how

00:37:21.600 --> 00:37:24.437
do you How do you handle the politics now?

00:37:24.520 --> 00:37:27.597
>> and terrifying. >> But if if
you were in the Yeah, exactly.

00:37:27.680 --> 00:37:29.717
Exactly. >> Honestly,
and if you ask me on any given

00:37:29.800 --> 00:37:31.557
day, I I I

00:37:31.640 --> 00:37:35.237
In fact, even intraday, I've gone from

00:37:35.320 --> 00:37:39.277
exhilaration to terror regarding AI. So,
I'm I'm going to be clear. I'm not like

00:37:39.360 --> 00:37:43.157
I'm not sort of some, you know,
I'm going to try to be sort of candid here

00:37:43.240 --> 00:37:47.877
and and just be some, oh,
everything's wonderful, some Panglossian

00:37:47.960 --> 00:37:51.837
view of the world where everything's,
you know, going to be great, no no no

00:37:51.920 --> 00:37:53.837
risk. I'm I'm saying there is some risk.

00:37:53.920 --> 00:37:58.357
Um >> I think we agree on that.
I guess I'm trying to be more sort of,

00:37:58.440 --> 00:38:03.517
you know, small bore and practical
and think about what can be done now to

00:38:03.600 --> 00:38:06.597
limit those risks. And you gave a very
strong argument for what you and and

00:38:06.680 --> 00:38:08.837
your fellow model makers could do.
What should the government do?

00:38:08.920 --> 00:38:13.317
>> would be good to for us to coordinate
and and just highlight issues

00:38:13.400 --> 00:38:17.997
that we see in each others' models
um to mitigate downside risk.

00:38:18.080 --> 00:38:20.477
>> Well, what should the Should the
government Should you be paying more tax

00:38:20.560 --> 00:38:23.597
to fund more transfer programs?

00:38:23.680 --> 00:38:26.917
>> Well, first of all,
I do I do pay a lot of tax. Um

00:38:27.000 --> 00:38:32.077
so, um I in fact I um

00:38:32.160 --> 00:38:36.717
I set a record for the most amount
of taxes ever paid by a by a human.

00:38:36.800 --> 00:38:38.957
>> That's also cuz you're
the world's richest man.

00:38:39.040 --> 00:38:42.710
>> Well, even before that. >> [laughter]

00:38:42.920 --> 00:38:46.277
>> So, my guess is that I will pay

00:38:46.360 --> 00:38:49.917
many trillions in tax.
And I'm fine with that.

00:38:50.000 --> 00:38:53.757
Um so yeah, um

00:38:53.840 --> 00:38:54.957
that's just the way it's going to be.

00:38:55.040 --> 00:38:58.437
But I don't think in the future
when money's relevant, I

00:38:58.520 --> 00:39:01.397
think taxation also
becomes somewhat irrelevant.

00:39:01.480 --> 00:39:04.677
>> about giving people a sense
that they have a share in the upside?

00:39:04.760 --> 00:39:09.477
>> Sure. >> Giving people shares in SpaceX
and Tesla, do you think that's the And

00:39:09.560 --> 00:39:11.517
there's a lot of talk
about this right now.

00:39:11.600 --> 00:39:15.397
>> Um well, one of the reasons
I took SpaceX public is so that

00:39:15.480 --> 00:39:20.397
the public could participate in they
could own a piece of SpaceX. Um

00:39:20.480 --> 00:39:23.997
whereas as a private company, we're very
limited in the number of shareholders

00:39:24.080 --> 00:39:27.797
we're allowed to have before going
public. So, as a public company,

00:39:27.880 --> 00:39:29.597
um we can have millions of shareholders.

00:39:29.680 --> 00:39:33.837
As a private company, we're we're limited
to just a few thousand shareholders.

00:39:33.920 --> 00:39:37.557
Um so >> Do you think there should be

00:39:37.640 --> 00:39:41.837
distribution of shares like, you know,
the Alaska wealth fund is there for which

00:39:41.920 --> 00:39:44.677
could you imagine a world and some
people talk about the Trump accounts

00:39:44.760 --> 00:39:48.677
being a kind of vehicle
for this that you a share of your

00:39:48.760 --> 00:39:51.597
equity goes to ordinary Americans?

00:39:51.680 --> 00:39:56.197
>> Right. Well, in in the taxes that like
that I that I pay, I mean I'll I'll

00:39:56.280 --> 00:40:00.917
you know, so for my stock
options, I pay 40%

00:40:01.000 --> 00:40:05.277
uh income tax on a national basis and then

00:40:05.360 --> 00:40:09.597
because I spend I don't know,
about 30% of my time in California,

00:40:09.680 --> 00:40:13.437
thereabouts, I pay an another
sort of a roughly a third of the

00:40:13.520 --> 00:40:15.517
California state taxes, which is

00:40:15.600 --> 00:40:18.117
I know I know around 15%. So, so I I pay

00:40:18.200 --> 00:40:21.157
about 45% in taxes just so people know

00:40:21.240 --> 00:40:25.197
know this, you know, I don't know
if it's quite so roughly half.

00:40:25.280 --> 00:40:29.197
>> Um then when I die,
there'll be another sort of

00:40:29.280 --> 00:40:33.437
>> 45% roughly half
that that will be taxed.

00:40:33.520 --> 00:40:36.677
Um so at at the end of the day I probably

00:40:36.760 --> 00:40:40.637
I'll have a quarter of what
I what I whatever I have. But

00:40:40.720 --> 00:40:44.077
the um

00:40:44.160 --> 00:40:47.917
I I think the only thing it gives
me is that for some period of time

00:40:48.000 --> 00:40:52.117
I would have control of the
direction of the companies.

00:40:52.200 --> 00:40:56.997
So it's not um that that's
pretty much all it it gets

00:40:57.080 --> 00:41:00.717
me and and until I guess
the AI is so smart that

00:41:00.800 --> 00:41:05.437
it's it's it's in charge. And at that
point controlling companies won't really

00:41:05.520 --> 00:41:07.397
matter. >> Well, let's
talk about that because I

00:41:07.480 --> 00:41:11.637
think one of the other things that worries
people about the direction we're

00:41:11.720 --> 00:41:15.077
heading in and I'm going to caricature
it but people think that the direction

00:41:15.160 --> 00:41:20.037
of the world is being shaped by a very
small number of incredibly powerful

00:41:20.120 --> 00:41:22.157
people and you are right
at the top of that list.

00:41:22.240 --> 00:41:25.117
>> But what what else is is this
different from the from the past?

00:41:25.200 --> 00:41:27.157
>> I agree with you. I think I think
[laughter] it was like that in the

00:41:27.240 --> 00:41:29.077
>> When when when wasn't that the case?

00:41:29.160 --> 00:41:31.037
>> ever thus.
>> It was ever thus, exactly.

00:41:31.120 --> 00:41:34.877
>> It is a as you've described very
vividly, it's an incredibly rapid

00:41:34.960 --> 00:41:38.197
change. And so let's start with control
of your company which you definitely

00:41:38.280 --> 00:41:40.797
have now. Um after the IPO you basically

00:41:40.880 --> 00:41:45.357
you have 80% odd of voting
shares and as I understand it

00:41:45.440 --> 00:41:49.517
you can only be removed by a vote of
shares which you yourself control. So so

00:41:49.600 --> 00:41:52.077
tell me if I'm wrong but I
think you can go crazy.

00:41:52.160 --> 00:41:56.797
>> in that bucket by the way. The The AI
companies are like Google's in that in

00:41:56.880 --> 00:42:00.077
that bucket. Alphabet
is controlled by Larry and Sergey.

00:42:00.160 --> 00:42:03.037
Um >> But I think yours is more
extreme. The inability to

00:42:03.120 --> 00:42:05.117
>> level of they have

00:42:05.200 --> 00:42:07.477
voting control of of that company.

00:42:07.560 --> 00:42:10.397
Mark Zuckerberg has control of a matter.

00:42:10.480 --> 00:42:14.397
And and you know, Mark is trying
very hard on the AI front and I think

00:42:14.480 --> 00:42:16.997
probably will will also have a leading

00:42:17.080 --> 00:42:19.637
AI he's made I think he released one

00:42:19.720 --> 00:42:25.037
last week. Um so I think
he's he's he'll be one of the

00:42:25.120 --> 00:42:27.917
you know Meta will be
one of the AI companies.

00:42:28.000 --> 00:42:29.597
>> But is it sensible for one person to

00:42:29.680 --> 00:42:32.717
have so much control over a company?

00:42:32.800 --> 00:42:35.157
>> Um well, it I really just need to make

00:42:35.240 --> 00:42:40.077
sure that I can focus on long-term

00:42:40.160 --> 00:42:44.477
By long-term I just mean
in the my 5 to 10 years. Um

00:42:44.560 --> 00:42:47.357
so one of the the challenges with being a

00:42:47.440 --> 00:42:51.997
public company is that it
is other sort of the pressure to to

00:42:52.080 --> 00:42:55.357
have uh great results
every single quarter.

00:42:55.440 --> 00:42:58.117
But and and and to not really invest in

00:42:58.200 --> 00:43:03.357
things that may only pay
off in 5 to 10 years. And

00:43:03.440 --> 00:43:06.997
um in the case of SpaceX, I need I I want

00:43:07.080 --> 00:43:09.877
to extend consciousness beyond Earth.

00:43:09.960 --> 00:43:12.877
And and have a

00:43:12.960 --> 00:43:17.597
growing civilization on the moon,
Mars, and elsewhere in the solar system.

00:43:17.680 --> 00:43:20.677
The return I I think ultimately
this would have a very high return on

00:43:20.760 --> 00:43:24.597
investment, but in the short term we'll
be spending all this money on a moon base

00:43:24.680 --> 00:43:28.917
or in a Mars base and then people are
like I you know, I'm going to I'm going to

00:43:29.000 --> 00:43:31.837
get all this pressure for like, "Well,
you missed your earnings this quarter

00:43:31.920 --> 00:43:35.637
cuz you spent too much on Mars." And I'm
like, "Yes." By the way, that was in the

00:43:35.720 --> 00:43:38.477
in the S1 prospectus and it's it's it's

00:43:38.560 --> 00:43:41.077
literally listed as a thing we will do.

00:43:41.160 --> 00:43:44.637
Um so we do so.
But but but they still get grumpy about

00:43:44.720 --> 00:43:48.037
these things, you know? Cuz they're like,
"Well, you had whatever, you know,

00:43:48.120 --> 00:43:51.237
you said your gross margin would be
this and it's actually slightly worse and

00:43:51.320 --> 00:43:55.077
then now now we're going to punish you
and and the short sellers all descend

00:43:55.160 --> 00:43:58.357
upon you and you get a lot of pressure
to public >> get a lot of that.

00:43:58.440 --> 00:44:00.877
>> Yeah, so I just
need some way to sort of

00:44:00.960 --> 00:44:04.757
you know, say like, "Listen, guys, we're
we're going to focus on 5 to 10 year or

00:44:04.840 --> 00:44:09.637
longer time horizon investments.
This will depress our short-term earnings,

00:44:09.720 --> 00:44:13.357
but the long I mean, if you take
a long-term view and there are investors

00:44:13.440 --> 00:44:16.077
we have many investors that do
take a long-term view. In fact,

00:44:16.160 --> 00:44:20.437
I'm a big fan of the the retail investors.
In fact, I find our retail investors

00:44:20.520 --> 00:44:22.917
um on balance have the have the

00:44:23.000 --> 00:44:26.037
the they're very insightful and and have

00:44:26.120 --> 00:44:30.477
very long-term views.
Um But but it part of this comes to you

00:44:30.560 --> 00:44:31.477
say, "What's the incentive structure?"

00:44:31.560 --> 00:44:36.437
Is it the for portfolio managers,
it's it's sort of understandable what why

00:44:36.520 --> 00:44:38.437
they put so much pressure on on a on a

00:44:38.520 --> 00:44:40.117
quarterly earnings because

00:44:40.200 --> 00:44:44.117
um their their compensation and their um

00:44:44.200 --> 00:44:48.677
and whether they get promoted or not
is dependent on how their portfolio does.

00:44:48.760 --> 00:44:51.677
And they they have they
only have a few years

00:44:51.760 --> 00:44:55.597
to to make these decisions. Um and then

00:44:55.680 --> 00:45:00.157
they're evaluated based on those decisions
over the course of sometimes only

00:45:00.240 --> 00:45:03.557
a year or less. And uh and so
since since they're under

00:45:03.640 --> 00:45:07.477
pressure to have immediate
returns, they then put

00:45:07.560 --> 00:45:10.957
companies under pressure to have
immediate returns. So once you

00:45:11.040 --> 00:45:12.437
understand the incentive structure, the

00:45:12.520 --> 00:45:14.917
behavior of naturally follows.

00:45:15.000 --> 00:45:20.437
>> What about though key man risk? I mean,
what happens you know, I heaven forbid,

00:45:20.520 --> 00:45:24.557
but what happens if you get run over
by the proverbial bus tomorrow? You know,

00:45:24.640 --> 00:45:27.517
it is you you have
total control over these

00:45:27.600 --> 00:45:31.077
companies. Do you have a plan for how
they would progress without you?

00:45:31.160 --> 00:45:33.717
>> I I think they would actually

00:45:33.800 --> 00:45:38.197
do very well for for for several years.

00:45:38.280 --> 00:45:40.237
Um

00:45:40.320 --> 00:45:45.277
And you know, there's a clear road
map for or space map, I should say, for

00:45:45.360 --> 00:45:50.357
SpaceX. Um And there's a a road map

00:45:50.440 --> 00:45:53.557
actually for Tesla literally. Um

00:45:53.640 --> 00:45:57.517
So I think I think there's you
know, if you look at Apple after

00:45:57.600 --> 00:46:00.357
Steve Jobs, you know, Steve Jobs had

00:46:00.440 --> 00:46:01.797
you know, he he he had a strategic

00:46:01.880 --> 00:46:05.597
vision and a bunch of products that um

00:46:05.680 --> 00:46:09.717
that that he had had ideas
for and Apple has done very well.

00:46:09.800 --> 00:46:13.917
Um but but the but when you
don't have a creative genius

00:46:14.000 --> 00:46:18.637
like Steve Jobs you don't
you you you what what you miss

00:46:18.720 --> 00:46:22.637
what's what Apple's missing are
the incredible breakthrough products.

00:46:22.720 --> 00:46:26.357
>> Yeah. >> So they make
amazing phones, you know,

00:46:26.440 --> 00:46:30.557
great computers um but Apple
hasn't come up with a sort of

00:46:30.640 --> 00:46:33.437
Steve Jobs Steve Jobs
level breakthrough product.

00:46:33.520 --> 00:46:38.197
>> So by analogy, that's what you
bring to all of your companies.

00:46:38.280 --> 00:46:43.117
You mentioned Mars, you know,
most people think of you

00:46:43.200 --> 00:46:46.917
as the man who wants to, you know,
create a colony on Mars so we have

00:46:47.000 --> 00:46:49.837
multi-planetary civilization.
>> saying that for such a long time.

00:46:49.920 --> 00:46:53.037
>> You have but now you seem more focused

00:46:53.120 --> 00:46:57.197
on AI and what AI will do in the next
10 years and you you sound much more

00:46:57.280 --> 00:46:59.477
excited about using the moon as a base

00:46:59.560 --> 00:47:03.117
from which to, you know, pop out orbital

00:47:03.200 --> 00:47:06.637
AI satellites. Have you
shifted your attention away

00:47:06.720 --> 00:47:08.757
from Mars or are you as interested in

00:47:08.840 --> 00:47:11.757
colonizing Mars as you ever were?

00:47:11.840 --> 00:47:16.117
>> I I'm I'm interested in the set
of actions that propagate consciousness

00:47:16.200 --> 00:47:19.237
into the future to the
maximum degree possible.

00:47:19.320 --> 00:47:23.157
So that the future light
cone of consciousness

00:47:23.240 --> 00:47:26.437
is as big as as big as we
can, you know, like like

00:47:26.520 --> 00:47:30.357
you know, there's a there's an exciting
future where there's a where there's a

00:47:30.440 --> 00:47:35.117
spacefaring civilization
out there among the stars and

00:47:35.200 --> 00:47:37.717
and the the great things
that are envisioned in

00:47:37.800 --> 00:47:42.477
the non-dystopian sci-fi

00:47:42.560 --> 00:47:46.757
books and movies come become real one day.

00:47:46.840 --> 00:47:51.717
I sort of want the Star Trek you
know Star Wars future. Star Wars was

00:47:51.800 --> 00:47:53.477
the first movie I ever saw
in the theater, by the way.

00:47:53.560 --> 00:47:57.237
>> Really? >> It was probably maybe had
a profound effect on me, I suppose. Um

00:47:57.320 --> 00:48:00.397
so, I remember I was I I was

00:48:00.480 --> 00:48:04.717
it blew my mind, honestly. I was I think
I was 6 years old and um I was like,

00:48:04.800 --> 00:48:06.397
"This is the greatest
thing I've ever seen."

00:48:06.480 --> 00:48:09.837
>> Has your life been
about making that happen?

00:48:09.920 --> 00:48:13.717
>> Yeah. I mean, it I didn't
think it would be, but at the

00:48:13.800 --> 00:48:17.557
when I was young, but um I'm my life
feels somewhat surreal. I I can't

00:48:17.640 --> 00:48:19.917
believe these things are
actually happening, you know.

00:48:20.000 --> 00:48:22.277
Um >> I think the world
feels somewhat surreal.

00:48:22.360 --> 00:48:27.477
>> in simulation theory.
Where it's like, "Well, this

00:48:27.560 --> 00:48:31.997
Like the the things I'm doing
are so preposterous that um

00:48:32.080 --> 00:48:33.917
it's hard to believe they're real.

00:48:34.000 --> 00:48:39.037
I mean, we've got Starship,
which is the largest, most powerful

00:48:39.120 --> 00:48:42.797
flying object ever made. Um

00:48:42.880 --> 00:48:45.637
and we intend to launch that

00:48:45.720 --> 00:48:50.317
more than once per hour. >> I know.
I read that. And we can, you

00:48:50.400 --> 00:48:52.877
know, people can quibble about how
quickly it happens, but I think based on

00:48:52.960 --> 00:48:55.877
your track record, we should
assume it will happen at some point.

00:48:55.960 --> 00:48:59.277
>> It will happen. >> I I'm
just sort of struck that when I

00:48:59.360 --> 00:49:01.397
when I I am as as blown away by this

00:49:01.480 --> 00:49:03.197
world as you are and I'm I watch it from

00:49:03.280 --> 00:49:06.757
a from a much more naive level, but

00:49:06.840 --> 00:49:09.437
for me, the whole evolution of AI and

00:49:09.520 --> 00:49:12.277
the capabilities of these models is in

00:49:12.360 --> 00:49:14.997
and of itself so mind-blowing.

00:49:15.080 --> 00:49:18.277
>> It's pretty much happening
as I as I thought thought it would.

00:49:18.360 --> 00:49:21.397
>> Is it? >> Yeah. >> Oh,
that's interesting. [laughter]

00:49:21.480 --> 00:49:23.517
>> Um

00:49:23.600 --> 00:49:27.117
And not just me, but many people um

00:49:27.200 --> 00:49:31.717
predicted this. Like, Ray Kurzweil predict
Ray Ray Kurzweil's predictions have

00:49:31.800 --> 00:49:34.717
been remarkably accurate.
>> Yeah, the singularity.

00:49:34.800 --> 00:49:37.037
>> Yeah. >> Approaching.
Let's go back and more

00:49:37.120 --> 00:49:40.597
immediately to the here and now. We've
talked about concentration of control in

00:49:40.680 --> 00:49:43.517
your companies. You made a very good
case for why you should have it. You

00:49:43.600 --> 00:49:49.517
also have, through Starlink, incredible
geopolitical power. I mean, I'm I'm one

00:49:49.600 --> 00:49:50.757
of your customers. >> Thank you.

00:49:50.840 --> 00:49:52.277
>> Excellent product.
>> I I hope you're right.

00:49:52.360 --> 00:49:55.637
>> But without them, the Ukrainians would
have been in serious serious trouble.

00:49:55.720 --> 00:49:57.757
>> Um yes, the Ukrainian government has

00:49:57.840 --> 00:50:00.717
publicly thanked Starlink many times for

00:50:00.800 --> 00:50:03.077
being instrumental in

00:50:03.160 --> 00:50:07.757
uh defending Ukraine. >> And you
recently decided to prevent the

00:50:07.840 --> 00:50:11.757
Russians in the occupied territories
from having access to Starlink.

00:50:11.840 --> 00:50:16.597
>> Um well, no that we do we
were we were never at any point

00:50:16.680 --> 00:50:19.677
selling Starlink to Russians.
Um but they were grabbing it, right?

00:50:19.760 --> 00:50:21.677
They were using it and you now >> Yes.

00:50:21.760 --> 00:50:25.957
Well, what they they they
essentially got Starlink units

00:50:26.040 --> 00:50:31.477
that they ordered through Ukraine
and then and then smuggled them into

00:50:31.560 --> 00:50:34.557
the into eastern Ukraine
where the Russians are.

00:50:34.640 --> 00:50:40.077
Um and were then using
them in some cases for

00:50:40.160 --> 00:50:44.397
attacks. Um Now, the

00:50:44.480 --> 00:50:49.157
in order to address this,
um what we worked with the Ukrainian

00:50:49.240 --> 00:50:53.717
government to create a white list
of approved Starlink terminals.

00:50:53.800 --> 00:50:57.877
Now, this has some downsides because
a whole bunch of people who were just

00:50:57.960 --> 00:51:01.997
innocent users of Starlink and just
trying to access the internet for work

00:51:02.080 --> 00:51:06.037
or for education now don't have Starlink.

00:51:06.120 --> 00:51:09.997
>> These are geopolitical decisions
of consequence. How do you feel about

00:51:10.080 --> 00:51:13.437
having that power? You could
determine the outcome of wars.

00:51:13.520 --> 00:51:18.357
How does that feel? And is it right
that you should have that power?

00:51:18.440 --> 00:51:24.077
>> Well, I I'm not sure we're we're
determining the ultimate outcome of the

00:51:24.160 --> 00:51:28.600
you know, that the say
the war in Ukraine. Um

00:51:29.320 --> 00:51:33.557
You know, I I I think I do hope
that that that that there's some sort of

00:51:33.640 --> 00:51:37.197
peace agreement reached soon because um

00:51:37.280 --> 00:51:39.357
for

00:51:39.440 --> 00:51:44.557
like two or three years now
the border has barely moved and um

00:51:44.640 --> 00:51:47.797
and every day there are
lots of people dying at the

00:51:47.880 --> 00:51:52.117
front line and um and I
think there's a lot of wishful

00:51:52.200 --> 00:51:56.157
thinking here of of people saying well
Russia should completely withdraw etc.

00:51:56.240 --> 00:51:58.317
But this is not realistic. Um I really

00:51:58.400 --> 00:52:02.157
think there needs to be some

00:52:02.240 --> 00:52:07.437
um you know a peace with that that is

00:52:07.520 --> 00:52:10.197
that you know we have
to be pragmatic about

00:52:10.280 --> 00:52:13.637
this. What we can't what I think would be

00:52:13.720 --> 00:52:17.957
terrible is if if there's if there's
another three years of people dying

00:52:18.040 --> 00:52:20.077
senselessly at the front lines.

00:52:20.160 --> 00:52:26.160
Um with with the and you
have to say like well

00:52:28.840 --> 00:52:33.117
you know how do you explain to the
the the children of of the men that

00:52:33.200 --> 00:52:36.477
are dying at the front lines. Russia
on the Russian side too by the way they're

00:52:36.560 --> 00:52:39.357
they're being conscripted
and sent there. I'm white

00:52:39.440 --> 00:52:42.957
offended by sort of fancy
diplomats who have

00:52:43.040 --> 00:52:45.797
you know seven course dinners in in

00:52:45.880 --> 00:52:50.277
you know in luxurious places while
people are dying at the front lines and

00:52:50.360 --> 00:52:54.197
who who pontificate about how Russia
should withdraw but they're not going to

00:52:54.280 --> 00:52:56.597
Russia's Russia's not going to withdraw.

00:52:56.680 --> 00:53:02.317
So there really needs to be an accepted
peace long along those lines.

00:53:02.400 --> 00:53:04.277
Um >> And do you think it's your

00:53:04.360 --> 00:53:09.357
responsibility to play a role in tipping
the scales towards what you see as the

00:53:09.440 --> 00:53:11.837
right way to peace? >> I

00:53:11.920 --> 00:53:17.157
I'd like I've been advocating peace
for a long time. I you know I I posted

00:53:17.240 --> 00:53:22.157
a few years ago that that that there
really needs to be some concession

00:53:22.240 --> 00:53:25.877
here to Russia.
Um This is not me being pro-Russia. I'm

00:53:25.960 --> 00:53:30.437
just being pragmatic um
or people will die for years

00:53:30.520 --> 00:53:34.877
uh with nothing achieved
and that is exactly

00:53:34.960 --> 00:53:39.477
what happened. >> I don't know. I mean
the Ukrainians >> Exactly what happened.

00:53:39.560 --> 00:53:42.917
>> The Ukrainians do are
doing pretty well right now at

00:53:43.000 --> 00:53:46.997
you know, with that incredible drone
capacity hitting inside Russia and the

00:53:47.080 --> 00:53:51.557
and the kind of general consensus is that,
you know, the the the balance of

00:53:51.640 --> 00:53:54.317
the war is shifting. But,
my point is not so much about what

00:53:54.400 --> 00:53:55.877
>> not going to retake those territories.

00:53:55.960 --> 00:53:58.917
It is not going to happen.
>> My interest is really on whether you

00:53:59.000 --> 00:54:03.437
think given that you you have through
Starlink, you have an incredible power,

00:54:03.520 --> 00:54:06.957
whether it's appropriate that you
should wield it. And that's

00:54:07.040 --> 00:54:09.637
quite rare that an individual actor has

00:54:09.720 --> 00:54:14.797
such power by turning on and off
a capability to shape conflict.

00:54:14.880 --> 00:54:18.720
>> Well, is there something you
think I should do differently?

00:54:19.520 --> 00:54:22.797
>> I mean, I I think you were right to
support Ukraine. I think that, you know,

00:54:22.880 --> 00:54:27.357
as you as it won't surprise you,
I I think that the Russian aggression was

00:54:27.440 --> 00:54:31.277
uh I, you know, for many, many reasons
an appalling aggression. And I I think

00:54:31.360 --> 00:54:34.077
that that was an important
thing that you did.

00:54:34.160 --> 00:54:38.717
>> Yeah, I mean, I did a a very deep dive
into the background of that conflict to

00:54:38.800 --> 00:54:41.037
understand exactly what was going on.

00:54:41.120 --> 00:54:45.320
Um and um

00:54:46.240 --> 00:54:50.397
I'm reminded of the old adage
that there are no angels in war.

00:54:50.480 --> 00:54:54.517
>> It's a similar tone, I think,
about domestic politics where you've

00:54:54.600 --> 00:54:58.037
you know, you've you've
played a big role domestically

00:54:58.120 --> 00:55:02.077
politically here. You spend a lot
of money on Donald Trump's re-election.

00:55:02.160 --> 00:55:05.357
After the election, you got very
actively involved, you know, through

00:55:05.440 --> 00:55:07.717
Dodge, you tried to cut a lot of

00:55:07.800 --> 00:55:10.477
spending, reshape the government.

00:55:10.560 --> 00:55:14.677
How do you feel that it wasn't hugely
successful in the sense that, you know,

00:55:14.760 --> 00:55:17.277
there wasn't nearly as much spending
cut as you tried. How do you as you look

00:55:17.360 --> 00:55:21.317
back on that?
Do you regret getting so involved in

00:55:21.400 --> 00:55:23.837
politics? Do you think
it was a good thing to do?

00:55:23.920 --> 00:55:26.477
>> I think I I think I got a little too
involved in politics, got carried away,

00:55:26.560 --> 00:55:31.917
frankly. Um so the um

00:55:32.000 --> 00:55:34.877
you know, the the the goal with the sort
of Dodge stuff was was to try to do

00:55:34.960 --> 00:55:39.077
something about the the deficit,
which is massive. Um

00:55:39.160 --> 00:55:44.077
where the interest payments
exceed the amount of money spent on

00:55:44.160 --> 00:55:47.797
the Department of War. Like basically,
if you add up all the money spent on

00:55:47.880 --> 00:55:51.157
um what is now the Department of War and

00:55:51.240 --> 00:55:55.477
the and and intelligence in the US,
uh that is smaller than the interest

00:55:55.560 --> 00:55:59.517
payments on the debt. So, that well,
we we we we need to sort of take a close

00:55:59.600 --> 00:56:02.957
look at spending, um and make
sure that uh money is being

00:56:03.040 --> 00:56:04.717
spent in sensible ways and that it is

00:56:04.800 --> 00:56:08.557
not wasted or spent on on fraudulent

00:56:08.640 --> 00:56:13.717
activities. Um and um I got
a lot of flak from this for this,

00:56:13.800 --> 00:56:16.997
obviously. Um but the the
the people don't understand

00:56:17.080 --> 00:56:19.877
just how um

00:56:19.960 --> 00:56:22.077
simple it was for for the DOS team to

00:56:22.160 --> 00:56:24.557
approve payments. Um

00:56:24.640 --> 00:56:29.797
if if we had literally we we we just
asked to have the contact information of

00:56:29.880 --> 00:56:33.557
the recipients, um or some
evidence that say money was

00:56:33.640 --> 00:56:37.597
going to um its intended purpose.

00:56:37.680 --> 00:56:40.797
Um but over and over
there we were we told oh

00:56:40.880 --> 00:56:45.317
we we need to wire money for some
sort of important cause in Africa.

00:56:45.400 --> 00:56:49.757
And we're like, "Okay, but the wiring
wiring instructions are to Deloitte &

00:56:49.840 --> 00:56:53.077
Touche in DC in Washington,
DC, not to Africa.

00:56:53.160 --> 00:56:56.557
So, we would like to send the money
to Africa. Can you please connect us with

00:56:56.640 --> 00:56:59.757
the recipients so we can
send them the money directly?"

00:56:59.840 --> 00:57:03.397
And then we'd get silence.
>> I I I have We don't need to relitigate

00:57:03.480 --> 00:57:08.357
DOS. I have no doubt that there
was there's a lot of waste in the US

00:57:08.440 --> 00:57:11.877
government. I think probably >> Much
worse than anyone could possibly imagine.

00:57:11.960 --> 00:57:15.517
>> But the way I I mean, I think there
is it is a fair criticism that the way you

00:57:15.600 --> 00:57:19.077
shut down a whole bunch of programs
probably in fact, certainly caused

00:57:19.160 --> 00:57:21.357
unnecessary suffering. I mean, >> No.

00:57:21.440 --> 00:57:25.157
>> No? >> No.
>> You You don't buy that at all.

00:57:25.240 --> 00:57:27.477
>> It It caused a lot of complaining.

00:57:27.560 --> 00:57:31.917
>> You're watching my interview with Elon
Musk. If you want to get early access to

00:57:32.000 --> 00:57:35.917
other interviews, subscribe to the
economist by clicking the link above.

00:57:36.000 --> 00:57:40.277
Now, back to the interview. You don't So,
you don't buy the idea that anybody,

00:57:40.360 --> 00:57:44.237
and to put it really crudely and simply,
any kid in Africa died because of the

00:57:44.320 --> 00:57:46.517
suddenness with which >> Zero.

00:57:46.600 --> 00:57:49.557
>> I just don't think that's right. I just
don't think my I mean, I've had people

00:57:49.640 --> 00:57:53.397
reporting in Africa >> How many
NGOs are there that could that

00:57:53.480 --> 00:57:57.797
could that could step in?
Cuz bear in mind, this it it was

00:57:57.880 --> 00:58:02.477
it happened quickly. So, this and there
are literally tens of thousands of NGOs.

00:58:02.560 --> 00:58:06.437
>> Yeah, not everyone functions as far
as as you do. If if if if funding is cut

00:58:06.520 --> 00:58:11.557
off for antiretrovirals overnight and it
just stops, then people will die when

00:58:11.640 --> 00:58:15.117
they don't have access to them. And it
just >> No, it No, only if they cannot get

00:58:15.200 --> 00:58:18.917
funding for that. But what
the Gates Foundation has $50 billion.

00:58:19.000 --> 00:58:21.597
What's what What What about that?

00:58:21.680 --> 00:58:25.117
What about any one of a number
of organizations? What about um

00:58:25.200 --> 00:58:28.877
the uh um >> Well, in many
of these countries, USAID

00:58:28.960 --> 00:58:33.917
>> whatever last name is must be Bezos
uh gave $50 billion. What about that?

00:58:34.000 --> 00:58:35.357
>> But you don't ask
me Let's be practical.

00:58:35.440 --> 00:58:38.517
In many of these countries, USAID
was the biggest funder. But I I don't want

00:58:38.600 --> 00:58:41.197
to have an >> USAID
was a political organization.

00:58:41.280 --> 00:58:44.117
>> It may also have been that, but it
was the biggest funder of of health care

00:58:44.200 --> 00:58:48.277
system. So, I I I am I'm
very sure that there's >> of

00:58:48.360 --> 00:58:53.357
of regime change. >> And and there
may well be benefits from having

00:58:53.440 --> 00:58:55.677
changed the system. >> funding
was not stopped. It was moved to

00:58:55.760 --> 00:58:58.157
the State Department. >> It was stopped.
There were huge numbers

00:58:58.240 --> 00:59:01.877
of things shut down in Africa that were
that were kind of health related.

00:59:01.960 --> 00:59:05.157
>> of this. These claims are false.

00:59:05.240 --> 00:59:07.997
They're nonsense. And and absolute

00:59:08.080 --> 00:59:11.117
nonsense. But but obviously

00:59:11.200 --> 00:59:12.757
just for

00:59:12.840 --> 00:59:15.957
when funding is stopped for a fraudulent

00:59:16.040 --> 00:59:20.077
organizations or for something, do you
think they're going to say Do you think

00:59:20.160 --> 00:59:23.917
they're going to say, "Well, we please
keep funding the fraud?" Or do you think

00:59:24.000 --> 00:59:28.397
they're going to come up with a sad
story that sounds sympathetic in order

00:59:28.480 --> 00:59:31.597
to get the funding restored. Obviously,
it would be the latter, and that's

00:59:31.680 --> 00:59:36.197
exactly what they did.
But, zero died because

00:59:36.280 --> 00:59:38.557
>> Well, that's what
I That's what I That's

00:59:38.640 --> 00:59:41.637
what I would take issue with.
I think you can I actually have a lot of

00:59:41.720 --> 00:59:46.197
sympathy for the change of the aid
regime that the US under the Trump

00:59:46.280 --> 00:59:47.917
administration has precipitated.

00:59:48.000 --> 00:59:50.677
However, I think the pace and the

00:59:50.760 --> 00:59:55.877
suddenness with which it happened will
almost certainly have caused unnecessary

00:59:55.960 --> 00:59:57.717
suffering and deaths.
And I think both of those could have been

00:59:57.800 --> 01:00:00.437
>> that's totally false.
>> Okay. Well, let's

01:00:00.520 --> 01:00:04.117
>> And obviously false
because there there are are

01:00:04.200 --> 01:00:07.717
the amount of funding we're talking
about uh is far less than what can be

01:00:07.800 --> 01:00:12.437
done that that that what can
be supported by a private uh

01:00:12.520 --> 01:00:16.997
by a foundation like the Gates
Foundation uh or McKenzie Bezos's

01:00:17.080 --> 01:00:20.317
foundation. Um she's given away
tens of billions of dollars.

01:00:20.400 --> 01:00:25.117
Um the Gates Foundation has tens of
billions of dollars. Uh they are capable

01:00:25.200 --> 01:00:28.877
of acting very quickly.
Um and if they didn't, I would say

01:00:28.960 --> 01:00:31.757
they're equally responsible.
>> I see. Okay.

01:00:31.840 --> 01:00:35.837
>> Obviously. >> All right.
Fair point. Um >> Zero people

01:00:35.920 --> 01:00:37.437
>> I hear I hear you loud and clear.

01:00:37.520 --> 01:00:39.677
>> Zero. >> Hear you loud and clear.

01:00:39.760 --> 01:00:43.997
>> Obviously. >> Uh domestically,
you you stepped away from

01:00:44.080 --> 01:00:47.277
those. It It was hard to do what you
were trying to do. You've just explained

01:00:47.360 --> 01:00:50.277
it. But, this administration has gone in

01:00:50.360 --> 01:00:54.477
directions that I suspect
you were not expecting

01:00:54.560 --> 01:00:56.357
2 and 1/2 years ago. You know, the >>
going to agree with I'm not going to

01:00:56.440 --> 01:00:59.437
agree with everything
any administration does.

01:00:59.520 --> 01:01:01.397
>> Are you disappointed
by this administration?

01:01:01.480 --> 01:01:05.197
>> No, I think this is a this
administration on balance uh is is

01:01:05.280 --> 01:01:09.117
excellent.
No No administration is perfect, but

01:01:09.200 --> 01:01:12.757
it's a vastly better than the alternative.

01:01:12.840 --> 01:01:17.637
>> Can we talk about Europe? A place
that you are very, very focused on.

01:01:17.720 --> 01:01:21.117
Um >> that focused on Europe.
>> I think judging by your posts, you're

01:01:21.200 --> 01:01:25.037
pretty concerned about Europe.
Uh you you support

01:01:25.120 --> 01:01:28.317
>> I I once in a while I post something.

01:01:28.400 --> 01:01:31.557
>> It's reasonably frequently. It's not
once in a while. But But I'm because I'm

01:01:31.640 --> 01:01:35.197
genuinely interested in this as someone
who lives in Europe, lives in London.

01:01:35.280 --> 01:01:37.717
>> Yeah. >> Uh you paint
a picture Well, first of

01:01:37.800 --> 01:01:40.917
all, you support not just the populist
right, but the far right. In fact, very

01:01:41.000 --> 01:01:42.597
fringe parties in some countries.

01:01:42.680 --> 01:01:44.757
>> support the normal people. >> Okay.

01:01:44.840 --> 01:01:47.077
>> What do you call the far right falsely?

01:01:47.160 --> 01:01:48.957
>> You think it's the normal person. Okay.

01:01:49.040 --> 01:01:51.877
>> Yes. >> Um >> Literally,
you could you could go back

01:01:51.960 --> 01:01:55.197
uh 10 or 15 years and these these these

01:01:55.280 --> 01:01:57.437
these policies were completely normal.

01:01:57.520 --> 01:02:01.117
In fact, you you uh one of the one
of the one of the tricks I think is is is

01:02:01.200 --> 01:02:06.037
funny. I think it's it's quite funny
is to like take a speech from Obama and

01:02:06.120 --> 01:02:11.077
Hillary and then and then go to someone
and say uh who's who's sort of um

01:02:11.160 --> 01:02:14.917
you know, on the sort of what I call
the lunatic left. Um and say, "What do you

01:02:15.000 --> 01:02:18.277
think of the speech by Trump?" And like,
"Oh wow, he's the worst person ever."

01:02:18.360 --> 01:02:22.957
Actually, that was a speech by Obama.
Or actually, that was a speech by Hillary.

01:02:23.040 --> 01:02:26.317
>> Uh I I I I I agree No,
I agree that the center

01:02:26.400 --> 01:02:28.237
of gravity in this country has shifted.

01:02:28.320 --> 01:02:30.357
Certainly, the Democratic Party shifted
to the left. But that's not what I'm

01:02:30.440 --> 01:02:32.637
talking about. I'm talking about people
like Rupert Lowe. I'm talking about

01:02:32.720 --> 01:02:36.397
>> please. It's It's actually just normal
people. Um it's just normal people. It's

01:02:36.480 --> 01:02:40.197
just normal people. And um the And And
here are the principles. And And tell me

01:02:40.280 --> 01:02:44.677
which of these sound terrible. Um that
we should have secure borders, that we

01:02:44.760 --> 01:02:49.597
should have safe cities,
uh that we should have sensible spending.

01:02:49.680 --> 01:02:53.757
Which of those three are are
are are are far right fringe?

01:02:53.840 --> 01:02:56.677
>> I would like to just
admonish you >> Do you

01:02:56.760 --> 01:02:58.397
>> and the media

01:02:58.480 --> 01:03:01.197
for the absurd characterization of the

01:03:01.280 --> 01:03:04.117
far right. It is false and misleading

01:03:04.200 --> 01:03:06.917
and nonsense. Please keep this part in.

01:03:07.000 --> 01:03:09.437
>> We'll keep all of this in.
But let me >> [laughter] >> Let me Let me

01:03:09.520 --> 01:03:11.877
>> Don't cut this part.
>> Let me push back on that. Let me push

01:03:11.960 --> 01:03:13.957
back on that because you have

01:03:14.040 --> 01:03:19.317
>> You know, you have What is it?
Close to 250 million followers on X.

01:03:19.400 --> 01:03:21.557
>> Yeah, cuz I'm >> You say
things like >> Popular guy.

01:03:21.640 --> 01:03:25.437
>> quote, "Civil war in Britain
is inevitable. Just a question of when."

01:03:25.520 --> 01:03:29.997
>> yes. >> I mean,
when were you last in the UK?

01:03:30.080 --> 01:03:33.997
>> Um a few years ago. >> Exactly.
So So I live there. I'm there

01:03:34.080 --> 01:03:35.397
the whole time. That is nonsense.

01:03:35.480 --> 01:03:38.397
>> in a cloistered existence. >> No,
I don't, actually. And the numbers

01:03:38.480 --> 01:03:41.037
also disprove what you say.
>> What numbers?

01:03:41.120 --> 01:03:44.597
>> All the numbers about crime. London
is a much safer city than any city in the

01:03:44.680 --> 01:03:48.197
United States. Violent crime in the UK
is not where you think it is. In fact,

01:03:48.280 --> 01:03:51.557
it's coming down. You You have painted,
and because you have so many followers,

01:03:51.640 --> 01:03:53.237
so many Americans believe it, a a kind

01:03:53.320 --> 01:03:57.317
of view that Europe and and the UK are

01:03:57.400 --> 01:04:01.357
kind of, you know, overrun with
terrifying Muslim immigrants. There's

01:04:01.440 --> 01:04:04.597
rape on every corner. There's
a sort of civilizational destruction.

01:04:04.680 --> 01:04:07.517
>> any of those things. >> not Well,
you said "Civil war in Britain

01:04:07.600 --> 01:04:10.677
is inevitable." And then >> Yes,
if current trends continue, it is.

01:04:10.760 --> 01:04:15.357
>> Why? >> Well, if if you have uh a

01:04:15.440 --> 01:04:16.837
large

01:04:16.920 --> 01:04:19.437
uh and growing, rapidly growing

01:04:19.520 --> 01:04:22.877
uh group of people whose beliefs are

01:04:22.960 --> 01:04:28.157
antithetical to Western beliefs,
um at some point there will be a

01:04:28.240 --> 01:04:31.477
reckoning. They They They
They will They will try

01:04:31.560 --> 01:04:36.157
to implement their views, and those
who oppose it will oppose it, and that's

01:04:36.240 --> 01:04:41.757
what you call civil war.
And obviously, this is the trend. Um and

01:04:41.840 --> 01:04:46.277
uh and and you absolutely do live in a
cloistered existence. And if you'd like,

01:04:46.360 --> 01:04:47.797
we can go on a tour of Britain.

01:04:47.880 --> 01:04:50.917
>> Sure. I I'd be very happy
to come on a tour of Britain.

01:04:51.000 --> 01:04:52.637
>> And then and you will see for yourself.

01:04:52.720 --> 01:04:56.637
>> country that I live in all the time,
and you haven't been to a few years ago.

01:04:56.720 --> 01:04:59.797
>> literally 100 times.
>> But when was the last time?

01:04:59.880 --> 01:05:00.677
>> Like literally a few years ago.

01:05:00.760 --> 01:05:02.677
>> You know why I this exercises me?

01:05:02.760 --> 01:05:05.517
Because you have really shaped opinion

01:05:05.600 --> 01:05:09.957
about the UK, and so many Americans I
come across, and I have conversations

01:05:10.040 --> 01:05:13.597
with them, and they have been And I just
had one the day, along the order of oh,

01:05:13.680 --> 01:05:16.677
London's such a dangerous place.
How can you know there there are sort of

01:05:16.760 --> 01:05:20.037
grooming gangs everywhere. There are
there's violence everywhere. This is a

01:05:20.120 --> 01:05:22.437
picture that people get from things that

01:05:22.520 --> 01:05:24.917
you post or that are posted on X. And

01:05:25.000 --> 01:05:27.357
for example, you you're posting for what

01:05:27.440 --> 01:05:30.157
was it? 48 hours of citizen vigilante.

01:05:30.240 --> 01:05:34.157
That film which, you know,
paints a dystopian I've watched it.

01:05:34.240 --> 01:05:36.357
>> Yes. >> It was not fun.
I watched it yesterday.

01:05:36.440 --> 01:05:39.877
It's not a very good movie. But it's
also paints a dystopian picture of

01:05:39.960 --> 01:05:44.677
Europe. And it glorifies vigilante
violence. It glorifies the murdering of

01:05:44.760 --> 01:05:47.797
a Muslim immigrant family.
I honestly think that it is

01:05:47.880 --> 01:05:50.077
irresponsible for you to put that up on X.

01:05:50.160 --> 01:05:52.437
>> I think that it's it's a movie.

01:05:52.520 --> 01:05:53.997
Um and it's worth watching.

01:05:54.080 --> 01:05:56.637
>> You're clearly suggesting Europe is

01:05:56.720 --> 01:05:59.917
dystopian place and you

01:06:00.000 --> 01:06:04.437
>> Yes, I mean why aren't you more

01:06:04.520 --> 01:06:08.157
doing more to fight censorship
in Europe? Why aren't you?

01:06:08.240 --> 01:06:11.557
>> I I >> seem to be doing anything
to fight censorship in Europe.

01:06:11.640 --> 01:06:14.117
>> I I think that there
are there are issues with

01:06:14.200 --> 01:06:18.557
>> in prison in the UK for social media
posts. Do you think that's appropriate?

01:06:18.640 --> 01:06:21.557
>> No, I mean we've written editorials
against it. But I don't think my point

01:06:21.640 --> 01:06:23.237
is with the UK.
>> of thing that I think is bad.

01:06:23.320 --> 01:06:26.637
>> think if you and I agree with you,
but I don't think you have to glorify

01:06:26.720 --> 01:06:29.277
vigilante violence and kind of, you

01:06:29.360 --> 01:06:34.157
know, post the things that you are doing
and paint an entirely inaccurate picture

01:06:34.240 --> 01:06:36.877
of Europe that you do. >> think it is.

01:06:36.960 --> 01:06:39.157
>> We should we should
do a tour of the UK.

01:06:39.240 --> 01:06:42.717
>> Yes, and one >> Cuz the UK
has actually been very good

01:06:42.800 --> 01:06:47.437
at integrating immigrants. The UK has
a is is very effective at doing that.

01:06:47.520 --> 01:06:50.157
There are problems. I'm the first
to admit that there are problems.

01:06:50.240 --> 01:06:51.277
>> Yes, there are major problems.

01:06:51.360 --> 01:06:52.677
>> There are problems, but it is not it

01:06:52.760 --> 01:06:55.837
really is not what you are

01:06:55.920 --> 01:07:00.477
describing to or suggesting to your 240
million followers it is. And I think

01:07:00.560 --> 01:07:04.797
there's a problem. >> No,
I think I'm I think I'm spot on.

01:07:04.880 --> 01:07:07.997
>> We'll we'll have to >> why you
have freedom of speech or at least

01:07:08.080 --> 01:07:10.357
>> You don't want people to can are wrong,
but your point is not [clears throat]

01:07:10.440 --> 01:07:13.477
that you're not just saying I have every
right to say what I want. You're saying,

01:07:13.560 --> 01:07:17.477
you know, I am right about Europe
and I really don't think you are.

01:07:17.560 --> 01:07:21.717
>> I think I am. >> I think
you will find that most people

01:07:21.800 --> 01:07:24.397
in Europe or many people in Europe don't
think you're right about it, but let's

01:07:24.480 --> 01:07:26.557
let's come to that. But >> Because
the media keeps ignoring what's

01:07:26.640 --> 01:07:31.197
ignoring reality. >> I don't think
that's true, either. I really I really

01:07:31.280 --> 01:07:32.837
>> by the TV and conventional media.

01:07:32.920 --> 01:07:37.237
>> but And why are you This is I guess
the even to me sort of the bigger picture

01:07:37.320 --> 01:07:40.877
question. Why do you
focus on this? Why are you

01:07:40.960 --> 01:07:45.197
interested in it? Because on the one hand,
you know, we started this conversation

01:07:45.280 --> 01:07:49.157
and you described to me this
extraordinary transformative

01:07:49.240 --> 01:07:55.037
amazing sort Yeah, civilization-changing
stuff that's happening in the next 10

01:07:55.120 --> 01:07:58.677
years. >> Yes. >> And yet
you're also, you know, a an

01:07:58.760 --> 01:08:02.837
ongoing participant
in the worst kind of tribalism

01:08:02.920 --> 01:08:05.357
uh you know, cesspit of social media.

01:08:05.440 --> 01:08:09.117
Like, how does that fit? Are you,
you know, on the one They're two very

01:08:09.200 --> 01:08:14.880
different views of the world. >> Yeah,
I don't think they're inconsistent.

01:08:15.200 --> 01:08:18.517
>> But why do you do the latter one,
the tribalism? I mean, people say that

01:08:18.600 --> 01:08:21.677
you're a racist. People say that you're,
you know, there's all kinds of things

01:08:21.760 --> 01:08:26.997
said about you, which I'm sure
you would disagree with. Um but

01:08:27.080 --> 01:08:29.437
I just don't understand what is the
motivation for doing that. Are you

01:08:29.520 --> 01:08:31.677
fermenting >> is is um

01:08:31.760 --> 01:08:36.837
she's half Indian and my my and I
have four children with her.

01:08:36.920 --> 01:08:40.437
Um you know, um

01:08:40.520 --> 01:08:42.077
one of them was named after an

01:08:42.160 --> 01:08:48.117
a famous Indian physicist. Um so I
I would say I'm not racist and also

01:08:48.200 --> 01:08:53.237
uh the you know, if you look at the people
that are employed at my companies,

01:08:53.320 --> 01:08:57.837
um you know, we have senior
executives of all races.

01:08:57.920 --> 01:09:00.877
Uh there's there's there's I don't
think there's any racism there.

01:09:00.960 --> 01:09:02.997
>> Are you anti-Muslim?

01:09:03.080 --> 01:09:06.197
>> Um I am

01:09:06.280 --> 01:09:10.200
I'm I'm against uh

01:09:10.960 --> 01:09:14.797
having if if if if people
are coming to a country

01:09:14.880 --> 01:09:19.077
with antithetical views

01:09:19.160 --> 01:09:23.157
um I I am against that.

01:09:23.240 --> 01:09:27.520
I I'm against that.
You know, I'm against um

01:09:28.440 --> 01:09:30.877
I I I'm against rape and murder.

01:09:30.960 --> 01:09:34.277
I'm against uh the

01:09:34.360 --> 01:09:37.717
imposition of

01:09:37.800 --> 01:09:41.877
rules and laws that are

01:09:41.960 --> 01:09:45.477
contrary to what we've

01:09:45.560 --> 01:09:52.477
come to accept in the West. Um and um

01:09:52.560 --> 01:09:56.477
and I think it's it's it's a crying
shame that the sort of traditional media

01:09:56.560 --> 01:10:00.197
like you um uh don't recognize this.

01:10:00.280 --> 01:10:04.517
>> First of all, I think
that you hugely exaggerate

01:10:04.600 --> 01:10:06.117
the challenges that Europe faces.

01:10:06.200 --> 01:10:10.077
Secondly, in areas I completely agree
with you. I I I think that, you know,

01:10:10.160 --> 01:10:13.437
the constraints on freedom of speech
in in the UK in certain areas are

01:10:13.520 --> 01:10:17.237
problematic. What worries
me is that the people you

01:10:17.320 --> 01:10:21.397
are supporting are really quite vile
people in some areas. Like some members

01:10:21.480 --> 01:10:23.277
of the AfD, Alice Weidel, whom I've met,

01:10:23.360 --> 01:10:26.277
is a perfectly thoughtful person. But in

01:10:26.360 --> 01:10:29.197
her party, are some real neo-Nazis. Are

01:10:29.280 --> 01:10:32.557
some really, really awful people. And

01:10:32.640 --> 01:10:37.197
your blanket support of this kind
of party with 240 million followers means

01:10:37.280 --> 01:10:40.837
that you have a power in Europe

01:10:40.920 --> 01:10:45.677
that I think perhaps you don't realize
the scale of impact you have. But you

01:10:45.760 --> 01:10:47.957
can >> I hope I hope
I'm having a big impact.

01:10:48.040 --> 01:10:51.957
>> But why? Why should you Why should you
be able to shape European politics? What

01:10:52.040 --> 01:10:56.557
You don't even live there. >> I think of
it as the sort of the West collectively.

01:10:56.640 --> 01:10:59.397
>> But do you see why
this makes people think

01:10:59.480 --> 01:11:02.997
that this man, who has all this power,
he's the world's richest man, he thinks

01:11:03.080 --> 01:11:07.157
he has a right to influence our politics.
It makes people It's why, you

01:11:07.240 --> 01:11:11.637
know, we started this conversation saying
superlatives, it's why people loathe you.

01:11:11.720 --> 01:11:12.717
I mean, some people do loathe you.

01:11:12.800 --> 01:11:15.917
>> Yes, some people do.
>> But do you understand why?

01:11:16.000 --> 01:11:18.797
Do you think what you're doing
is helpful for Western democracy?

01:11:18.880 --> 01:11:22.477
>> maybe some people do loathe me. I And
that's And that's probably true. I don't

01:11:22.560 --> 01:11:27.517
care. Um the uh but but the fact that uh

01:11:27.600 --> 01:11:30.997
as you pointed out a quarter
billion people follow me is

01:11:31.080 --> 01:11:35.437
that I think a lot more people
actually like me than don't.

01:11:35.520 --> 01:11:37.277
And I think a lot more people hate you

01:11:37.360 --> 01:11:41.357
and the media uh more than you realize.

01:11:41.440 --> 01:11:45.437
Do you realize the media is despised? Do
Do you realize that that journalists how

01:11:45.520 --> 01:11:50.277
the view of journalists is is is like

01:11:50.360 --> 01:11:53.837
it's it's like the a favorable
view of journalists is like 15%.

01:11:53.920 --> 01:11:57.797
>> in the United States. >> You So
so the shoe is on the other foot.

01:11:57.880 --> 01:12:01.717
They hate you. They hate you

01:12:01.800 --> 01:12:05.477
far more than they hate me. >> I
think there is certainly lots of trust

01:12:05.560 --> 01:12:10.237
in journalism. I think in part and there
have been mistakes by journalists and

01:12:10.320 --> 01:12:12.357
journalism. I totally acknowledge that.

01:12:12.440 --> 01:12:14.637
But I also think that the kind of

01:12:14.720 --> 01:12:19.557
polarized environment, particularly
on social media, fostered by the kinds of

01:12:19.640 --> 01:12:23.717
things you are pushing on X
actually helps this. Makes it worse.

01:12:23.800 --> 01:12:26.480
>> I don't think so.

01:12:27.480 --> 01:12:30.357
I think I think you're
the problem, the media.

01:12:30.440 --> 01:12:35.837
Um and And And speaking of loathing, you
are loathed far more than I am. And you

01:12:35.920 --> 01:12:37.277
don't seem to even realize that.

01:12:37.360 --> 01:12:40.237
>> I think very few people know who I am
relative to you, so I don't think I play

01:12:40.320 --> 01:12:43.437
in the same game. >> collectively. Um

01:12:43.520 --> 01:12:47.880
so I mean, you're hated.

01:12:48.080 --> 01:12:50.237
Do you even Do you even know that?

01:12:50.320 --> 01:12:53.757
You probably don't.
>> I I like to think that I'm aware of

01:12:53.840 --> 01:12:57.597
my own personal shortcomings and the
economist shortcomings and I'm our world

01:12:57.680 --> 01:13:00.077
view shortcomings. But what we stand for

01:13:00.160 --> 01:13:03.397
as you as you know is liberal values,

01:13:03.480 --> 01:13:08.077
traditional classical liberal values
where we worry about concentrations of

01:13:08.160 --> 01:13:11.197
power and we believe
in the individual and the

01:13:11.280 --> 01:13:16.037
dignity of the individual.
I and individual freedom. I firmly believe

01:13:16.120 --> 01:13:19.717
that and I have long thought that you
were a classical liberal. But when I

01:13:19.800 --> 01:13:22.477
read what you post classical liberal

01:13:22.560 --> 01:13:24.877
J.S. Mill, limited government,

01:13:24.960 --> 01:13:26.957
individual freedom.

01:13:27.040 --> 01:13:30.037
But when I read what you post

01:13:30.120 --> 01:13:34.397
on your feed, I sometimes wonder
whether you really do believe in that.

01:13:34.480 --> 01:13:35.277
Um

01:13:35.360 --> 01:13:38.917
I literally said, "Look, I believe in uh

01:13:39.000 --> 01:13:43.517
safe cities, secure borders,
and sensible spending

01:13:43.600 --> 01:13:44.877
with

01:13:44.960 --> 01:13:50.877
with respect to government." Um That
doesn't make me right-wing at all. Um

01:13:50.960 --> 01:13:55.477
in the least. >> Does the US or do
European countries score better on those?

01:13:55.560 --> 01:13:57.757
>> The US is better. >> No. >> Yes.

01:13:57.840 --> 01:14:02.557
>> Safe cities, much more dangerous
in the US than the UK. Sensible spending,

01:14:02.640 --> 01:14:06.877
bigger deficit here, out of control, you
managed you you you failed to reduce it.

01:14:06.960 --> 01:14:10.557
Secure borders, now under the Trump
administration, I agree with you, before

01:14:10.640 --> 01:14:13.917
not. But on two out of three,
unequivocally, Europe is in a better

01:14:14.000 --> 01:14:15.677
place. Yet you seem to paint a picture

01:14:15.760 --> 01:14:21.317
of Europe being a hellhole.
>> I don't think Europe's a hellhole.

01:14:21.400 --> 01:14:22.597
>> Okay, we can agree on something.

01:14:22.680 --> 01:14:24.037
>> When when did I ever say that?

01:14:24.120 --> 01:14:27.197
>> Well, the picture you paint of Europe
>> just you're making up things.

01:14:27.280 --> 01:14:32.037
>> No, I'm not. You you say you say
civil war in Britain is inevitable and

01:14:32.120 --> 01:14:36.037
then you you've you basically
proselytize for this citizen

01:14:36.120 --> 01:14:39.797
vigilante film which really does paint
Europe as a hellhole. It just starts

01:14:39.880 --> 01:14:42.437
with you know ominous
music and it says Europe.

01:14:42.520 --> 01:14:48.280
>> Well, there are there was out
of control rape and murder.

01:14:49.040 --> 01:14:52.357
>> It's more dangerous in this
country than it is in Europe.

01:14:52.440 --> 01:14:56.357
>> Uh where where there is danger
in this country is is because

01:14:56.440 --> 01:14:59.117
we we we fail to incarcerate

01:14:59.200 --> 01:15:01.957
uh repeat violent offenders.

01:15:02.040 --> 01:15:04.637
Um and that's uh you know, the classic

01:15:04.720 --> 01:15:09.197
sort of Soros DA uh or judges
that fail to incarcerate

01:15:09.280 --> 01:15:13.037
repeat violent offenders.
And if somebody gets arrested

01:15:13.120 --> 01:15:15.797
50 times, literally

01:15:15.880 --> 01:15:20.637
um for violent crimes, uh and and
eventually they're going to kill

01:15:20.720 --> 01:15:23.557
somebody. So you So you either you either

01:15:23.640 --> 01:15:26.277
incarcerate them or people will die.

01:15:26.360 --> 01:15:29.157
>> I think the prevalence of guns
has quite a lot to do with it, too.

01:15:29.240 --> 01:15:32.117
I mean, this is a much more dangerous
country. Murder rates are much higher in

01:15:32.200 --> 01:15:35.517
large part because there's so
much There are so many guns.

01:15:35.600 --> 01:15:38.437
>> Well, let me give you an interesting
number. There are more people that die

01:15:38.520 --> 01:15:41.917
in Europe from heat deaths than gun
than people that die from gun deaths in

01:15:42.000 --> 01:15:47.397
America. Why aren't you uh writing
an article about air conditioning?

01:15:47.480 --> 01:15:49.637
>> Well, we've done several on air
conditioning, and we had on our cover

01:15:49.720 --> 01:15:54.557
last week um a whole a whole long
piece about, you know, why the

01:15:54.640 --> 01:15:56.277
>> Pro-regulation? >> Yeah,
we're very pro-regulation.

01:15:56.360 --> 01:15:57.997
>> Okay, good. Well, good. >> Very
pro-regulation. You should read us.

01:15:58.080 --> 01:16:00.357
Absolutely. I'm I I think I I think you

01:16:00.440 --> 01:16:04.357
have a a a view of The Economist that

01:16:04.440 --> 01:16:07.757
is not entirely accurate. So, I will

01:16:07.840 --> 01:16:11.797
You come look at England, start
reading The Economist, and I will um

01:16:11.880 --> 01:16:18.837
you know, stand corrected
on my criticism of you in this area.

01:16:18.920 --> 01:16:20.277
>> The There There are

01:16:20.360 --> 01:16:23.197
There are many parts of Britain um

01:16:23.280 --> 01:16:27.397
and Europe which are not a

01:16:27.480 --> 01:16:30.317
safe. You You You can't just walk around.

01:16:30.400 --> 01:16:32.397
>> But there are far more in this country.

01:16:32.480 --> 01:16:34.997
Far more. >> No, I don't think so.

01:16:35.080 --> 01:16:37.357
>> All right, let's go and get
my fact-checkers to check then.

01:16:37.440 --> 01:16:39.237
>> Yeah, I don't I
don't think that's true.

01:16:39.320 --> 01:16:43.997
Um >> Let's move on from this because
uh I think I've got a sense of why you're

01:16:44.080 --> 01:16:46.997
doing what you're doing.
>> The point is not that that that the

01:16:47.080 --> 01:16:50.037
question is not is that is is
America perfect? Question is

01:16:50.120 --> 01:16:52.717
does Europe have a serious
and growing problem?

01:16:52.800 --> 01:16:54.157
>> is which is which is more dangerous.

01:16:54.240 --> 01:16:57.397
That's where we started this. >> Europe
has a serious and growing problem.

01:16:57.480 --> 01:17:01.077
I'm I'm I'm not going to say
that the United States has no

01:17:01.160 --> 01:17:05.477
problems. Of course it does. Um but

01:17:05.560 --> 01:17:08.557
you know, we we should all
try to to address these issues.

01:17:08.640 --> 01:17:11.477
>> We can agree on that. We should all
try to address the problems. My point was

01:17:11.560 --> 01:17:14.637
simply that I think you exaggerate
the scale of Europe's problem and that

01:17:14.720 --> 01:17:18.997
>> I think you people are asleep at the
switch and do not realize the magnitude

01:17:19.080 --> 01:17:20.837
of the issue.

01:17:20.920 --> 01:17:23.317
>> I also think that your interventions,

01:17:23.400 --> 01:17:28.357
however well-intentioned they may be,
risk making Europe's problem a lot worse.

01:17:28.440 --> 01:17:32.437
>> How? >> Because you are putting
your support behind people who are

01:17:32.520 --> 01:17:36.997
proto-authoritarians, who are
not trying simply to you know,

01:17:37.080 --> 01:17:38.997
make sure there is better integration of

01:17:39.080 --> 01:17:41.197
Muslims or to to secure the borders, but

01:17:41.280 --> 01:17:46.517
who have very clear fascist tendencies
and you will cause huge divisions within

01:17:46.600 --> 01:17:48.517
Europe. >> I think what
you're saying is total

01:17:48.600 --> 01:17:54.117
fiction. Um and really I I'm
simply saying that there needs to be

01:17:54.200 --> 01:17:57.477
there needs to that Europe
has to have secure borders.

01:17:57.560 --> 01:18:00.437
>> Okay. >> and so not let
in vast numbers of people

01:18:00.520 --> 01:18:03.997
who with views that are antithetical to

01:18:04.080 --> 01:18:08.437
the accepted views in in Europe and and
where they will where they will not be

01:18:08.520 --> 01:18:13.277
some level of uh safe cultural
integration. This is simply common sense.

01:18:13.360 --> 01:18:15.997
>> If all you are saying is Europe should
have secure borders, I completely agree

01:18:16.080 --> 01:18:20.557
with you. Then then I have
misunderstood what your

01:18:20.640 --> 01:18:22.357
interventions have been on social media

01:18:22.440 --> 01:18:27.877
cuz that is not what the view I have not
what I would conclude having seen what

01:18:27.960 --> 01:18:30.237
the people you are supporting and the
things you're doing. But if that's your

01:18:30.320 --> 01:18:33.437
goal, I share your goal that Europe
should have secure borders. I think

01:18:33.520 --> 01:18:36.557
that's perfectly agreeable.
That gets me to a question.

01:18:36.640 --> 01:18:40.277
>> And let me let me explain what what
the fundamental forcing function here is

01:18:40.360 --> 01:18:43.517
uh for why the borders
are not secure, which is

01:18:43.600 --> 01:18:48.317
uh When you have a welfare state that

01:18:48.400 --> 01:18:52.277
provides substantial
benefits uh to anyone who

01:18:52.360 --> 01:18:56.317
comes there, um you create
a very powerful economic

01:18:56.400 --> 01:18:59.797
forcing function for anyone
who is below that standard of

01:18:59.880 --> 01:19:02.517
living to come to Europe.

01:19:02.600 --> 01:19:08.517
Um it is it is a an obvious
powerful driving force.

01:19:08.600 --> 01:19:11.717
Um and um people will come there if if

01:19:11.800 --> 01:19:16.877
their standard of living is uh
from what they get in Europe is above

01:19:16.960 --> 01:19:18.717
where they are elsewhere in the world.

01:19:18.800 --> 01:19:21.597
It's very basic. >> agree with that.
I don't I don't disagree.

01:19:21.680 --> 01:19:24.597
>> very basic incentive structure. Um

01:19:24.680 --> 01:19:27.877
>> Economic conditions, economic

01:19:27.960 --> 01:19:30.757
influences determine a lot of migration.

01:19:30.840 --> 01:19:34.277
I think asylum is a separate issue,
but I think you I think you're absolutely

01:19:34.360 --> 01:19:38.077
right. There are there are um pressures
for migration. However, there are other

01:19:38.160 --> 01:19:40.437
things that you haven't focused on,
which is migration brings a lot of

01:19:40.520 --> 01:19:43.797
benefits. And it also particularly does
so in a continent like Europe, which is

01:19:43.880 --> 01:19:46.677
a rapidly aging continent. So, we can

01:19:46.760 --> 01:19:51.037
agree that completely uncontrolled
borders are a bad idea.

01:19:51.120 --> 01:19:56.277
But, I am I think Europe
benefits from having

01:19:56.360 --> 01:19:58.357
immigration, controlled
immigration. It brings

01:19:58.440 --> 01:20:00.637
>> Yes, I I I'm in favor of immigration.

01:20:00.720 --> 01:20:03.317
Obviously, I'd be hypocritical if I
wasn't since I am an immigrant to the

01:20:03.400 --> 01:20:06.437
United States.
Um I'm in favor of immigration if the

01:20:06.520 --> 01:20:10.877
person immigrating um is going
to be a productive member of society,

01:20:10.960 --> 01:20:14.877
um if they're going to be honest, and if
they're not going to mistreat people in

01:20:14.960 --> 01:20:17.837
that country they're going to.

01:20:17.920 --> 01:20:19.117
>> I can agree on all of those.

01:20:19.200 --> 01:20:21.757
>> Yes. >> But, I think
Europe does better on that

01:20:21.840 --> 01:20:24.997
than you give it credit for. That's I
think where we disagree. But, let's move

01:20:25.080 --> 01:20:28.957
on. Tell me what do you think
people get most wrong about you?

01:20:29.040 --> 01:20:33.237
>> Well, I mean, you got a you just got
a whole bunch of things wrong about me.

01:20:33.320 --> 01:20:38.200
Pretty much everything in the last Just
rewind the last 20 minutes is all wrong.

01:20:38.560 --> 01:20:42.317
>> So I got we got I got
wrong [laughter] your your

01:20:42.400 --> 01:20:44.317
>> Let's not be repetitive here.

01:20:44.400 --> 01:20:47.437
>> Okay. So So that political
side of you people get most wrong.

01:20:47.520 --> 01:20:49.917
>> Yeah, my political views are very are

01:20:50.000 --> 01:20:52.757
centrist. They're not

01:20:52.840 --> 01:20:55.477
far right. And frankly calling it far

01:20:55.560 --> 01:20:59.197
right I think is an insult uh

01:20:59.280 --> 01:21:03.237
to to history, an insult to many peoples,

01:21:03.320 --> 01:21:04.517
and it is

01:21:04.600 --> 01:21:07.917
an outrageous falsehood. Um

01:21:08.000 --> 01:21:11.997
It's It's completely absurd. I I
literally have stated my position, which

01:21:12.080 --> 01:21:15.077
is that And And I've been very clear
about this, which is that there need to

01:21:15.160 --> 01:21:19.557
be secure borders, that um If If
there's If there's an influx of

01:21:19.640 --> 01:21:21.277
immigrants, they they they they need to

01:21:21.360 --> 01:21:24.997
be people who will agree that the with

01:21:25.080 --> 01:21:29.237
with with Western ideals, with Western
values. And if they don't, and if And if

01:21:29.320 --> 01:21:31.637
there's a long-term uh

01:21:31.720 --> 01:21:34.717
growth of people who do not or or in

01:21:34.800 --> 01:21:36.757
some cases are adamantly opposed to

01:21:36.840 --> 01:21:39.277
Western values, you are setting yourself

01:21:39.360 --> 01:21:42.277
up for civil war. This is obvious.

01:21:42.360 --> 01:21:45.397
A child can see this.
So So But this is This is not a

01:21:45.480 --> 01:21:48.157
right-wing view.
I'm simply stating reality.

01:21:48.240 --> 01:21:52.797
>> Well, I think we can both agree that
evidence will be able to decide which of

01:21:52.880 --> 01:21:56.157
us is more or less right. And we will
find evidence for both of our views in

01:21:56.240 --> 01:21:58.597
different parts of Europe.
The question is what the overall is.

01:21:58.680 --> 01:22:00.237
>> I'm very good at predicting the future.

01:22:00.320 --> 01:22:01.677
I'm not perfect, but I'm very good at

01:22:01.760 --> 01:22:03.677
predicting the future. Uh my batting

01:22:03.760 --> 01:22:07.797
average is is very high for a human. Um

01:22:07.880 --> 01:22:09.757
But this this does have somewhat of of a

01:22:09.840 --> 01:22:12.957
Cassandra effect, where where people

01:22:13.040 --> 01:22:19.840
don't believe that what I'm saying
is likely to be true um because

01:22:20.360 --> 01:22:24.837
you know, I I I can see into the future
with a higher bet not perfectly, but

01:22:24.920 --> 01:22:28.837
with a higher probability
of success than most people.

01:22:28.920 --> 01:22:32.237
So, and including things
that people believe to be

01:22:32.320 --> 01:22:34.077
in some cases counterintuitive.

01:22:34.160 --> 01:22:37.517
Um and uh again, I'm not saying I'm always

01:22:37.600 --> 01:22:42.397
right. I certainly am not.
But but my batting average is very good.

01:22:42.480 --> 01:22:46.597
Um But but this this
this does have, like I

01:22:46.680 --> 01:22:50.477
said, a Cassandra effect where people
don't realize that what I'm saying will

01:22:50.560 --> 01:22:54.157
come to pass. Um

01:22:54.240 --> 01:22:58.277
Maybe not exactly in the timing that I
thought, but it will come to pass.

01:22:58.360 --> 01:23:04.397
Um and uh Yeah.

01:23:04.480 --> 01:23:08.560
The history will write the verdict.

01:23:08.680 --> 01:23:11.317
But I think you will see that

01:23:11.400 --> 01:23:13.637
very little of what I've said

01:23:13.720 --> 01:23:15.637
um is inaccurate.

01:23:15.720 --> 01:23:22.320
Um and a supermajority of what I've said

01:23:22.560 --> 01:23:24.997
is very accurate and will come to pass.

01:23:25.080 --> 01:23:27.237
Unless there are there's
some change in direction.

01:23:27.320 --> 01:23:28.917
>> Do you have higher confidence level in

01:23:29.000 --> 01:23:33.597
your expectations of the future for AI,

01:23:33.680 --> 01:23:35.957
the first part of our conversation, or

01:23:36.040 --> 01:23:39.357
the latter part, the political part?

01:23:39.440 --> 01:23:42.600
>> Um the

01:23:43.040 --> 01:23:46.557
I mean, AI is digital superintelligence

01:23:46.640 --> 01:23:49.997
is called the singularity,
like a black hole, because

01:23:50.080 --> 01:23:53.557
you don't know what's going to happen
after that. It's and it sort of sucks in

01:23:53.640 --> 01:23:56.517
everything. So, I think

01:23:56.600 --> 01:23:58.477
really

01:23:58.560 --> 01:24:03.720
in terms of the the really
big macro effects here,

01:24:04.280 --> 01:24:08.520
AI and robots will really
dominate everything.

01:24:09.400 --> 01:24:14.757
So, on on a on a time scale
that is less than 10 years.

01:24:14.840 --> 01:24:17.437
>> Which probably renders this other
stuff that we've been talking about less

01:24:17.520 --> 01:24:21.957
important and less relevant. >> The the AI

01:24:22.040 --> 01:24:27.920
AI robotics revolution will happen

01:24:29.200 --> 01:24:32.357
sooner than it will it will
happen sooner than civil

01:24:32.440 --> 01:24:36.317
war in in in in Britain.
I would say civil war in Britain is

01:24:36.400 --> 01:24:40.157
probably 20 years away. Um

01:24:40.240 --> 01:24:44.877
and the AI robot singularity

01:24:44.960 --> 01:24:47.797
is 10 years away. >> Well,
hopefully these benign all

01:24:47.880 --> 01:24:51.317
powerful AIs will prevent such outcomes.

01:24:51.400 --> 01:24:54.477
>> They probably will. >> Eamonn I
don't agree with everything you say,

01:24:54.560 --> 01:24:56.837
but it's always absolutely [music]
fascinating to talk to you. Thank you

01:24:56.920 --> 01:24:59.557
very much indeed. >> Well,
you're welcome. I'm not boring.

01:24:59.640 --> 01:25:02.760
>> You're definitely not boring.