00:00:00
All right everybody, emergency podcast
00:00:02
time. Episode 263 of All-In. We have
00:00:04
Emil Michael, the under secretary of war
00:00:08
for research and engineering working
00:00:10
directly for Pete Hexa. We had to get
00:00:13
this out to you on Thursday night
00:00:15
because it is an emergency pod. One of
00:00:18
my old besties. Emil Michael is here.
00:00:20
Emil and I uh were part of team Uber
00:00:23
back in the day. He was Travis's
00:00:25
right-hand man. Some might say fixer.
00:00:28
And Emil Michael is now the under
00:00:30
secretary for war
00:00:33
here in the United States, serving his
00:00:35
country like our bestie David Sachs.
00:00:38
Welcome to the program for the first
00:00:39
time. Neil Michael, how you doing,
00:00:41
brother?
00:00:41
>> I'm doing good. Uh I hope it was more
00:00:43
than a fixer, but you know,
00:00:46
>> raising 20 billion fixer. I mean, you
00:00:48
got it done. You got it done. The
00:00:50
hardest. He would give you the hardest
00:00:51
things. Yeah. Just
00:00:52
>> That's right. Fair enough.
00:00:53
>> If it was hard, and that's what a fixer
00:00:55
is.
00:00:55
>> An operational axe. That's what they
00:00:58
call them.
00:00:58
>> All right. Sure. Uh, in Brooklyn, we
00:01:00
call them fixers. With us again,
00:01:01
>> a rain maker.
00:01:03
>> There's that, too. There's that, too.
00:01:05
Making it happen. With us again. Chim
00:01:08
Polyhapatia. How are you, brother?
00:01:11
>> Great.
00:01:11
>> Yeah. Look at that smile. What do you
00:01:14
got going on? You got some pokers in the
00:01:17
fire.
00:01:18
>> Okay.
00:01:19
>> I'm not going to say in the coming
00:01:20
weeks, I think some news is going to
00:01:22
drop. That's my prediction. I don't have
00:01:23
any.
00:01:23
>> Are you loving tweet mogging that's been
00:01:25
going on this week? So good. So good.
00:01:28
>> Oh my god.
00:01:29
>> So good.
00:01:30
>> He's looks maxing by default, but he's
00:01:32
been mogging
00:01:33
>> the Gooners.
00:01:35
>> Yeah. So funny. What was your favorite?
00:01:38
>> The one I sent you this morning that you
00:01:39
said. What did you say? So funny.
00:01:41
>> Are you collecting your losses? Tax
00:01:45
harvesting.
00:01:45
>> What did you say? Oh my god.
00:01:47
>> Chimat said uh Oh my god. It was just
00:01:50
like Yes, I did. Yes, I did.
00:01:51
>> Yes, I did. Yes. Someone said something
00:01:53
to Chabat. He's like dropped in. Why is
00:01:55
everyone so mad at Jamoth? All he did
00:01:57
was lose billions in retail investors
00:01:58
money by proep one page spaxs. It's not
00:02:02
like he then told them to enjoy their
00:02:03
capital losses or anything. Give the man
00:02:05
a break. Jama's response. Yes, I did.
00:02:08
>> All right. Piling on is your sultan of
00:02:11
science. Everybody's favorite. Had a
00:02:13
great
00:02:15
>> some great science that he brought to
00:02:17
the show last week. Freedberg, how are
00:02:18
you doing?
00:02:20
>> Oh, yeah. I've been traveling this week
00:02:21
back at home.
00:02:22
>> All right. Sax is out today. He's very
00:02:24
busy on Capitol Hill. We'll talk about
00:02:26
what he's up to next week.
00:02:27
>> Let's go. Come on. Let's go. Let's go.
00:02:29
>> Let's go. Go, Jason.
00:02:30
>> All right. The US and Israel launched a
00:02:31
joint attack on Iran on Saturday. Today
00:02:34
is day six of Operation Epic Fury. Iran
00:02:38
Supreme Leader Ali Hameni
00:02:42
was killed within hours of the
00:02:44
operation. 40 senior officials have also
00:02:46
been killed. Death toll so far. About a
00:02:48
thousand people according to reports.
00:02:50
Tragically, six US Army Reserve soldiers
00:02:53
were killed following a drone strike on
00:02:55
a base in Kuwait. A US submarine sank an
00:02:58
Iranian ship off the coast of Sri Lanka.
00:03:00
This is the first torpedo kill since
00:03:02
World War II. Why were at war been a bit
00:03:05
of a moving target in a debate. First
00:03:07
explanation from Rubio. He said Israel
00:03:09
was going to attack and the US had no
00:03:12
choice but to participate. Later walked
00:03:14
that back. Trump made it clear this is
00:03:17
not a regime change effort, but we're
00:03:20
doing this to stop terrorism and the
00:03:22
development of ICBMs by obviously a
00:03:26
pretty crazy group of individuals and
00:03:29
obviously nuclear bombs which we blew up
00:03:31
a couple weeks ago. Trump also mentioned
00:03:33
the people of Iran should seize the
00:03:35
moment quote and take their country
00:03:38
back. Pegsth believe is sure boss meal
00:03:41
said quote this is not a so-called
00:03:43
regime change war but the regime sure
00:03:46
did change and the world is better off
00:03:48
for it so here's an interesting poly
00:03:50
market right now US forces enter Iran
00:03:55
this is boots on the ground by the end
00:03:57
of March 40% chance by the end of the
00:04:00
year 59% chance so the idea that we're
00:04:02
not going to have boots on the ground
00:04:04
the sharps on poly market believe we
00:04:06
will will the Iranian regime fall by
00:04:11
June 30th 39% chance according poly
00:04:13
market and by the end of the year 51%
00:04:16
chance so Emil I guess there are two
00:04:19
questions people really want to know
00:04:21
I'll leave off why we're doing this I
00:04:22
think President Trump has been pretty
00:04:24
clear now but how long is this going to
00:04:27
take is the one question and are we
00:04:29
going to have to have boots on the
00:04:30
ground maybe what is success here
00:04:35
>> um I think the the president tal talked
00:04:38
about this is a weeks not months kind of
00:04:41
operation and it's aimed at essentially
00:04:45
disarming the the regime uh or the
00:04:48
country from uh in such a way that they
00:04:52
can't supply Hezbollah, Hamas, um Muslim
00:04:56
Brotherhood, all the kind of terror
00:04:58
groups that get sponsored by weapons and
00:05:00
money from Iran, not to mention the
00:05:02
nuclear bit. And that's why you see from
00:05:06
the reporting they're going after the
00:05:08
depots the the you know we went after
00:05:11
nuclear sites before they're a
00:05:14
prodigious drone maker. These like huge
00:05:16
one-way attack drones that can go hund
00:05:19
you know hundreds and hundreds of miles.
00:05:21
Um lots of ballistic missiles uh that
00:05:25
are aimed at every country in the Middle
00:05:27
East as you've seen they've attacked
00:05:28
them. So
00:05:30
I think that's one. In terms of boots on
00:05:32
the ground, there's no scenario where we
00:05:34
have some protracted boots on the ground
00:05:37
Afghanistan, Iraq 2 like scenario.
00:05:40
>> Freeberg, your thoughts on this war.
00:05:43
Obviously, a lot of people voted for
00:05:47
Trump in order to have the peace
00:05:49
dividend that he was in his first term,
00:05:52
absolutely the peace president. And now
00:05:54
here we are. Eight countries have been
00:05:56
bombed and we've had two leaders deposed
00:05:59
and one of those two have been killed.
00:06:02
Your thoughts, Freeberg?
00:06:03
>> I think the president and the
00:06:05
administration have probably the biggest
00:06:07
meetings of the term coming up in China
00:06:10
in April. My
00:06:13
estimation based on the conversations
00:06:14
and the comments made by the president
00:06:17
before he came into office and since
00:06:19
he's been in office is that finding a
00:06:21
grand bargain or a deal with China is
00:06:23
probably one of his top priorities. And
00:06:25
if you think about the importance of
00:06:27
that, is the US going to wade into a
00:06:30
giant global conflict led by a US China
00:06:34
rift or is the US going to find some
00:06:36
grand bargain? I think he would probably
00:06:38
have a preference for the grand bargain.
00:06:40
And that being the case, I think you
00:06:42
could look at the in the context of
00:06:44
Maduro and the actions in Iran as
00:06:47
creating maximal leverage going into
00:06:49
those negotiations that
00:06:51
>> the reason for that free bird
00:06:53
>> 90% of the oil that comes out of Iran
00:06:54
goes to China
00:06:57
and there's been a long developing and
00:06:59
developed relationship between Maduro's
00:07:02
government and China and these are big
00:07:06
economic drivers or support the economic
00:07:09
driving in China. So creating leverage
00:07:12
by having significant influence or
00:07:15
damage or destruction to those supply
00:07:17
chains for China gives the United States
00:07:20
footing to be able to negotiate a better
00:07:22
deal for America. I would imagine that
00:07:24
the president's intention here isn't to
00:07:25
go and decide who should be in charge
00:07:28
and drive regime change and end in a
00:07:29
multi-year conflict with Iran. But
00:07:32
ultimately, if there's some transaction
00:07:34
with China that gets everyone out of
00:07:36
this and puts the US on a strong footing
00:07:39
where American businesses can sell into
00:07:40
China, which is very challenging as
00:07:42
everyone knows today. And there's
00:07:44
parity, regulatory par, economic and
00:07:47
trade parity between the US and China.
00:07:49
there's a point of view on what happens
00:07:50
with Taiwan and availability of key
00:07:52
technologies like semiconductors. I
00:07:54
think it could be a win-win and I think
00:07:57
that a deal with China could be the
00:07:59
crowning achievement of this
00:08:00
administration particularly going into
00:08:02
the midterms. So the timing is right and
00:08:04
I think that's probably a core part of
00:08:06
the motivation here. Chimath, your
00:08:08
thoughts on this action and why we're
00:08:12
doing it. You've heard obviously the
00:08:14
president has his position. We're not
00:08:16
doing regime change. It's a secondary
00:08:18
effect obviously, but we want to stop
00:08:20
those ICBMs and nuclear bombs from being
00:08:22
developed and we want to stop terrorism.
00:08:25
Additionally, Freeberg says, "Hey, we're
00:08:27
framing this great, you know, discussion
00:08:32
we're going to have with she and China
00:08:34
and oil is part of that." Where where do
00:08:36
you think you stand on all this?
00:08:38
>> I'll build on both what Emil said and
00:08:40
what Freeberg said. I don't think this
00:08:43
is about regime change and I don't think
00:08:45
it's about a local regional conflict. I
00:08:49
think if you take a step back and zoom
00:08:51
out, the most important thing that we
00:08:54
did in the last 3 months was by taking
00:08:58
out Maduro and by taking out the Iranian
00:09:04
leadership, we have created enormous
00:09:06
leverage as Freedberg said
00:09:09
with China. Now, why is that important?
00:09:12
Because I think all of this centers
00:09:14
around that geopolitical discussion.
00:09:17
Last night, something important
00:09:19
happened, which is that the official
00:09:20
Chinese
00:09:22
bureaucracy posted what their GDP
00:09:25
targets were, and it was shocking to
00:09:27
anybody reading it because what we saw
00:09:30
was that they guided to a range of 4.5
00:09:33
to 5%.
00:09:35
which if you look at the historical
00:09:37
context of that growth is the lowest
00:09:40
that it has ever been in about 30 years.
00:09:44
So three decades. So before they entered
00:09:47
the WTO
00:09:48
and the question that one should ask
00:09:50
yourself is when a country that's
00:09:51
growing at 8 9 and 10% start to grow at
00:09:54
half that rate yet have double the
00:09:57
number of people and double the GDP what
00:09:59
happens? you already have incredibly
00:10:02
high domestic unemployment, especially
00:10:04
youth unemployment.
00:10:06
Does it become more or less chaotic? And
00:10:08
I think the historical artifacts of
00:10:09
every other country would show that it
00:10:12
will become more chaotic. If you have
00:10:13
that as a starting point, what is it in
00:10:16
China's best interest to do? And I think
00:10:18
it becomes obvious that the right thing
00:10:19
to do would be to invade Taiwan. Why?
00:10:23
because you start to create a sinkhole
00:10:25
that occupies your people, that occupies
00:10:27
resources, that can get domestic
00:10:29
production up and running, that can
00:10:31
start to generate a war machine. And you
00:10:33
see the economic impact of war machines
00:10:35
in any country during any conflict.
00:10:38
And if I had to guess, just to build on
00:10:40
what Emil said, the president saw that
00:10:42
and I think what they did can be
00:10:44
summarized in this chart which I sent to
00:10:46
Nick. So if your goal is to prevent war
00:10:50
with China, which is a massive global
00:10:53
conflict, which could be nuclear, which
00:10:56
could be cataclysmic,
00:10:59
how would you do it? And this chart
00:11:02
paints one way to do it. If you look at
00:11:04
the conditions inside of the Chinese
00:11:06
economy,
00:11:07
the most interesting takeaway is that
00:11:10
they are enormously
00:11:12
dependent on imported oil. So about 20%
00:11:17
of their economy, but it's not 20% of
00:11:19
their economy because it's 100% of these
00:11:22
critical things that create GDP,
00:11:24
logistics, transportation, aviation,
00:11:26
feed stock inputs.
00:11:29
And of that 19%
00:11:32
about a fifth of it comes exclusively
00:11:34
from Iran and Venezuela. And now all of
00:11:37
that is off the table.
00:11:39
So if you take that and then you see
00:11:41
what Steve Whitoff and Jared Kushner and
00:11:43
Josh Greenbond have been doing, which is
00:11:46
trying to get a deal done in Russia, and
00:11:47
you put all of these things together,
00:11:49
because by the way, if you add Russia
00:11:50
into that mix, it's about 40% of China's
00:11:53
oil, not only do you red dollarize, not
00:11:56
only do you stop the funneling of all of
00:12:00
these illicit oil funds to creating
00:12:02
chaos all around the world, but you hem
00:12:05
in China going into a massive moment at
00:12:08
the end of March, beginning of April,
00:12:10
where as Freeberg said really astutely,
00:12:13
there is the potential for a grand
00:12:14
bargain and I think that secures global
00:12:17
safety in that that is a huge thing for
00:12:19
America.
00:12:20
>> Emil, how much does this have to do with
00:12:22
China?
00:12:24
I think you know my instinct is and I'm
00:12:27
not speaking for the administration on
00:12:29
this is that's a second order benefit to
00:12:32
some of these things like the um you
00:12:35
know and you said eight eight conflicts
00:12:37
there have not been eight conflicts
00:12:39
there is like we inherited Gaza we
00:12:42
inherited Russia Ukraine um Venezuela
00:12:45
was its own operation and then you could
00:12:48
sort of attach to it the drug boats that
00:12:50
were coming out of that as like one big
00:12:51
operation
00:12:53
um and then the Houthies was just Biden
00:12:55
was ignoring the Houthies. They were
00:12:56
just shooting at our ships. So that was
00:12:58
like very limited in terms of like stop
00:13:01
shooting our ships. We need freedom of
00:13:03
the seas and that wasn't sort of a, you
00:13:06
know, so that's something any president
00:13:08
should be doing generally. I think um
00:13:11
Iran being the one, you know, material
00:13:14
conflict outside of Venezuela. So it's
00:13:16
not it's not that many. And and how long
00:13:18
did Venezuela last? It was one raid.
00:13:21
one.
00:13:21
>> I guess that's a really important
00:13:23
>> few hours. This is this is an important
00:13:25
note I think for you Emil to sort of uh
00:13:28
>> I've explained to us there's a new
00:13:31
>> approach here with regard to the these
00:13:35
actions which is no boots on the ground
00:13:38
and we seem to uh and and you of course
00:13:41
have better information than anybody
00:13:42
else does. I don't think anybody would
00:13:44
have known Venezuela would have gone as
00:13:46
well as it did and so far and listen we
00:13:49
got a long way to go with Iran. This has
00:13:51
gone very well as as well. So explain to
00:13:54
us what you know and what you the
00:13:57
president and HGV know that we don't
00:14:00
that makes these two operations go so
00:14:02
smoothly. What what is it then? There's
00:14:05
obviously some new technology here in in
00:14:06
the case of what happened in Venezuela.
00:14:08
>> Yeah. Bes besides the discombobulator,
00:14:11
what we've got is a very well-trained
00:14:16
military. Like the the global war and
00:14:18
terror was disaster in so many respects,
00:14:23
but the people now who are fighting that
00:14:25
are generals now. And so they've learned
00:14:28
a lot of lessons. And you compare that
00:14:30
to the Chinese military. They don't have
00:14:32
a lot of experience. In fact, the the
00:14:34
decapitation they did in the Chinese
00:14:36
military, the one guy they took out was
00:14:37
the one guy who had experience in
00:14:39
Vietnam. So, they don't have conflict
00:14:41
experience, and that matters because you
00:14:43
understand going in what are the things
00:14:46
that could go wrong. Um, and then you
00:14:48
you have incredible technology, space,
00:14:51
air, land, sea, cyber, um, all kinds of
00:14:56
effects that you could bring together.
00:14:58
And so you imagine uh a hundred guys
00:15:01
goes into the most fortified compound in
00:15:03
Venezuela where the president is, you
00:15:06
know, take him and his wife out safely
00:15:08
and are out with no KAS.
00:15:12
>> Incredible.
00:15:12
>> I mean, it's just incredible, right? So
00:15:14
stunning.
00:15:14
>> Yeah. So, and they they these things
00:15:17
these war games have been on the on the
00:15:18
shelf for a long time. Every every
00:15:20
scenario has been planned for years
00:15:23
ahead of time. Midnight hammer in Iran
00:15:25
was planned years ahead of time in terms
00:15:27
of how would you do it if you were going
00:15:29
to do it.
00:15:30
>> Um, and then you keep refreshing the
00:15:32
tactics, tactics, techniques, and
00:15:34
procedures and you're updating them. So,
00:15:36
we have a very sophisticated way of of
00:15:38
doing these things to minimize loss of
00:15:40
life and maximize success.
00:15:42
>> Can I ask a question? I don't want to I
00:15:43
don't want to derail this conversation,
00:15:44
but is the discombobulator real? Like,
00:15:47
what can you say about the discombobul?
00:15:50
>> I completely I was like obsessed with
00:15:52
this when I saw it on X. I was like,
00:15:54
"What is this thing?" I mean, I need it
00:15:56
in my house. Like, I just want to push a
00:15:58
button of all these my kids.
00:15:59
>> And that's just for when Helm Youth
00:16:01
shows up.
00:16:01
>> Oh my god.
00:16:02
>> Not meant for your kids or if they're
00:16:04
behaving badly. No, it's can't talk
00:16:06
about it.
00:16:06
>> Emil, do you think we would have been
00:16:07
able to pull off that mission as
00:16:09
successfully as we did 5 years ago, 10
00:16:11
years ago? Has the technology improved
00:16:14
that quickly that this is not something
00:16:16
that's been possible historically? And
00:16:17
how does that change the the pacing and
00:16:20
the face of war for the next couple of
00:16:22
years? I'd say no. It wasn't only a
00:16:25
technology maturation from five years
00:16:28
ago. It's the uh rules of engagement.
00:16:31
The rules of engagement that we used to
00:16:34
have uh there was some I mean if you
00:16:36
read about them some of them were insane
00:16:37
like if in Afghanistan if the guy had a
00:16:40
small gun you had to have a small gun
00:16:43
and you know there was this parody in
00:16:45
weird ways.
00:16:46
>> And when you're like well but is the
00:16:48
objective to have like a fair fight or
00:16:50
an unfair fight? Um, well, if you're on
00:16:52
our side, you want it to be unfair. So,
00:16:54
the rules of engagement were relaxed to
00:16:57
be
00:16:57
>> Who writes those, Emil? Who sits in an
00:16:59
office and says you can't shoot back if
00:17:01
a combatant is shooting at you if you
00:17:04
aren't matched gun forgun?
00:17:06
>> Yeah. I mean, who writes that crazy
00:17:08
policies that are written into military
00:17:10
department? And that's why when
00:17:12
Secretary Hicks talks about um this kind
00:17:16
of thing and what was happening with him
00:17:18
when he was in Afghanistan, if you ever
00:17:19
read his book in Iraq, he's like the
00:17:22
rules of engagement were so punishing
00:17:24
that we were we were at risk all the
00:17:26
time cuz you had to have like a legal
00:17:28
understanding of what was happening in
00:17:30
every minute in the battlefield as
00:17:32
opposed to well your job is to you know
00:17:34
take out these guys and protect these
00:17:36
guys. Here's your munitions. here's like
00:17:38
the the the red lines and then like in
00:17:41
the middle of that go use your judgment
00:17:43
your commanding officer use your
00:17:45
judgment on how to win and we kind of
00:17:47
gotten back to that use your judgment
00:17:50
push responsibility field still have
00:17:52
your red lines but other than that the
00:17:53
objective is the objective it's more of
00:17:55
a colon Powell approach it's like go all
00:17:58
in have a clear objective come out use
00:18:01
overwhelming force and we were not doing
00:18:03
that for the last four years
00:18:05
>> and then going back to the the face of
00:18:07
war going forward forward. My
00:18:08
understanding is that there have been
00:18:10
more drones deployed by the United
00:18:12
States this past week than we've done in
00:18:14
the history of military activity. Is
00:18:17
that right? And like how does that
00:18:18
really change things going forward here?
00:18:20
>> It it changes it big. Well, so so the
00:18:23
Predator drone was the first big drone
00:18:25
program like 1015 years ago. It was this
00:18:27
big honken drone. Um and then if you
00:18:30
remember Obama would take out some of
00:18:32
these al-Qaeda leaders with drones on
00:18:34
their balcony and things like that. Uh I
00:18:37
think we uh President Trump took out
00:18:39
Sulammani with a drone near his car.
00:18:41
That was the beginning. And then Iraq uh
00:18:44
sorry the Russia uh Ukraine war happened
00:18:46
where it's like drone on drone. 70% of
00:18:48
the the casualties are for drone but
00:18:51
because of drones. So um drone on drone
00:18:55
warfare, robot on robot warfare, those
00:18:57
things are the future for sure,
00:18:59
>> right? And that's why companies like
00:19:00
Ander are companies like Anderoll is
00:19:02
because they're making unmanned systems.
00:19:04
A and this has been something you've
00:19:06
specifically been very focused on and
00:19:09
you tweeted today a little bit about a
00:19:11
competition. We'll play a little video
00:19:12
here. And this Lucas lowcost unmanned
00:19:14
combat attack systems.
00:19:17
>> It used to take a lot of time. It
00:19:19
certainly wasn't startup time to get new
00:19:23
product into the channel for our
00:19:25
military to use. Explain what program
00:19:27
you're running here. Feels like the
00:19:28
DARPA self-driving challenge all over
00:19:31
again and what these drones cost. I know
00:19:33
there's a company making them for I
00:19:35
think $35,000. Am I correct?
00:19:37
>> I mean the the small drones like I'm
00:19:39
holding right there and that are way
00:19:41
cheaper than that. The Lucas oneway
00:19:43
attack drone which can go 5 6 700 miles
00:19:47
at the speed of an airplane
00:19:49
>> carry a big warhead. Those are like 50
00:19:52
$80,000
00:19:53
depending on what kind of equipment you
00:19:54
put on you put on it. But the we've have
00:19:57
a drone dominance program and the real
00:19:59
and we basically have to build an
00:20:01
arsenal for for drones. Now are we
00:20:04
likely to have a territorial conflict
00:20:05
like Russia Ukraine with Canada and
00:20:07
Mexico? No. But but we do want to take
00:20:10
out drug drones at the border and we
00:20:13
want but long one-way attack drones are
00:20:16
important for you know any kind of major
00:20:19
conflict like you're seeing in Iran. but
00:20:21
also to protect military bases for
00:20:23
America 250 World Cup uh Olympics in 28
00:20:28
like there's other there's a lot more
00:20:29
uses of drones for surveillance not just
00:20:32
for you know for combat
00:20:34
>> there you're showing drones that are
00:20:36
sort of human operated but how much of
00:20:38
this should basically be AI so that it's
00:20:41
just some computer vision and again back
00:20:43
to what you said before a model
00:20:46
understands the rules and the red lines
00:20:48
but otherwise is be on task and
00:20:50
accomplish your mission. How much of it
00:20:52
is one versus the other?
00:20:53
>> I mean, it I believe that a
00:20:55
sophisticated drone war is going to be
00:20:57
drone swarms controlled by AI to some
00:21:00
degree or another, right? To what degree
00:21:02
the control matters? Like for example,
00:21:05
drones have decoys. They could spit out,
00:21:07
you know, they could dazzle. They could
00:21:09
put out things. So, how do you
00:21:11
discriminate what's a drone and how to
00:21:13
hit it? You know, you could use AI for
00:21:15
that because it's learned, you know, how
00:21:17
to do automatic target recognition, for
00:21:19
example. Um, and then also could it
00:21:22
identify a person and that and and does
00:21:24
that make it safer? So, it's going after
00:21:26
actually someone you want to get and not
00:21:28
someone you don't want to get. Um, so
00:21:31
there's a lot of uses for AI at the
00:21:32
edge, if you will. Um, in the future
00:21:35
here, the the Ukrainians and Russians do
00:21:37
something called like a kill box where
00:21:39
they lose comms because it's jammed for
00:21:41
this drone and then it just starts going
00:21:43
in a box and looking for, you know, a
00:21:46
the person they're trying to get and
00:21:48
they're trying they're starting to use
00:21:49
AI to do that.
00:21:50
>> Wow.
00:21:51
>> And China has this uh ability already
00:21:55
probably times some magnitude. Yeah.
00:21:58
They have drone swarms because they can
00:22:01
they can force the companies that make
00:22:04
them, not just DJI, to interoperate. So
00:22:07
interoperating drones called
00:22:10
heterogeneous autonomy, right? You take
00:22:11
different kinds of drones and how they
00:22:13
communicate with one another and then
00:22:15
make sure they're not going after the
00:22:16
same target is like a pretty complex
00:22:19
thing that they're definitely working
00:22:20
on.
00:22:21
>> And let's talk uh about the fidelity of
00:22:25
these. Obviously, AI is a new
00:22:28
technology. They it can make mistakes.
00:22:30
Anybody who uses it on a day-to-day
00:22:32
basis might experience a hallucination.
00:22:34
How confident are you in the AI Ukraine
00:22:37
and Russia conflict? They obviously are
00:22:40
not going to be as thoughtful maybe as
00:22:41
we are in putting this together. They're
00:22:43
in a hot war right now. But we as the
00:22:46
United States have to be very thoughtful
00:22:47
about this. So, how confident are you
00:22:49
that this isn't going to make a mistake?
00:22:50
I think that's the key to a lot of this
00:22:53
debate. and and when will it be you know
00:22:55
perfect defined as much better and I
00:22:59
guess this dovetales with the
00:23:00
self-driving
00:23:02
you know thoughts it has to be a
00:23:04
magnitude better than a human so when
00:23:05
will this be a magnitude more accurate
00:23:07
than you know when we have uh make a
00:23:10
mistake as a as a military and we kill a
00:23:12
civilian
00:23:12
>> yeah no it's it's a good question and um
00:23:16
I don't know when that moment hits that
00:23:18
FSD moment where it get kind of gets
00:23:20
better certainly not there and you
00:23:23
wouldn't want to take huge risk with
00:23:25
that in like you know there there's a
00:23:27
gradation of when you would use that and
00:23:28
what kind of risk you were trying to
00:23:30
take or not. If you were trying to take
00:23:32
out a drone using AI using a like a
00:23:35
laser or something you'd be pretty like
00:23:37
okay making mistakes because you just
00:23:38
missed the drone you know like whatever
00:23:41
with a with a with the laser laser goes
00:23:42
off it's all over. Um if you were doing
00:23:45
something more sophisticated in a
00:23:46
population area have densely populated
00:23:49
area you'd take less risk. So, we're
00:23:51
developing procedures, tactics for each
00:23:53
scenario. And this is part of the debate
00:23:55
I had with anthropic, which is we need
00:23:58
AI for things like Golden Dome.
00:24:01
>> Chinese hypersonic missile comes up.
00:24:02
You've got 90 seconds before it
00:24:04
separates and all kinds of decoys and
00:24:07
you don't know where the actual payload
00:24:09
is and you want to get it hit it from
00:24:12
space. and a human can't doesn't have
00:24:14
the reaction time, doesn't have the may
00:24:17
not be able to discriminate with their
00:24:18
own eyes what they're going after.
00:24:21
That's a pretty, you know, lowrisk thing
00:24:22
because it's in space and you're just
00:24:24
trying to hit something that's trying to
00:24:25
hit you. So, I think in the next 10
00:24:28
years, you're going to see a lot of
00:24:29
these applications develop AI to one
00:24:30
degree or another so long as we think
00:24:33
it's safe and it's not going to do the,
00:24:34
you know, make mistakes.
00:24:36
>> Before we get on to the anthropic
00:24:38
discussion, and we really appreciate you
00:24:40
coming here and my lord, this has been
00:24:41
so informative. So thank you Neil for
00:24:43
coming here and explaining to the
00:24:44
American public and to us what you're
00:24:46
working on. It really makes us uh I
00:24:48
think speak for everybody really
00:24:50
confident in what you're doing and it's
00:24:51
so great that you've you know left the
00:24:53
private sector to do this.
00:24:55
>> What I would say just very quickly Emil
00:24:57
is I think that not enough people
00:24:58
understand that the American military
00:25:01
has had to fight with one hand tied
00:25:02
behind their back. Just that little
00:25:05
>> insight that you just gave about
00:25:06
Afghanistan
00:25:08
to me seems so scary because the men and
00:25:11
women that sign up for the American
00:25:13
military. They're doing this to fight on
00:25:16
behalf of this country. They deserve a
00:25:18
lot more than being sent there and all
00:25:20
of a sudden being given this rule book
00:25:22
and say, "Do your best." And it's like,
00:25:24
"Oh, wait. You violated 19 rules trying
00:25:26
to protect America. Do your job." That's
00:25:28
insane.
00:25:29
>> Let's just
00:25:30
>> It's really insane in some cases. I I my
00:25:33
my belief is that's what the frustration
00:25:36
for those soldiers who were out there in
00:25:37
those wars had more than anything. There
00:25:40
was the broader frustration what are we
00:25:41
doing here and then the secondary
00:25:43
frustration is while I'm here why can't
00:25:45
I do my job?
00:25:47
>> Yeah. Is there is there much of a debate
00:25:49
internally and sorry Jake before we move
00:25:51
forward on this
00:25:52
>> regarding this idea of full autonomy in
00:25:56
military action? I don't want to speak
00:25:57
ahead to the anthropic point, but it was
00:26:00
something that the media seemed to say
00:26:02
was part of Daario's concern is that
00:26:05
when you press the button and hand over
00:26:06
complete autonomy and there's a kill
00:26:09
action that you're now giving to a robot
00:26:11
or to some autonomous system, do we then
00:26:13
kind of have a moral issue at hand? And
00:26:17
is that something that's kind of debated
00:26:19
or discussed? And is that the right way
00:26:20
to think about the framing of what goes
00:26:21
on?
00:26:22
>> I mean, we're not even close to there
00:26:24
yet, right? where like this the systems
00:26:26
are not we we wouldn't feel that a
00:26:29
system uh that would have sort of like
00:26:33
real risk for a civilian is ready to
00:26:36
launch yet. So we're not even debating
00:26:38
that. We're just trying to get basic
00:26:40
autonomy in drones, basic autonomy in
00:26:42
underwater unmanned vehicles, basic
00:26:45
autonomy that, you know, you've heard of
00:26:47
this collaborative aircraft that fly
00:26:49
along with a the jetcraft so that it has
00:26:51
more firepower, but it's still tethered
00:26:53
to what the jet does. So, we're
00:26:55
>> incredible. Yeah.
00:26:56
>> Yeah. So, we're just at the very
00:26:57
beginning of this stuff,
00:26:59
>> but for Golden Dome's a good example of
00:27:01
like, yeah, who can oppose that? Like,
00:27:03
it's the only way to get out a threat
00:27:04
like that. Um, so who could oppose if
00:27:08
you have a military base and you have a
00:27:10
bunch of soldiers sleeping that you have
00:27:11
a laser that can take down drones
00:27:13
autonomously on that? So there's it's
00:27:16
it's pretty scenario by scenario, but I
00:27:18
don't we're not having a lot of debate
00:27:19
because the Skynet thing is so not a
00:27:22
realistic thing at this moment, right?
00:27:25
Except if one thing I did tell the
00:27:27
anthropic guys, I was like, you know, or
00:27:29
I'd tell any company, your models are
00:27:32
getting stolen by the Chinese. They're
00:27:34
going to unguard rail them and use them
00:27:36
against us. And then you want our models
00:27:38
to be less capable against your models.
00:27:41
It's sort of
00:27:41
>> they're not going to be thoughtful. In
00:27:42
other words, they're going to go for it.
00:27:44
And you know, if we just benchmark this
00:27:46
against where we were at, you know, but
00:27:49
101 15 years ago, there was the Wikileak
00:27:53
of collateral murder, I think they
00:27:55
called it, where we tragically had an
00:27:58
Apache take out some journalists. And
00:28:00
this technology even applied today
00:28:03
probably would have avoided that in my
00:28:04
mind. Yeah. Like we have enough that
00:28:07
when you're targeting not drones but you
00:28:10
know people on the ground with an Apache
00:28:12
this would have probably avoided that.
00:28:14
>> Yeah. Or or you know the Kuwaiti
00:28:17
aircraft hitting you know an American
00:28:19
aircraft making a mistake because it
00:28:21
doesn't have the identification. I mean
00:28:24
it's the same self-driving argument to a
00:28:26
degree. like self-driving could save
00:28:28
lives even though it's scary to look at
00:28:30
a car without a human behind the wheel
00:28:32
but there's tons of scenarios where it's
00:28:34
a way better safer option more precise
00:28:38
um than the alternative
00:28:39
>> all right before we move on to the Dario
00:28:41
thing and anthropic and that brew haha
00:28:43
there was one piece that we haven't
00:28:44
addressed with this interaction Freeberg
00:28:46
Chimath which is uh the Israeli
00:28:49
government and their desire to take out
00:28:54
this regime And us according to Tucker
00:28:58
Carlson and a large contingent of the
00:29:01
MAGA base they feel that we are captured
00:29:04
by this group. Does Israel have too much
00:29:07
influence over the United States with
00:29:09
regard to these actions in the Middle
00:29:11
East? This is you know a big debate
00:29:13
within the party within the Republican
00:29:15
party within the MAGA constituent. Hey
00:29:17
we number one we don't want these wars.
00:29:19
Number two is Israel driving this thing
00:29:21
to the point of Rubio's quotes that hey,
00:29:23
we're doing this because Israel is going
00:29:25
anyway. I think we should address it
00:29:27
here. Not that I have a personal stake
00:29:28
in this. I'll give my personal opinion
00:29:29
at the end.
00:29:30
>> I don't think the president is captured
00:29:32
by Israel in the least. I think he
00:29:34
decides what is in the best interest of
00:29:36
the United States. And if
00:29:40
Israel can be a part of that, then
00:29:43
they're a part of it. And look, let's be
00:29:45
clear, they're incredibly capable. And
00:29:48
so in something like this, to be able to
00:29:51
incorporate the intelligence of MSAD,
00:29:54
what you're seeing today in this
00:29:55
operation Epic Fury, we're four days in,
00:29:59
Iran has been 90% depleted of all of
00:30:01
their munitions. It looks like they're
00:30:02
just firing no more missiles out from
00:30:05
Iran to anywhere else. There's fleets of
00:30:08
drones and planes just waiting.
00:30:09
Everybody knew where the Iranians were.
00:30:12
It's great that when we make a decision
00:30:14
on something that we need to do, we can
00:30:16
rely on our allies. I think the opposite
00:30:18
question should also be asked like what
00:30:20
was the UK doing? Why is Spain
00:30:21
pontificating? Why was Europe taking the
00:30:23
weekend off before they could even issue
00:30:25
a statement? Why don't you ask that
00:30:27
question?
00:30:27
>> Yeah. No, it's equally valid question.
00:30:30
You know, uh Freeberg, do you want to
00:30:32
get in on this or no?
00:30:34
>> No.
00:30:34
>> I'm a Jew. No one's going to care what I
00:30:35
have to say. They're either going to
00:30:37
they're either going to be like totally
00:30:39
like or they're going to say this guy's
00:30:40
a Jew. We shouldn't listen to him. So,
00:30:42
like let's move on. Go ahead.
00:30:43
>> Yeah. Yeah. Yeah.
00:30:44
>> Emil, any thoughts on on this? I do want
00:30:47
to know from Emil though like you know
00:30:49
is this Iron Dome working this laser in
00:30:52
Israel system is this operational and if
00:30:54
so is there any success metrics you can
00:30:56
share around it
00:30:57
>> I mean I think that the golden sorry the
00:31:00
golden iron beam was the first
00:31:01
generation of of the Israeli air air
00:31:04
defense thing and then they're build
00:31:06
building Iron Beam and and I think it's
00:31:08
still earlyish but yeah it works they're
00:31:11
they're a technologically sophisticated
00:31:14
country that's very small that has like
00:31:17
a reason to invest in these things and
00:31:19
they have a lot of smart people to do
00:31:21
them. So I think I think it's good. But
00:31:23
>> does it primarily work on rockets? And I
00:31:25
guess I just want to understand the
00:31:26
logical evolution of this because in the
00:31:28
80s and 90s there was a lot of
00:31:30
conversation about space-based lasers
00:31:32
that could shoot ICBMs out of the sky to
00:31:34
avoid, you know, global nuclear war and
00:31:38
we could always take out every nuclear
00:31:39
warhead delivered on an ICBM. Is that
00:31:42
technology feasible? Is there a place in
00:31:44
the near future where we could see
00:31:46
basically maximal global deterrence
00:31:48
using these systems, either groundbased
00:31:50
or space-based to take out hypersonic
00:31:52
missiles?
00:31:53
>> I think I think the the harder but more
00:31:57
valuable problem to solve would be the
00:31:58
space-based way of doing it because then
00:32:01
um you could get at any kind of almost
00:32:04
any kind of threat that hits space, but
00:32:06
you still need a ground layer because
00:32:08
there's cruise missiles that could come
00:32:10
at you, there's drones and so on. to so
00:32:12
we call them multi-layers like how do
00:32:14
you how do you get every kind of weapon
00:32:16
at every layer
00:32:18
but you know directed energy lasers as
00:32:22
they get more powerful you could take on
00:32:24
a bigger weapon farther away right so
00:32:26
those uh so those technologies as that
00:32:30
as they improve it gets more and more
00:32:32
capable and I think all these defense
00:32:34
systems uh are going to get more and
00:32:36
more capable to get more and more of a
00:32:38
variety of weapons at farther farther
00:32:40
standoff which is what want you want it
00:32:42
to you don't want to shoot it when it's
00:32:43
right over Tel Aviv. You want to shoot
00:32:45
it, you know, when it's still over their
00:32:47
their land ideally.
00:32:48
>> Are the laser interceptors in the field
00:32:51
today. There's reports that they are.
00:32:53
>> I I think there's some I think they've
00:32:55
demonstrated some of them.
00:32:56
>> Got it. And and is this our technology
00:32:58
or Israel's technology? Because
00:32:59
President Trump said, "Hey, that's
00:33:02
actually our technology." Is there any
00:33:04
insight there?
00:33:04
>> We we have collaborations with Israel on
00:33:07
some of this stuff. They have their own.
00:33:09
We have our own. Um, so it's not this is
00:33:12
uh but they're good at tech, we're good
00:33:16
at tech. There's certain there's certain
00:33:17
ways you get part of our system and part
00:33:19
of their system because it's like a he
00:33:21
it's a quickly evolving part of of
00:33:24
science right now. How do you cohhere
00:33:26
beams of light to like get distance? How
00:33:29
to use high-owered microwave to like
00:33:31
just drop drones in their tracks? Um
00:33:34
there's lots of different ways to get at
00:33:35
some of these things. Um and and yeah, a
00:33:38
lot of it's ours. uh and a lot of and
00:33:40
some of it's theirs.
00:33:41
>> Yeah. And uh to to the earlier question,
00:33:44
you know, I I am pro-regime change if it
00:33:47
can be done thoughtfully and obviously
00:33:50
isolating a dictator, that's the best
00:33:52
thing you can do. We've done that
00:33:53
successfully with, you know, Putin, Kim
00:33:56
Jong-un, etc. Keep diplomacy up. But if
00:33:59
there is a moment in time where you
00:34:01
could free the people of Iran after 50
00:34:04
years of being subjugated by these
00:34:07
lunatics and dictators, I'm all for it.
00:34:09
And I actually trust President Trump to
00:34:11
make that decision. I know this may
00:34:13
sound crazy. People think like I'm a
00:34:15
libtart or something because of the way
00:34:16
my besties frame me on this program,
00:34:18
which is completely inaccurate. I'm an
00:34:20
independent.
00:34:20
>> You are.
00:34:21
>> I actually I'm completely independent
00:34:24
and I am just based on my voting and I'm
00:34:26
not on either one of these sides. I am
00:34:28
pro President Trump and I trust his
00:34:30
judgment. I think he has more
00:34:31
information than us. I think you have
00:34:32
more information. I actually trust you
00:34:34
guys to do it thoughtfully and there
00:34:35
obviously was a window here. Israel can
00:34:37
have their own, you know, motivation.
00:34:39
That could be the China motivation, but
00:34:40
there's also spreading democracy, which
00:34:42
might be the least of people's concerns
00:34:45
here, but that's on the top of my list.
00:34:47
I would like to see the people of Iran
00:34:49
free. Just to build on your point,
00:34:50
Jason, the thing that Emil said before,
00:34:52
which I think is important as well, is
00:34:54
we have an enormous amount of learnings
00:34:56
about what happened in Iraq. Yeah. We
00:34:58
also have a ton of learnings between the
00:35:00
Iran Iraq war and a ton of learnings in
00:35:02
53 when US and the British deposed Mosad
00:35:06
in the or at least foremented that and
00:35:08
put the sha and then the sha was booted
00:35:10
up.
00:35:11
>> Yeah.
00:35:12
>> If you take those three chapters in
00:35:13
Iranian history or that regional
00:35:15
history, there's a ton to learn. And to
00:35:19
your point, there is a way to affect
00:35:21
what we need to do without creating some
00:35:25
20-year forever war. There was an
00:35:27
incredible tweet. I don't know if you
00:35:28
guys saw this. Somebody said, "So every
00:35:31
war doesn't have to be
00:35:34
>> three decades and trillions of dollars
00:35:35
to your friends in Virginia, Maryland,
00:35:37
and DC. Did you guys see that tweet?"
00:35:39
It's true. These things can be one and
00:35:42
done in and out.
00:35:44
And if President Trump succeeds here, I
00:35:46
just want to also give him some flowers
00:35:48
here. The people of Venezuela and the
00:35:51
people of Iran being free represent
00:35:53
about 5% of the people in the world
00:35:54
living under an autocracy, under a
00:35:57
dictator. If those both flip back to
00:36:00
democracies, he'll have done more for
00:36:02
the spread of democracy than any
00:36:03
president for many decades, perhaps in
00:36:06
our lifetime. This would be incredibly
00:36:08
noble. Incredibly noble, incredibly
00:36:10
just. And
00:36:11
>> would you in the human rights set
00:36:14
>> want him to get the Nobel then?
00:36:16
>> Absolutely. Give him all the Nobels.
00:36:18
Like literally if you can free people,
00:36:21
>> all of them. Give him every prize. Give
00:36:23
him an Oscar.
00:36:24
>> Physics, chemistry,
00:36:27
he can have everything.
00:36:29
>> Physics, philosophy.
00:36:30
>> Jay Cal's an independent. When's the
00:36:32
last time you voted for a Republican
00:36:33
presidential candidate? Just curious.
00:36:36
>> Um,
00:36:37
yeah.
00:36:38
>> Say it. Say it.
00:36:39
>> No, no, no, no. Um,
00:36:41
>> Montdale.
00:36:42
>> No, no, I didn't. I would have voted
00:36:44
for, if I was of age, I would have voted
00:36:46
for um I wouldn't have voted for the
00:36:47
Bushes. I voted for the moderates. Um,
00:36:52
uh, uh, obviously Clinton and Obama.
00:36:54
>> Oh, we're playing the would have, should
00:36:55
have game.
00:36:56
>> I would have voted for Reagan in
00:36:58
>> I would have bought Nidia at $4.
00:37:00
>> Well, no. And I I didn't vote for Kamla,
00:37:02
so I'll leave it at that. But I voted
00:37:04
probably
00:37:05
at that. Why don't you say that you
00:37:06
voted for President Trump?
00:37:07
>> Just say you voted for President Trump.
00:37:09
I don't want to complicate things,
00:37:10
>> but you did. So, just say it.
00:37:12
>> I didn't vote for Kla. I'll leave it at
00:37:13
that. All right.
00:37:14
>> It's so weird that you that you'll say
00:37:16
you're a moderate, but you won't say
00:37:17
that you voted for President Trump.
00:37:18
>> I am supporting President Trump in about
00:37:20
60 70% of what he does. Uh, let's leave
00:37:22
it at that. Three, two. All right. Let's
00:37:24
talk economic
00:37:26
impact of oil and insurance. Oil has
00:37:28
rose to $84 a barrel Wednesday
00:37:30
straightfor
00:37:33
basically a standstill at this point.
00:37:35
Here's the clip. You can see the traffic
00:37:37
slowing down and then hey, some of the
00:37:39
dots are even going away. That could be
00:37:41
uh ships were taking out.
00:37:45
Unless the straight opens, 3.3 million
00:37:47
barrels of daily production would be
00:37:49
lost early next week. And then there's
00:37:53
insurance companies. They've all
00:37:54
cancelled
00:37:56
the war risk coverage of vessels in the
00:37:59
Gulf effective March 5th. Super tanker
00:38:01
traffic dropped 94% within the first 48
00:38:04
hours. Trump said the US would provide
00:38:06
political risk insurance for all
00:38:08
maritime trade through the Gulf,
00:38:10
especially energy. Freeberg, your
00:38:12
thoughts on the economic second order
00:38:14
effects that we're starting to
00:38:16
experience here and over the next four
00:38:18
weeks could be um you know intense and
00:38:21
acute.
00:38:21
>> The modern insurance market emerged
00:38:24
specifically to solve the risks of
00:38:26
maritime trade. So in the 17th century,
00:38:29
Lloyds of London, which was a coffee
00:38:30
shop in London, where all the maritime
00:38:33
traders would get together and they talk
00:38:34
about, hey, what's the safest route so
00:38:36
pirates don't get our ship and so you
00:38:38
don't run into weather. That's where
00:38:39
they would kind of have these
00:38:40
conversations and eventually they
00:38:41
started underwriting the risks of the
00:38:43
shipping uh routes and giving each other
00:38:46
guarantees. They said, "Hey, if you make
00:38:48
this route, great. You pay me a certain
00:38:50
amount and if you don't make the route,
00:38:52
I'll pay you the loft value." And that's
00:38:54
how Lloyds of London, which is the kind
00:38:56
of world's biggest reinsurance market,
00:38:58
started today. Lloyds of London has 78
00:39:00
what are called syndicate members. And
00:39:02
these are kind of these pools of
00:39:03
reinsurance that underwrite big crazy
00:39:05
risks like maritime insurance for folks
00:39:08
that are moving oil tankers through the
00:39:10
straight of Hermuz, which the IRGC just
00:39:13
announced they're shutting down. When
00:39:15
the IRGC announced that they were
00:39:17
shutting down the straight of Hermuz,
00:39:19
there's a significant risk of all the
00:39:20
mines going in the straight and the
00:39:23
ships getting attacked and blown up. So
00:39:24
loss of value. The insurance premium
00:39:27
spiked initially from a quarter% so
00:39:29
0.25% of the value of the ship to 1.25%.
00:39:34
So it went up by like 5x and so folks
00:39:36
had to pay a lot more of the value of
00:39:37
their ship in order to continue the
00:39:39
routes and get guarantees that they'll
00:39:41
make it through. And then all of the
00:39:43
markets started to shut down. So once
00:39:45
the conflict got heavier, everyone said,
00:39:47
"Let's shut this thing down." And that's
00:39:48
a obviously a massive risk to energy
00:39:50
prices globally, which drives inflation
00:39:52
and puts US economic security at risk.
00:39:55
And so this is a brilliant move. I would
00:39:57
say the US government stepped in with
00:39:59
the US International Development Finance
00:40:01
Corporation, which was actually funny
00:40:03
enough started a couple of years ago
00:40:05
like in 2019 or something like that as a
00:40:08
kind of output of one of the agencies
00:40:10
that provided credit from USAD. much
00:40:13
talked about USA ID. And so they're
00:40:15
they're leveraging the credit capacity
00:40:18
of this old US ID agency to go out and
00:40:21
say to all the shipping companies, hey,
00:40:23
we'll give you insurance on your routes.
00:40:26
And the reason they need it is the
00:40:27
shipping companies are levered. They
00:40:30
take on debt to buy the ships. And the
00:40:32
debtors require that they have insurance
00:40:35
or else they're not allowed to take the
00:40:37
routes because the debtors are
00:40:38
ultimately going to be out the money.
00:40:39
And so the shipping companies themselves
00:40:41
need to have insurance. And so this
00:40:44
provides a market that has now gone
00:40:46
away. Very smart. And ultimately a lot
00:40:48
of people are saying this could actually
00:40:50
reshore or onshore maritime insurance
00:40:53
back to the United States and create an
00:40:55
entirely new insurance industry here in
00:40:58
the US that has historically been served
00:41:00
almost exclusively by European
00:41:02
syndicates and European partners. And it
00:41:04
actually creates a big economic
00:41:06
opportunity as this war dies down for
00:41:08
American insurance companies and
00:41:10
American brokers to basically be the
00:41:12
underwriters and the guarantors of this
00:41:14
sort of insurance and create a new
00:41:15
industry. So that's super super
00:41:17
interesting kind of side story on what's
00:41:18
going on here. All right, some breaking
00:41:20
news here folks via Bloomberg. The
00:41:22
Pentagon has formally notified Anthropic
00:41:25
that it's been deemed a supply chain
00:41:27
risk. This has never happened to an
00:41:29
American company. It has happened to
00:41:31
Russian companies. uh and Chinese
00:41:34
companies Huawei and for background the
00:41:37
department of war cancelled Anthropic
00:41:40
$200 million contract on Friday and said
00:41:43
they would do this. The dispute came
00:41:44
down to two clauses according to sources
00:41:46
and we have one of the principles here.
00:41:49
So we will hear directly from him in a
00:41:51
moment. Anthropic had two concerns.
00:41:54
Number one, fully autonomous weapons aka
00:41:56
murderbots. As we previously discussed,
00:41:59
Daario didn't feel that their technology
00:42:01
was reliable yet and wanted some
00:42:04
assurances. The second thing Anthropic
00:42:06
said was uh they were concerned about
00:42:09
mass surveillance of Americans because
00:42:11
they believe this technology is uniquely
00:42:14
powerful and uh it's can do things
00:42:19
beyond what a series of webcams or a
00:42:21
network of 7-Eleven cameras can do.
00:42:25
Pentagon said they wanted all lawful
00:42:28
use. Dario, you're welcome to come on
00:42:30
the program next week or any time to
00:42:32
give your side of the story, but this
00:42:33
week we have Emil. Emil, your thoughts
00:42:36
and explain to us what happened here and
00:42:38
how this broke down.
00:42:40
>> It's worth a little history, short
00:42:41
history. So,
00:42:44
if you remember the Biden executive
00:42:46
order on AI, which was this crazy
00:42:49
executive order that limited the amount
00:42:51
of compute any model company could do
00:42:54
and was essentially grandfathered in a
00:42:57
few a small number of AI companies that
00:42:59
they were going to designate the winners
00:43:01
and everyone else was out so they could
00:43:03
have more control on what they did. Um,
00:43:06
Anthropic was one of those winners. Um,
00:43:08
and then they were smart. act as a good
00:43:11
sales strategy to sell in to the most
00:43:14
sensitive parts of the US government
00:43:16
like all of our combatant commands sent
00:43:19
central command that's doing the Iran
00:43:21
fight now the Indoaccom command which is
00:43:24
sort of responsible for China several of
00:43:26
the intelligence agencies and they did
00:43:28
forward deployed engineers palunteer
00:43:30
style so they're very got very sticky um
00:43:33
to the workflows and all that and so I
00:43:36
came in and I got the AI portfolio for
00:43:39
department in August and I said, I just
00:43:42
want to see the contracts. You know, the
00:43:44
old lawyer in me and I looked at
00:43:46
contracts. I was like, holy cow, they
00:43:48
say you can't use them for you can't use
00:43:50
them to plan a a kinetic strike. You
00:43:53
can't use their AI model to move a
00:43:55
satellite. You can't There was a a 20
00:43:58
page.
00:43:58
>> You can't do a war game scenario with
00:44:00
it.
00:44:00
>> You could do a scenario, but you can't
00:44:02
like let's suppose you're writing a plan
00:44:04
saying like if this happens, here's what
00:44:06
we would do. and it might involve a
00:44:08
kinetic strike which causes harm to a
00:44:10
human. So like, well, what do you think
00:44:12
these folks do? You know, we use the
00:44:14
Department of War. This is what we do.
00:44:16
And so I said, okay, well, I've got to
00:44:18
number one, have direct relationships
00:44:20
with these companies, not just through
00:44:21
Palunteer, because I want to use it more
00:44:23
broadly. And then number two, I need to
00:44:26
have the terms of service be rational
00:44:28
relative to our mission set. So we
00:44:30
started these negotiations and and took
00:44:33
three months and I had to sort of give
00:44:37
them scenarios about like this Chinese
00:44:39
hypersonic missile example. They're
00:44:40
like, "Okay, we'll give you an exception
00:44:42
for that." Well, how about this drone
00:44:44
swarm? We'll give you an exception for
00:44:45
that. And I was like, "The exceptions
00:44:47
doesn't work. I I can't predict for the
00:44:50
next 20 years what all the things we
00:44:51
might do use AI for." Um, and so so all
00:44:55
lawful use seems like a good thing. If
00:44:58
Congress wants to act, great. We have
00:44:59
our own internal policies like we'll
00:45:02
follow them. We're not knuckle draggers
00:45:04
here. We want we don't want to hurt
00:45:06
people unnecessarily. So, you know, it's
00:45:08
our province to decide how we fight and
00:45:10
win wars um so long as they're lawful.
00:45:14
And I think at some point it turned into
00:45:17
a PR game for them because they were not
00:45:21
going to win this intellectual battle of
00:45:25
well, we're going to stop you. we're
00:45:26
going to use our judgment because we
00:45:28
think Congress is behind and impose it
00:45:30
on the US military. And it became this
00:45:33
like let's find the issues that are most
00:45:35
inflammatory, robot weapons and mass
00:45:39
surveillance. I mean, like we're the
00:45:40
Department of War. We're not the FBI.
00:45:41
We're not Homeland Security. We're not
00:45:43
>> You're not allowed to legally spy on
00:45:45
Americans.
00:45:46
>> Yeah. You're not you're not. So it's so
00:45:48
so you're like and then what it came
00:45:51
down to on that issue just as an
00:45:52
anecdote is they didn't want us to bulk
00:45:56
collect public information on people
00:45:59
using their AI system and they wrote it
00:46:02
in a way that I was like so you're
00:46:03
telling me before we got to bulk collect
00:46:05
if someone types in you know Shimat's at
00:46:08
LinkedIn it's I'm using public available
00:46:12
information that I would be violating
00:46:14
your terms of service like yeah well
00:46:15
okay let's rewrite it. was months of
00:46:17
this like stuff. Um, which which was
00:46:21
sort of interminable and then the
00:46:23
trigger point was after the Maduro raid,
00:46:27
one of their execs called Palanteer who
00:46:29
we buy ourselves through and asked them
00:46:32
uh was our software used in that raid
00:46:35
which is by the way classified
00:46:36
information anyway. So we're trying to
00:46:38
get classified information and implying
00:46:41
that if there was used in that raid that
00:46:43
that might violate their terms of
00:46:44
service. So they wanted to enforce, this
00:46:47
is very important here. Yeah.
00:46:48
>> They wanted to enforce their terms of
00:46:50
service. They went behind your back to
00:46:52
try to collect information to then
00:46:56
>> maybe pull your license for their
00:46:58
technology.
00:46:59
>> But you know, and it wasn't by behind my
00:47:01
back. I don't want to accuse them of
00:47:02
that. Palunteer is the prime contractor
00:47:04
this sub. Um, but it raised enough alarm
00:47:07
with Palenteer who's got a trusted
00:47:09
relationship with the department to tell
00:47:10
me and I'm like, "Holy, what if this
00:47:14
software went down, some guard rail
00:47:16
kicked up, some refusal happened for the
00:47:19
next fight like this one and we left our
00:47:22
people at risk and I had so I went to
00:47:24
Secretary Hath. said this would happen.
00:47:26
And that was like a wo moment for the
00:47:28
whole leadership at the Pentagon that
00:47:30
we're potentially so dependent on a
00:47:32
software provider without another
00:47:34
alternative that has the right or
00:47:36
ability to do to not only shut it off,
00:47:38
maybe it's a rogue developer who could
00:47:40
poison the model to make it not do what
00:47:42
you want uh at the time or sort of trick
00:47:44
you because you have to trick it. I mean
00:47:46
all these things that we know we were
00:47:48
about models or hallucinate purposefully
00:47:50
or do or not follow instructions like
00:47:53
some insider threat stuff. So then that
00:47:56
culminated in the Tuesday kind of
00:47:58
dramatic meeting with Hexath and
00:48:00
Secretary Hexath and me and and Daario
00:48:03
um with the Friday deadline that that
00:48:06
got blown and I never thought they
00:48:07
really wanted to make it. M is is the
00:48:10
model entirely hosted by Anthropic or
00:48:13
just explain to us technically does this
00:48:15
sit in a cloud that Palunteer runs for
00:48:18
you guys? Um is there really technically
00:48:20
a way that employees at Anthropic could
00:48:23
kind of interfere intervene in the use
00:48:25
of the model?
00:48:26
>> Yeah. So they put their model in AWS
00:48:30
GovCloud
00:48:32
>> GovCloud. Yeah.
00:48:32
>> And then Palunteer serves it from there
00:48:35
and they refresh it. They held the
00:48:38
control plane for the model. So, so
00:48:41
yeah,
00:48:41
>> they can change the model weights if
00:48:43
they want. They can do whatever they
00:48:44
want.
00:48:45
>> Yeah.
00:48:45
>> The insight into this thing is
00:48:46
unbelievable. Not just governments, but
00:48:50
now if you're running a company, the
00:48:51
reality is that what anthropics showed,
00:48:55
which by the way is their right at some
00:48:57
level, is that they are going to have a
00:49:01
political perspective and a set of terms
00:49:04
that reflect their philosophy and that
00:49:07
that philosophy can change on a dime.
00:49:09
But what the government did was also
00:49:11
completely reasonable, which is we can't
00:49:13
rely on you if you're going to be
00:49:16
completely unreliable and
00:49:20
disallow things that are reasonable.
00:49:23
I'll give you a different example to
00:49:24
make the point.
00:49:26
There's a state that wants to run some
00:49:28
healthcare program, but they're a
00:49:30
prolife state.
00:49:33
You can't conduct abortions in that
00:49:34
state. Does that mean that the anthropic
00:49:37
engineers can decide, you know what,
00:49:39
we're pro-choice, so we're going to
00:49:41
change the access model and the
00:49:42
capability of that model inside of that
00:49:44
state. Is that allowed? Should that be
00:49:47
allowed? At one level, you'd say, "This
00:49:50
is a private company. They're allowed to
00:49:51
choose." But what that really means is
00:49:53
for the government, for all the states,
00:49:55
for any city, for every company, you
00:49:57
cannot choose to only use one of these
00:50:00
things because it is just a matter of
00:50:02
time until some person inside of one of
00:50:05
these companies goes on some lunatic
00:50:07
moral tirade and then jeopardizes your
00:50:10
business against something that is
00:50:13
nothing about law, but is everything
00:50:14
about subjectivity. That is the huge
00:50:17
thing that this thing tore open this
00:50:19
weekend. So if you're not figuring out
00:50:21
how to be multimodel and agnostic across
00:50:23
these models, you're taking on enormous
00:50:25
business risk after Friday because you
00:50:27
can't tolerate that these folks will do
00:50:29
that. It's too critical of a technology.
00:50:32
By the way, this is deplatforming all
00:50:34
over again. Remember what happened when
00:50:36
you didn't like what was said? Now all
00:50:38
of a sudden you were deplatformed. This
00:50:40
is that times a thousand because this is
00:50:42
not about posting on social media. This
00:50:45
is about using fundamental technology to
00:50:47
either advantage or disadvantage your
00:50:48
business. Emil.
00:50:50
>> Yeah. I mean, I think I described it the
00:50:52
other way the other day as these c the
00:50:56
leaders of these companies say they're
00:50:58
going to cause 50% white collar
00:51:00
unemployment. This is as powerful as a
00:51:02
nuclear bomb. You it's like 50,000
00:51:05
geniuses in a data center. So you could
00:51:07
have a small country coers the world
00:51:09
into its whatever. So you're like, "Holy
00:51:10
cow." All right. So this is a general
00:51:12
substrate of intelligence of technology
00:51:15
that's applicable to a lot of things.
00:51:17
Very generalized. It's not like workday
00:51:20
like HR software we could just use a
00:51:22
competitor. This is going to be part of
00:51:24
our everyday life in so many different
00:51:25
ways and the controlling the like what
00:51:29
whether it has a moral conscience. I
00:51:30
mean anthropic has its own constitution.
00:51:33
It has its own soul. It's not the US
00:51:36
constitution. So you're subject to that
00:51:39
plus whatever whims and how that
00:51:40
changes. And that's a scary thought for
00:51:43
for Americans generally. Um and I think
00:51:46
that did come through a little bit
00:51:47
today. And in the coming years it's
00:51:49
going to be a bigger and bigger deal.
00:51:51
>> So take us through OpenAI software,
00:51:54
Gemini software and Gro software. Have
00:51:58
they push back on any use or are they
00:52:01
like Dell or Apple? They sell you a
00:52:03
computer and you have the computer and
00:52:05
you can use it as you will. Have any of
00:52:08
those given you any push back? So
00:52:10
Grock's all in for all awful use cases
00:52:12
across all classified and unclassified
00:52:14
networks as you'd expect because and you
00:52:16
know Elon's truth seeking. We want truth
00:52:19
and Department of War. We don't want
00:52:20
ideology
00:52:21
>> because ideology will mess with
00:52:23
operational decisions like you you don't
00:52:25
want anyone to anything to be fake or
00:52:27
tilted.
00:52:28
>> We're we're surgeing Google and
00:52:30
>> we have them we have Google for all
00:52:32
lawful use cases on on classified
00:52:34
networks and we're trying to move them
00:52:36
to classified networks. are just they
00:52:38
have to build out infrastructure because
00:52:39
the stuff's complicated.
00:52:41
>> So they're in compliance in terms of
00:52:43
what you're looking for as a partner.
00:52:44
And then
00:52:45
>> I guess the last one is OpenAI and Sam
00:52:47
seems to be
00:52:48
>> just characteristically playing both
00:52:50
sides a bit trying
00:52:53
to his credit
00:52:56
um I called him and said I need a
00:52:58
solution if this thing goes sideways. I
00:53:01
need multiple solutions. I'd like you to
00:53:03
be one of them. And he's like okay well
00:53:05
what can I do for the country? is like,
00:53:07
I need to get you up running as soon as
00:53:09
I can. And he was he was trying to
00:53:12
protect anthropic to his credit. He was
00:53:14
like, don't do don't call him a supply
00:53:16
chain risk. That's bad for the industry.
00:53:18
Let me maybe I can negotiate terms that
00:53:20
they'll find acceptable. Uh but he's in
00:53:23
the middle because they're they compete
00:53:24
for the same researchers. So, a lot of
00:53:27
this comes down to this thousand
00:53:29
researchers like baseball players that
00:53:32
get traded between these companies.
00:53:33
>> Moneyball. Yeah. These are the best of
00:53:35
the best. It's a very moneyballish sort
00:53:36
of thing and there's not that many of
00:53:38
them and you lose
00:53:40
>> 20% of them and all a sudden, you know,
00:53:42
they launched Claude Code before you
00:53:44
launched Codeex or something like that
00:53:46
and then the numbers changed pretty
00:53:47
dramatically. So, he was being a real
00:53:49
patriot um to his credit and trying to
00:53:52
him, you know, help anthropic while they
00:53:54
were trashing him and recruiting from
00:53:56
his company. And I I am not biased. I
00:53:58
just I want all of them. I want to give
00:54:00
them all the same exact terms uh because
00:54:03
I need redundancy. I want to see if they
00:54:06
diverge or not or did they if they
00:54:08
converge maybe I only need two over time
00:54:10
but we don't know it's too early
00:54:11
>> but why why keep them in the mix? So if
00:54:14
there's clearly like a difference of
00:54:16
operations and philosophy and how they
00:54:19
want to run their business and there's
00:54:21
other models is is their model
00:54:23
particularly good at particular
00:54:25
applications that make it important to
00:54:28
keep it in the mix given that there are
00:54:30
three or four other kind of alternatives
00:54:31
here.
00:54:32
>> Anthropic you mean?
00:54:33
>> Yeah. Oh well because number the number
00:54:35
one reason we were having this
00:54:37
conversation at all was because they
00:54:38
were deeply embedded. So now I have to
00:54:40
unentangle them and the other companies
00:54:44
have not gone as heavy enterprise
00:54:46
enterprise sales forward deployed
00:54:48
engineers government business. So
00:54:51
they're have to catch up not on
00:54:52
necessarily the capability of the model
00:54:54
but just how do you serve the government
00:54:56
>> the Bible is just way ahead on that
00:54:58
>> right
00:54:59
>> but the models themselves you don't
00:55:00
think are uniquely advantaged or do you
00:55:01
have a view on that at this point? I
00:55:03
don't have a view on that. I I don't
00:55:04
think they're, you know, I mean,
00:55:06
certainly cloud code was was innovative
00:55:09
and ahead. That's true. Um, but I do I
00:55:12
believe in 12 months Codeex is not going
00:55:15
to be close. I think it will be.
00:55:17
>> I think you're right. There's an
00:55:18
asmmptoing that's happening. If you just
00:55:20
look at the like the confidence interval
00:55:22
on how overperforming or
00:55:24
underperformance some of the leading
00:55:25
models are, the error bars are
00:55:27
shrinking, right? the confidence
00:55:28
intervals like these things are all kind
00:55:30
of becoming the same eventually they're
00:55:33
all getting access to enough power
00:55:35
enough compute they're generating
00:55:38
similar results it turns out which I
00:55:39
think you would expect so even more
00:55:42
important that you have a complexion of
00:55:44
models the other thing Emil I don't know
00:55:46
if you saw this but they posted about
00:55:49
the revenue ramp of anthropic
00:55:52
and well I have a small software company
00:55:55
called 8090 and I asked the team. Let's
00:55:57
go look at our opex. I posted it because
00:56:00
I was so shocked at these numbers. Our
00:56:01
costs have more than tripled since
00:56:04
November of 25. Between the inference
00:56:07
cost that we pay AWS, which is
00:56:10
ginormous, between our cost with cursor,
00:56:13
between anthropic, we are just spending
00:56:15
millions.
00:56:17
>> So now more per unit and more more in
00:56:19
aggregate
00:56:20
>> both. But the problem is that my costs
00:56:23
are going up 3x every three months. My
00:56:25
revenues are not
00:56:28
>> token use is very addicting.
00:56:30
>> Yeah. And by the way, because everybody
00:56:32
has gotten infatuated with what we call
00:56:33
these Ralph Wiggum loops, like just like
00:56:35
send the thing off and like it'll just
00:56:37
go figure something out. A, it never
00:56:39
figures anything out, and B, you just
00:56:41
get this ginormous bill from Cursor. So,
00:56:43
one of the things we had to do was just,
00:56:44
we had to say, guys, you got to
00:56:46
deprecate cursor because you're just
00:56:47
wrapping cloud code and charging us way
00:56:49
too much for these tokens. But I don't
00:56:51
know if you're seeing any of this thing
00:56:53
where like the tool usage, it's so great
00:56:55
to use these tools. Let's be honest,
00:56:56
it's super fun. It's like you feel like
00:56:59
a genius,
00:57:00
>> but then the ROI of these tools are
00:57:02
really important. I'm not sure that
00:57:03
that's as much of an issue for you or
00:57:05
not in in
00:57:08
>> it will be it. It will be
00:57:11
>> for sure.
00:57:12
>> As people find more and more use cases,
00:57:14
the use cases get more sophisticated. So
00:57:16
the next marginal thing you have it do
00:57:19
is likely to be harder and therefore be
00:57:21
more consumptive. Right.
00:57:23
>> Right. Right.
00:57:24
>> Let me just ask Emil, the important
00:57:26
question that I think triggered a lot of
00:57:27
the news this week is why then designate
00:57:30
them a supply chain risk? Why not just
00:57:33
abandon them, move on, use the other
00:57:35
vendors? Like why take this kind of
00:57:37
punitive action?
00:57:38
>> Yeah. So I I don't view it as punitive
00:57:40
and I'll tell you why. It's if their
00:57:44
model has this policy bias, let's call
00:57:46
it, based on their constitution, their
00:57:49
culture, their people and so on. I don't
00:57:52
want Loheed Martin using their model to
00:57:55
design weapons for me. I don't want the
00:57:58
people who are designing the things that
00:58:00
go into the the componentry to come to
00:58:02
me because if that po if you believe the
00:58:05
risk of poisoning threat yes it can
00:58:08
enter into any part of the defense
00:58:11
enterprise
00:58:12
>> but it's just the defense enterprise so
00:58:15
Boeing wants to use anthropic to build
00:58:18
commercial jets have at it Boeing wants
00:58:20
to use it to build fighter jets I can't
00:58:23
have that because I don't trust what the
00:58:25
outputs may be because they're so wedded
00:58:27
to their own policy preferences.
00:58:30
>> I guess a dovetail to that is why
00:58:31
couldn't this have been handled
00:58:34
quietly? Is this anthropic who made this
00:58:36
a public spat or was it the
00:58:39
administration that made it a public
00:58:41
spat or two to tango? I mean, they have
00:58:44
a very good sophisticated press
00:58:46
operation and like really good and
00:58:50
painting us as doing mass surveillance
00:58:53
where where where their issue was like
00:58:56
some commercial database thing that
00:58:58
someone else could buy. They didn't want
00:58:59
us to buy to use it, which I'm not even
00:59:01
sure we buy them except to do recruiting
00:59:03
for soldiers. And you know, we run
00:59:05
schools, hospitals, we do a lot of
00:59:06
things at DoD. We don't just fight wars.
00:59:09
and um and the the way they were able to
00:59:12
characterize these two things which are
00:59:13
genuinely scary to people but were not
00:59:16
the real issues. Um it was really the
00:59:20
you worry I worried about them shutting
00:59:22
off our system at a moment of need or
00:59:24
them messing with our system in motar
00:59:27
>> came to mind is if they are selling you
00:59:30
batteries and you need to use the
00:59:31
batteries or the laptops however you
00:59:33
need to use them lawfully okay that
00:59:36
should be enough for them unless they
00:59:38
are peacenicks and they don't want to be
00:59:40
involved in selling weapons which by the
00:59:41
way was Google's position for many years
00:59:43
they just didn't want to be involved in
00:59:44
it because to your point they want to
00:59:47
recruit talent that is also aligned with
00:59:49
that. So there's just seems to be maybe
00:59:52
this isn't the right partner for the
00:59:54
Department of War.
00:59:56
>> Yeah, you should if you if you don't
00:59:57
want your stuff to be used for
00:59:59
department war stuff, you shouldn't be
01:00:00
selling to the Department of War.
01:00:02
>> Pretty sure it's in the name. It's in
01:00:05
the name.
01:00:06
>> Well, and then also I have to say when
01:00:09
you know you said, "Hey, we don't know
01:00:10
what how we're going to use this thing."
01:00:12
Like immediately came to mind was like
01:00:14
911. you you have to go check with them,
01:00:17
you know, if you find out there's
01:00:19
another 911 unique, you know, black swan
01:00:22
event that's going to occur and you have
01:00:24
to go clear it with them. Like you
01:00:26
That's
01:00:26
>> That was literally the comment. That was
01:00:28
literally the comment when I was Yeah.
01:00:30
So I was in a room of 20 people. So this
01:00:31
is not undeniable if everyone want Daria
01:00:34
wants to deny it.
01:00:36
>> And I was giving these scenarios, these
01:00:37
Golden Dome scenarios and so on. And
01:00:39
he's like, "Just call me if you need
01:00:41
another exception." And you know, I'm
01:00:42
like, but what if the balloon's going up
01:00:45
at that moment and it's like a decisive
01:00:47
action we have to take. I'm not going to
01:00:49
call you to do something. It's like not
01:00:51
rational. And
01:00:53
>> yes,
01:00:54
>> uh so it that was another holy cow
01:00:56
moment of like how they think about it.
01:00:59
>> That just means that what he wants to be
01:01:00
is the secretary of war.
01:01:02
>> That's right.
01:01:03
>> He wants to be the the god king there, I
01:01:05
guess. Yeah.
01:01:06
>> You can't do that. The thing that shocks
01:01:08
me, Emil, I don't know. you maybe you
01:01:10
can't say anything but guys you can
01:01:11
comment on this. It's clear that
01:01:13
Anthropic just lost all the Republicans
01:01:17
but I think that if they think that they
01:01:19
have the Democrats that's fleeting as
01:01:21
well because I think progressive
01:01:23
Democrats fundamentally just hate
01:01:26
Silicon Valley and technology and so
01:01:27
there's no way they're going to let some
01:01:28
god king over here that they don't
01:01:31
control either. And so in both ways, I
01:01:33
think they accidentally may have pissed
01:01:35
off every constituent. The longer term
01:01:37
fallout amongst them and progressives
01:01:39
will come home to roost because as the
01:01:41
progressives want more control and these
01:01:43
guys push back on them, they're just
01:01:44
going to fall into the same situation.
01:01:49
>> Yeah. I mean, it's an interesting
01:01:51
perspective. I think if you don't want
01:01:54
to be involved in war that you're right,
01:01:56
I think you mentioned this like three
01:01:57
times, Jimoth.
01:01:58
Don't sell bullets if you don't want to
01:02:00
be in but you can't call Smith and
01:02:02
Wesson and say can I The other thing is
01:02:05
what the hell was the senior management
01:02:07
and the board talking about over these
01:02:10
last few days because to me it would
01:02:12
have sounded insane. So then the
01:02:13
question is were people just so
01:02:15
breathless to buy this revenue curve?
01:02:17
What is the board doing? What is the
01:02:19
senior management really doing? What do
01:02:21
you change guys? What do you think you
01:02:22
would tell them if you were sitting
01:02:24
inside of the board of
01:02:25
>> Enthrop if you're an investor, you're on
01:02:27
the board, what do you say to Daario
01:02:29
when he says, "Hey, I need to dictate to
01:02:32
Emil and Hexth how they use my tool and
01:02:34
everybody else is just saying lawful use
01:02:36
as the standard." What's your coaching
01:02:38
advice?
01:02:39
>> Well, it's also a very unusual
01:02:41
circumstance because I don't think any
01:02:43
business in history has grown as fast as
01:02:45
they have in the last 90 days. So,
01:02:47
they've added what was it? 6 billion of
01:02:50
ARR.
01:02:51
>> Yeah. in a month or something.
01:02:53
>> I mean, that's absurd. Like, I mean,
01:02:55
absurd. It's absurd. It's a great
01:02:56
product. Open Claw has driven a lot of
01:02:58
this.
01:02:59
>> If you're on the board,
01:03:00
>> you're closing your eyes. Yeah.
01:03:01
>> You're shutting the up. You're just
01:03:03
shutting the up cuz something's
01:03:04
working.
01:03:05
>> You're actually
01:03:07
>> I think he's off doing his thing and
01:03:08
they're going to let him do it. And I
01:03:09
don't think that company's worth 350
01:03:11
billion anymore.
01:03:13
God knows what it's worth.
01:03:14
>> Oh. Oh, that's interesting. Where do you
01:03:17
If you get put a block of stock right
01:03:18
now, where do you put a bid in? I'll
01:03:20
tell you where I
01:03:20
>> Oh my god. I had I had this conversation
01:03:22
at dinner two nights ago. It's like you
01:03:23
have to pick between OpenAI at their
01:03:25
current mark,
01:03:28
anthropic at their current mark, or
01:03:29
Google. And it's either multiple from
01:03:33
here or net market value creation from
01:03:36
here because those are actually two very
01:03:37
different conversations.
01:03:38
>> Explain the difference.
01:03:39
>> I think the net market valuation because
01:03:41
Google's already worth three trillion.
01:03:43
So if they double, they've added three
01:03:45
trillion, but I think Google is the bet.
01:03:47
I think Google is the market value
01:03:49
creator bet. But I think anthropic is
01:03:51
the multiple bet. I think anthropic is a
01:03:53
trillion five market cap at the end of
01:03:55
the day.
01:03:56
Unless this blows them up,
01:03:58
>> you're still buying the 5x versus the 3x
01:04:00
kind of thing.
01:04:01
>> You'd buy the 5x instead of the 2x.
01:04:03
>> But if you get put a block of stock now,
01:04:04
do you buy it at the last post or do you
01:04:06
buy it at a discount or do you just say,
01:04:08
"Ah, I just buy it at the last post."
01:04:09
>> Anthropic is worth a lot more than 350.
01:04:12
That's for sure.
01:04:12
>> I I that it's undervalued compared to
01:04:14
chat.
01:04:15
>> They just added six billion in the last
01:04:17
month. And I will tell you anecdotally,
01:04:18
anecdotally,
01:04:20
>> I am everyone I talk to is on co-work.
01:04:23
Everyone is like gone deep on this.
01:04:25
Everyone's amazed and shocked and
01:04:27
actively using it. And everyone's saying
01:04:28
the same thing, which is anthropic, may
01:04:30
actually be fulfilling the promise of
01:04:32
AI. I will also say that it's only going
01:04:34
to take 90 days for Google to flip on a
01:04:38
virtual version of co-work. And once
01:04:40
Google like has this integrated with G
01:04:42
Suite and you have a virtual hosted
01:04:44
version of co-work sweeps the market
01:04:46
with this same competitor. But right now
01:04:48
Coowwork is such an incredible product
01:04:50
and everyone's saying the same thing.
01:04:51
It's like giving Elon giving truth to
01:04:53
AI. Elon said something with respect to
01:04:57
Grock which was that
01:04:59
he expects it to exceed all of these
01:05:01
coding models probably in the May spin
01:05:05
but for sure by June.
01:05:07
So to your point free like what like I
01:05:10
guess my question guys to you is like
01:05:12
what happens okay what do you guys do em
01:05:15
what do you do when all the models
01:05:16
asmtote let's just say by October of
01:05:18
this year let's just say I can guarantee
01:05:20
you just for the thought exercise by
01:05:23
October all the models are the same do
01:05:26
you just take a complexion of them all
01:05:28
and say great we're going to build some
01:05:30
governance layer around it and now we're
01:05:33
indifferent or
01:05:35
>> what do you do
01:05:35
>> I would love to be indifferent because
01:05:37
then I could compete on price, right?
01:05:39
And then then I have and then I have one
01:05:42
one main and one redundant or two mains
01:05:45
and I'd need at least two.
01:05:48
>> Yeah.
01:05:49
>> Anthropic's not going to be one of them
01:05:50
if they continue sort of with their
01:05:52
their sort of posture. So then it would
01:05:55
be three. And if one gets wobbly from a
01:05:57
policy scenario too because they all,
01:06:00
you know, except for Elon's is based in
01:06:01
San Francisco and has that that vibe to
01:06:04
it. So, uh, you kind of want to have two
01:06:06
or three at any given time. And yeah,
01:06:08
then I then you price compete them. I do
01:06:10
think Google has a long-term strategic
01:06:13
advantage, not because of not only
01:06:15
because of their consumer thing, but
01:06:16
because they have their own cloud.
01:06:18
>> So, between them, they don't have the
01:06:20
margin on top of the cloud that
01:06:22
Anthropical have to pass on. So, it's an
01:06:24
interesting economic uh
01:06:26
>> uh proposition from them.
01:06:30
And just to build on your point,
01:06:32
Freedberg, after you finish your uh
01:06:34
insightful comments here, pull this up,
01:06:37
Nick. Almost on quue, Freedberg. You're
01:06:40
such an oracle. Here is the announcement
01:06:42
from Google. Google Workspace is now
01:06:45
integrated for agents and 40 agent
01:06:48
skills were included today. Emilio,
01:06:51
you've been great today. Super honest.
01:06:53
Daario's uh position. I'm going to give
01:06:55
you some fast balls here. Daario says,
01:06:58
"The real reason the Pentagon and Trump
01:07:03
admin do not like us is that we haven't
01:07:05
donated to Trump. While Open AI Greg
01:07:08
have donated a lot, here's Claude's
01:07:12
answer to that claim." Here's nine
01:07:14
companies and their activities with the
01:07:18
administration from the inauguration to
01:07:20
attending the inauguration to the White
01:07:22
House CEO dinner to the Melania
01:07:24
documentary. If you go through and you
01:07:26
look at these nine companies, Microsoft,
01:07:30
Apple, Tim Apple, Nvidia, Amazon, they
01:07:33
have all participated.
01:07:35
There's one company that hasn't
01:07:37
participated and that's Anthropic. Are
01:07:39
is Anthropic being singled out because
01:07:42
they are not genulecting and because
01:07:45
they're not paying the cover charge.
01:07:46
People say this administration is pay
01:07:49
for play. That's the accusation he's
01:07:51
making. I'd say maybe there's a cover
01:07:53
charge. Nobody likes to pay it, but the
01:07:55
other companies have. What do you think
01:07:57
here?
01:07:57
>> I mean, it's literally one of the
01:07:58
dumbest things I've ever heard. I I
01:08:02
truly just because I'm like, I'm in the
01:08:05
Department of War. I need to win wars.
01:08:07
If you help me win wars and I don't have
01:08:09
to waste time transitioning you out, I'm
01:08:12
that makes me thrilled. Um, and it sort
01:08:14
of it's a criticism on me because it's
01:08:17
not like Trump, President Trump dipped
01:08:19
in and he's like, "Hey, Emil, by the
01:08:21
way, those guys didn't get any money.
01:08:22
you can't use them anymore. I mean,
01:08:24
obviously, it's sort of like like
01:08:26
invention in his own mind. It's like I
01:08:28
don't know if people sleep at night if
01:08:29
those thoughts get in there. Um and and
01:08:33
I was trying to work with them. Why
01:08:34
would I spend three months trying to
01:08:35
negotiate with them to get to a simple
01:08:37
standard if I would have just said,
01:08:38
"Okay, guys, you're out. Bye." So, I I
01:08:42
think it's just some internal
01:08:44
psychosis. That's the only way I can
01:08:46
explain that.
01:08:47
>> Okay. It could be on Dario that he's
01:08:49
antagonistic to the administration both
01:08:51
with respect to how he operates
01:08:53
commercially and it's also reflected in
01:08:55
the fact that he doesn't want to support
01:08:56
the administration.
01:08:57
>> I have a different theory.
01:08:59
>> I think that they have a massive
01:09:01
instance of co-work internally that
01:09:04
helps them come up with business
01:09:06
strategy. And I bet you there's like
01:09:09
some element of AI that says, "Yeah, you
01:09:11
should do it. Do it. It just makes
01:09:12
sense.
01:09:13
>> Zig where they zag and get more press."
01:09:15
And so now there's some some
01:09:17
cladbot telling them to basically tell
01:09:19
the department of war to pound sand.
01:09:21
It's gonna turn out to be the stupidest
01:09:23
decision.
01:09:24
>> Listen, if I was chairman of the board
01:09:25
of that company, I pull Dario aside and
01:09:27
I'd say, "Listen, you're obviously a
01:09:28
genius. We obviously have the best tool
01:09:30
in town. This is not a battle you can
01:09:33
win and it makes no sense. You're going
01:09:35
to come across as not being patriotic."
01:09:38
And Tim Cook is showing up for the
01:09:40
Melania premiere. Would it kill you to
01:09:42
support the president? Would he kill you
01:09:43
to show up? Look what happened when
01:09:46
Biden Look what happened when Biden
01:09:47
excluded Elon that ankled him. Show up
01:09:50
for the president. Show up for America
01:09:53
and be a patriot. You don't have to
01:09:55
donate, but be a patriot and show up for
01:09:56
the dinners.
01:09:57
>> That's terrible advice. Here's my
01:09:58
advice.
01:09:59
>> Okay, here's your advice. Okay.
01:10:00
>> Hey Dario, call a meal back right now
01:10:03
and say, "You know what? Sorry, weed up.
01:10:06
We're gonna own this and we're going to
01:10:08
put out a press release that says we
01:10:11
support our customers use of our models
01:10:15
to do everything and anything that's
01:10:17
lawful. Number one, and number two, that
01:10:20
our terms of service are written in
01:10:22
stone and that you can expect solidity
01:10:25
and reliability from us. And this was
01:10:28
just a misstep.
01:10:29
>> Camille, how do you respond?
01:10:31
>> I mean, I would say that's what I've
01:10:33
always wanted. I need a reliable, steady
01:10:35
partner that gives me something that'll
01:10:38
work with me on autonomous because
01:10:40
someday it'll be real and we're starting
01:10:43
to see earlier versions of that and I
01:10:45
need someone who's not going to wig out
01:10:47
in the middle and we're just at the
01:10:49
early stages and it's rational. But
01:10:52
then,
01:10:53
>> you know, you called President Trump in
01:10:54
your 5,000word essay on Friday a wannabe
01:10:57
dictator.
01:10:59
>> You're going to have to apologize to
01:11:00
more people than just me. Yeah, maybe
01:11:02
time to rewrite the position here. Uh,
01:11:04
let let's just say Kumbaya, everybody.
01:11:07
Kumbaya, we solved the problem. And look
01:11:08
who's on the line. Surprise guest.
01:11:10
Daario's here. I thought I would
01:11:11
surprise everybody. Nick, pull Dario up.
01:11:13
No, he's not here.
01:11:14
>> What's your view on how the industrial
01:11:17
supply chain for hardware components and
01:11:20
systems is coming along in the United
01:11:21
States? Because my understanding is
01:11:23
we're trying to reduce dependency on
01:11:25
Chinese manufactured components. Where
01:11:27
are we with respect to where we need to
01:11:30
get to in the US manufacturing supply
01:11:32
chain?
01:11:33
>> We are early days. Um critical minerals
01:11:37
you see you've seen the action around
01:11:39
that. Um you'll start to see so I have
01:11:42
the office of strategic capital which
01:11:44
has 200 billion in lending authority.
01:11:46
And what we're trying to do is is it's
01:11:49
like a treasuries plus 100 bips loan to
01:11:52
companies, show them that the department
01:11:54
needs their solid rocket motors, their
01:11:57
batteries, their fiberglass, like all
01:11:59
the things that we that we're heavily
01:12:01
dependent on for our defense industrial
01:12:03
base that are completely outsourced to
01:12:05
China and domesticate them here. Um, and
01:12:09
we've got uh a bunch of great people
01:12:11
running it. So, but but it's early days.
01:12:13
is going to take for the rest of the
01:12:15
term to get um I think we'll get
01:12:18
critical minerals done before the rest
01:12:20
of the term where where we have the
01:12:23
access to what we need to from US or
01:12:26
allied countries. Um but from batteries
01:12:30
is like the next problem I'm trying to
01:12:31
solve. For example, batteries are
01:12:32
totally outsourced both technologically
01:12:34
and from lithium to China. Um, and
01:12:37
there's like, you know, kind of call it
01:12:39
20 critical things. If I could get to
01:12:42
all of them at some level, but then
01:12:43
it'll take a few years for them to like
01:12:45
build plants and do that stuff. But
01:12:46
there there it's it's very important. I
01:12:48
hope whatever administration comes next
01:12:50
continues it because I'm all free
01:12:53
market, but but we outsource so much
01:12:56
that um, you know, it crippled sort of
01:12:59
the the kind of the assembly part of
01:13:02
putting all these things together. Do we
01:13:04
have a munitions risk right now given
01:13:06
the conflict that we're involved in?
01:13:08
>> We don't have a munitions risk, but um
01:13:11
we do need to plus up because
01:13:14
the Europeans are taking a long time to
01:13:16
contribute like Ukraine. Russia has
01:13:18
consumed a lot of munitions from like
01:13:21
all over the world and then uh obviously
01:13:25
these conflicts we've been in and
01:13:28
um we need to have like the next
01:13:30
generation we're still there's still a
01:13:32
large degree we're fighting with 1980
01:13:34
cold war weapons
01:13:36
>> right and not modern weapons and so we
01:13:38
need to plus up those things that to to
01:13:41
regenerate them I mean our nuclear
01:13:43
missiles are 50 years old some of the
01:13:46
planes are 40 years old so all has to be
01:13:48
renewed.
01:13:50
>> Do you think um just speak to the
01:13:52
venture capitalists in the audience?
01:13:54
>> Are we in the early stages of this kind
01:13:56
of defense tech boom? Is defense tech
01:13:59
wellunded at this point or is it kind of
01:14:01
too hypy and bubbly and that's not
01:14:03
really the issue? It's not about funding
01:14:05
the companies. It's about funding some
01:14:06
of the further upstream uh issues that
01:14:08
we're facing. What's what's your view on
01:14:10
where we are there?
01:14:10
>> There's more defense tech venture
01:14:12
capital than ever by you know 3x more
01:14:15
than last year. So, you know, it it's
01:14:18
growing. What I need to do and what the
01:14:20
department needs to do is have some of
01:14:22
these companies win big contracts quick
01:14:24
like whether you know and sure um uh
01:14:28
Seronic sure like bunch of these
01:14:30
companies so that more money flows in
01:14:33
more entrepreneurs do it and I could buy
01:14:35
more because genuinely I do think
01:14:38
warfare is going from big car carrier
01:14:41
ships that cost20 billion dollars and a
01:14:44
decade and a half to build to mass
01:14:47
traitable
01:14:48
lowcost um things and that's what these
01:14:52
new these new entrance can do. So we
01:14:55
need those to succeed so that the
01:14:56
flywheel goes with venture capital money
01:14:58
entrepreneurs capabilities
01:15:00
>> in that sense and what I've heard as
01:15:02
kind of the explainer for this is we're
01:15:04
moving from the old primes to the new
01:15:06
primes that there's going to be a small
01:15:08
set of big winners and then obviously
01:15:10
lots of seconds and and subs and
01:15:12
whatnot. Is that really how this
01:15:14
market's going to evolve? So, are we
01:15:15
going to end up with Andrew, Palanteer,
01:15:17
and maybe three or four others, and
01:15:18
that's where most of the value is going
01:15:20
to acrue from a market perspective?
01:15:22
>> I mean, Andrew and Palanteer want that,
01:15:24
and I joke with them all the time about
01:15:25
it, but I I want I definitely want at
01:15:28
least a second layer that's innovative
01:15:31
and trying to disrupt the first layer
01:15:32
all the time. I met a mom and pop like
01:15:35
wholly owned company that that makes
01:15:37
these missiles called Rams that are we
01:15:39
really sell and send send to Ukraine and
01:15:41
they do it with like 30 people and they
01:15:43
can do a thousand a year because they've
01:15:45
designed a manufacturer and it's
01:15:46
awesome. So I want companies like that
01:15:48
to continue innovating maybe and then b
01:15:52
buys them but the the one of the reasons
01:15:54
the primes are such a small number it's
01:15:57
not the only but it's one is they learn
01:15:59
how to contract with the government.
01:16:00
They learned how to go through the
01:16:01
bureaucracy and that became a
01:16:04
competitive advantage. I'm trying to
01:16:05
take that competitive advantage away.
01:16:07
>> That's a really important point. How do
01:16:09
you disassemble all that bureaucracy so
01:16:12
that product innovation can actually get
01:16:14
to you?
01:16:15
>> Yeah. So we we did a big So part of it
01:16:20
comes down to requirements reform. What
01:16:22
used to happen is people like oh we need
01:16:24
a new fighter jet. So, Army, Navy, Air
01:16:26
Force put in the requirements and were
01:16:28
you know it would we needed to be
01:16:31
stealthy to hold a missile to hold four
01:16:34
humans and you know it became this
01:16:37
unbuildable thing but the contractor
01:16:40
didn't care because they're getting paid
01:16:41
cost plus so like sure I'll fulfill your
01:16:44
requirements two years from now you're
01:16:46
like that was never engineered properly
01:16:48
it'll be another few years late and a
01:16:50
couple more billion dollars so we're
01:16:52
trying to change that to I tell you my
01:16:54
common operational problem. I need a
01:16:56
bunch of missiles that go 500 miles or
01:16:57
more that have this kind of blast. Come
01:16:59
to me with solutions, as little
01:17:01
requirements as possible on that side.
01:17:03
And on the contract piece, trying to get
01:17:05
to as close to commercial contracts as
01:17:08
possible. And this is going to take, and
01:17:10
this is where the startups are so good,
01:17:12
they'll do fixed cost pricing. They'll
01:17:14
do, you know, pay you don't pay me as
01:17:17
much if I deliver late. You pay me more
01:17:18
if I deliver early.
01:17:20
>> It's very disruptive to the existing
01:17:22
system. Yeah.
01:17:22
>> Super disruptive. But that's that's what
01:17:24
I'm I'm like waking up every day trying
01:17:27
to do.
01:17:28
>> So you could put out as something
01:17:29
saying, "Hey, the straight of horses is
01:17:30
super important. We need to keep it
01:17:32
open. We need these type of devices to
01:17:35
keep it open." But come to us with your
01:17:36
ideas and let them be creative
01:17:39
entrepreneurs as opposed to, you know,
01:17:41
just trying to goose the profits. Yeah.
01:17:43
It's really brilliant.
01:17:44
>> Yeah.
01:17:45
>> Emil, you also oversee DARPA. Yeah.
01:17:47
>> Yeah. DARPA is the father of the modern
01:17:49
internet and it's created a lot of
01:17:52
really critical technologies. Can you
01:17:53
talk about what's going on in there? Are
01:17:55
there interesting things that you think
01:17:57
our audience should know about that
01:17:58
you're trying to push forward?
01:18:00
>> I mean there's so it's probably my
01:18:02
favorite part of of my my office is like
01:18:05
because there that's where you it's sort
01:18:08
of like it's still a very honored
01:18:10
profession to be part of DARPA. like you
01:18:12
know being a being in government service
01:18:14
for a long time is sort of reduced in
01:18:16
its stature since the Manhattan project
01:18:19
now because now now if you're a great
01:18:22
ass you know uh someone who wants to do
01:18:23
rockets and stuff you go to SpaceX DARPA
01:18:26
still has the best of the best and so
01:18:28
the most creative ideas happen there one
01:18:31
of the things that they're working on
01:18:32
that's public is they're trying to use
01:18:34
biology to synthesize critical minerals
01:18:36
so so how do you so how can you just
01:18:38
pull them out of ground use biology to
01:18:40
do it so you don't need to do all this
01:18:42
crazy messy dirty refining that would
01:18:45
like change the game big time on our
01:18:47
ability to get the critical minerals we
01:18:49
need faster and leaprog the Chinese in
01:18:51
terms of tech. Um,
01:18:54
so they're doing a lot of that kind of
01:18:55
stuff. They're deep in cyber cyber
01:18:58
attacks are are the next huge threat
01:19:00
with AI, right? The what what we saw
01:19:02
with the creating all these agents to
01:19:04
attack systems that anthropic happened
01:19:06
to them. Um, so they're they're they're
01:19:09
working on that's there's not a ton I
01:19:11
can talk about kind of DARPA because
01:19:13
it's so it's so classified, but those
01:19:15
are a couple examples for you.
01:19:18
>> All right, speaking of classified, uh,
01:19:19
just two quick questions before we wrap
01:19:21
here. Are there aliens? And what are you
01:19:22
going to tell us? And number two, uh, in
01:19:25
all seriousness, I I'm curious, what
01:19:28
have you learned about China and where
01:19:30
they're at and the threat there and our
01:19:32
ability to counter it? like give us some
01:19:35
idea of where we're at as a country cuz
01:19:38
we hear a lot of hyperbolic stuff.
01:19:41
They're building this incredible mobile
01:19:43
small navy. They've got hypersonics.
01:19:45
They're just way ahead of us. You know,
01:19:47
we hear these things. But realistically,
01:19:49
are we competitive?
01:19:51
>> Um I fought Well, I'll answer your first
01:19:53
question, which I fought for the alien
01:19:55
portfolio. I didn't get it.
01:19:58
>> Work to do more work to do.
01:20:01
All the guys on my team were like,
01:20:02
"Dude, you got to get this for us.
01:20:04
Please talk to the secretary. We want to
01:20:06
do this."
01:20:08
>> But I but I I was like, you as long as I
01:20:10
had 100% access to everything, I would
01:20:12
do it because that would be it would be
01:20:14
amazing, right?
01:20:15
>> Sabers would be a game changer.
01:20:17
>> Um but on the second one, uh it is true
01:20:20
that Chinese have had the greatest
01:20:23
military buildup in world history in the
01:20:25
last 15 years. and we're asleep at the
01:20:28
wheel to some degree because we're
01:20:30
focused on global war and terror. So,
01:20:32
they've advanced without sort of us
01:20:34
thinking about threat. That being said,
01:20:37
our operational expertise and our space,
01:20:40
like we have some sophisticated stuff,
01:20:44
you know, our subs, our space layer, um
01:20:47
we still have the best stuff in the
01:20:49
world that does, you know, but but we
01:20:52
have to make sure that gap doesn't
01:20:53
narrow,
01:20:54
>> right? We can't be complacent. We should
01:20:57
sleep well at night knowing you're
01:20:58
there. Yeah.
01:20:59
>> Knowing President Trump's allocating
01:21:00
money towards this and he's decisive in
01:21:02
his actions. But we cannot be complacent
01:21:04
trying to I feel like this week was a
01:21:07
true reminder of how fortunate we are to
01:21:09
have the defense that we have for the
01:21:11
United States. When you look at what
01:21:13
happened in Dubai and in Doha and in Tel
01:21:17
Aviv and you see how people in their
01:21:19
residential homes are getting attacked
01:21:22
and bombed, you realize just how
01:21:24
fortunate we are to have all of the
01:21:25
layers of protection that we have by our
01:21:27
government. And I've actually come
01:21:29
around to this quite a lot.
01:21:30
>> I'm a true kind of arguably libertarian
01:21:33
at heart, small government, but the one
01:21:35
thing that I've realized is so critical
01:21:37
for us to have the freedom to do all the
01:21:39
things we want to do is defense. And so
01:21:42
I think it's an amazing institution,
01:21:44
very valuable to the United States.
01:21:46
Emil, thank you for what you do.
01:21:47
>> Yeah, thank you. Really appreciate you
01:21:49
coming on and being so candid and
01:21:51
thoughtful and insightful. This has been
01:21:53
a fun amazing episode. We'll see you
01:21:56
next time. Byebye.
01:21:57
>> Love you, boys. Byebye.
01:22:00
>> Let your winners ride.
01:22:08
We open sourced it to the fans and
01:22:09
they've just gone crazy with it.
01:22:11
>> Love you. Queen of
01:22:15
>> yours.
01:22:20
Besties are gone.
01:22:23
>> That is my dog taking your driveway.
01:22:28
>> Oh man. Myasher will eat me.
01:22:30
>> We should all just get a room and just
01:22:32
have one big huge orgy cuz they're all
01:22:34
just useless. It's like this like sexual
01:22:35
tension that you just need to release
01:22:37
somehow.
01:22:41
>> Your feet.
01:22:44
We need to get merch. I'm going all in.
01:22:53
I'm going all in.
