00:00:00
hey everybody hey everybody before we
00:00:01
drop this week's amazing episode just
00:00:04
some quick information for you saak and
00:00:07
chamoff both had the week off so we had
00:00:10
two Amazing Friends of the pod on Gavin
00:00:12
Baker from at trees and Joe Lonsdale
00:00:14
from 8 VC what a treat to have them on
00:00:16
the program but there is some big news
00:00:19
huh freeberg yeah well the news dropped
00:00:21
after we recorded so it's not going to
00:00:23
be referenced and the show might feel a
00:00:25
little stale but here we go our friend
00:00:28
David Sachs is the White House Ai and
00:00:31
cryptar of the United States of America
00:00:34
congratulations to our boy amazing super
00:00:36
thrilled that is a big reason why he
00:00:38
wasn't able to join us for the show he
00:00:39
was in the middle of getting this news
00:00:41
ramped up yes it came out we had already
00:00:43
recorded yeah take that into account as
00:00:45
we go into the show yes and uh according
00:00:47
to Trump's truth social post saxell
00:00:50
quote guide policy for the
00:00:52
administration in artificial
00:00:54
intelligence and cryptocurrency he will
00:00:56
also lead the presidential Council of
00:00:58
advisers for science and Technology huh
00:01:01
how great is that freir wow science
00:01:03
Corner in the White House can't wait
00:01:05
it's going to be fun ex pretty awesome I
00:01:08
have no announcements uh the rumors
00:01:10
about me becoming press secretary are
00:01:12
obviously premature but if ask to serve
00:01:14
I will serve my country okay let's get
00:01:16
to it don't worry besties all in isn't
00:01:19
going anywhere we'll be here every week
00:01:21
for you except maybe Thanksgiving or a
00:01:23
holiday now and again all right let's
00:01:25
start the show huh free bar great
00:01:26
episode let's go all right everybody
00:01:29
Welcome to the all in podcast I am your
00:01:32
host Jason calacanis how you doing
00:01:34
freedberg I'm hanging in there I'm
00:01:36
waiting for the tsunami to hit it's
00:01:38
gonna we're about 40 minutes away I'm a
00:01:40
little anxious right now more anxious
00:01:43
than normal is what you're saying like
00:01:44
on a scale of one to freeberg this is
00:01:48
like as anxious as you get with this
00:01:50
tsunami I don't know what's gonna happen
00:01:52
this current warning shows a 5 foot
00:01:55
water rise or 5 meter I can't tell God I
00:01:59
really hope we don't this episode and
00:02:00
there's like a total
00:02:03
disaster it we're not going to make
00:02:06
light of it it we get warnings once in a
00:02:08
while just to give everybody a little
00:02:09
perspective here I'm at David Sax's
00:02:12
house I've taken over as you can see I
00:02:14
got my Monclair hat on
00:02:16
freeberg I uh I found Sax's robe see
00:02:20
this year it was Sax's robe but I got a
00:02:22
red Sharpie and I just crossed it out
00:02:23
and I put Jal on it and then I I went
00:02:26
down and I was talking to Chef does sax
00:02:29
know that you're staying at his house he
00:02:30
knows I'm staying here but does he
00:02:32
though he doesn't know that I found the
00:02:34
actual
00:02:37
cavar let your winners
00:02:39
[Music]
00:02:44
ride we open source it to the fans and
00:02:47
they've just gone crazy
00:02:48
[Music]
00:02:52
with with us today chamath couldn't make
00:02:55
it this week and had surgery and now he
00:02:57
looks like Gavin okay yes yeah
00:03:02
and saaks is taking a day off he's he's
00:03:06
he's got the day off today I think he's
00:03:08
just winning too much with us two
00:03:10
substitutes to jump in here for the team
00:03:13
cackling you can hear uh the cackling
00:03:15
Joe Lan there's Joe lale take my mon
00:03:17
Clair hat off here
00:03:19
J I'm also winning too much but I'm
00:03:22
happy to be here last minute as a as a
00:03:24
sub for David J J Joe lale for people
00:03:26
who don't know is to the right of David
00:03:29
saaks he is venture capitalist and he
00:03:31
started a couple of companies you might
00:03:33
have heard of them paler I what are the
00:03:35
other Greatest Hits adapar adapar just
00:03:37
across trillion dollars reported on it
00:03:39
this month it's a good real company too
00:03:41
yeah open go sold this year we we saw
00:03:43
start of a bunch of Cong good yeah um
00:03:45
Joe uh gets he's you're a little bit
00:03:48
lower profile than all these
00:03:50
accomplishments it's pretty incredible
00:03:52
co-founded how many billion dollar
00:03:54
companies have you co-founded now I
00:03:55
don't know maybe six or so six or so uh
00:03:58
also with us Gavin Aker from a treaties
00:04:01
he is a hedge fund manager he does
00:04:04
Private he does public one of the
00:04:06
smartest guys I know great analyst
00:04:08
welcome to the program Gavin friend of
00:04:09
the Pod thank you Jay Cal happy to be
00:04:12
here great to see everybody yeah we are
00:04:15
experiencing a huge Trump bump the
00:04:17
election is over the cabinet is being
00:04:21
assembled and we're seeing do the
00:04:23
department of government
00:04:24
efficiency as well as maybe regulations
00:04:27
being pulled down and and we'll we'll
00:04:29
get to to our first topic about our new
00:04:32
SEC chair but just generally speaking
00:04:34
aan as a market participant for a living
00:04:38
what's your take on what we've seen over
00:04:40
the last three weeks since the election
00:04:42
results came in so if they execute
00:04:46
unstated plans
00:04:48
and there are some of the world's
00:04:52
greatest execution machines involved um
00:04:56
you know Elon generally does what he
00:04:57
says he is going to do MH
00:05:01
um like I this is going to be awesome
00:05:04
for America for markets for the world
00:05:07
and the analogy I keep coming back to is
00:05:10
Sai Nella taking over a CEO of Microsoft
00:05:13
Microsoft was a monopoly incredibly
00:05:15
advantaged it had just been horribly
00:05:17
mismanaged for years all he had to do to
00:05:20
start winning was stop doing really
00:05:22
really dumb
00:05:25
things and that's an incredible place to
00:05:28
be you know America like we're the
00:05:29
greatest country country you know we've
00:05:30
got you know oceans on two sides
00:05:33
peaceful neighbors incredible natural
00:05:35
resources you know completely you know
00:05:37
can produce our our own food and energy
00:05:40
like in many ways most privileged
00:05:41
country on Earth but sometimes with
00:05:43
great privilege comes great like
00:05:45
stupidity and California to me would be
00:05:48
a leading example of that most in many
00:05:50
ways most privileged state in America
00:05:52
and a predator away with bad
00:05:55
policies and I do think one thing that
00:05:58
everyone of all political Stripes agrees
00:06:01
on is there are too many regulations
00:06:04
that result in far too many
00:06:06
administrators far too much complexity
00:06:08
and an inability to build things in
00:06:11
America so you know it was used very
00:06:15
effectively and as it should have been
00:06:17
against the Harris Administration that
00:06:19
they'd approved $42 billion for Rural
00:06:21
Broadband hadn't built anything had
00:06:23
approved $40 billion for Ev charges or
00:06:25
whatever it was hadn't done anything and
00:06:28
it's not that they didn't want to you
00:06:31
know dig trenches and do Broadband they
00:06:34
didn't want to make EV Chargers their
00:06:37
own regulations prevented them so I just
00:06:40
think
00:06:41
deregulation and
00:06:43
simplifying regulations and the tax code
00:06:47
is going to lead to an immense amount of
00:06:50
growth which is something that all
00:06:52
Americans should be happy about Joe you
00:06:54
want to maybe give us your take I know
00:06:57
obviously you're very vocal you're
00:06:59
obviously a conservative run a think
00:07:01
tank and have strong feelings on all
00:07:03
this but there is consensus I think
00:07:05
amongst all Americans that we don't like
00:07:07
fraud waste abuse and those items those
00:07:09
seem to be consensus building uh
00:07:11
efficiency and paying low lower taxes
00:07:13
all of that seems like something almost
00:07:16
all Americans most Americans can get
00:07:18
behind so I guess maybe a little bit of
00:07:21
your take on what we've seen in the
00:07:23
markets and the plan and then also are
00:07:25
you going to be involved in this at
00:07:28
all well listen listen J I think I do
00:07:30
agree I think almost all reasonable
00:07:32
Americans can see that our growth is
00:07:34
ridiculously constrained and I think
00:07:36
Jeff Bezos was saying yesterday like we
00:07:38
need a growth oriented mindset if we're
00:07:40
g to get out of our debt problems out of
00:07:41
our deficit problems so it's nice to see
00:07:43
everyone kind of coming on the team and
00:07:44
saying yeah we need to fix these things
00:07:46
I think what people don't understand is
00:07:48
like just how broken the government
00:07:51
bureaucracy bureaucracies and
00:07:52
regulations are right it's not like
00:07:54
they're kind of sort of bad like it's
00:07:56
it's almost like they're companies that
00:07:57
went bankrupt like think of the worst
00:07:58
company you know Silicon Valley they
00:08:00
went bankrupt like 30 years ago and
00:08:02
imagine if someone just like kept
00:08:04
pumping money into that worst company
00:08:05
you know over like 30 years to keep
00:08:07
hiring people so like Yahoo or AOL like
00:08:11
some Legacy
00:08:12
company was failing it take the worst
00:08:15
Department there and then like the worst
00:08:16
Department gets the most money so it's
00:08:18
like it's like it's it's like more I
00:08:21
guess you can use the word now
00:08:22
it's more than anything you've
00:08:24
seen in a long time and and I think one
00:08:26
of the important things is America used
00:08:28
to have these really hard test for 100
00:08:30
years you know this in 1883 we said you
00:08:32
know what we shouldn't just hire our
00:08:34
friends if someone's going to run
00:08:35
something in government let's have these
00:08:37
tough tests kind of like China China did
00:08:38
for thousands of years and and and for a
00:08:40
hundred years we had really hard test if
00:08:42
when we went to the moon when we fought
00:08:43
the world wars you know we did them
00:08:45
inhan project the people making those
00:08:47
decisions and hiring people had to pass
00:08:49
things that really only like five 10% of
00:08:51
people could pass who took them and then
00:08:53
in the late 70s we said not only are
00:08:56
these tests racist so we can't test
00:08:57
people anymore of any background we also
00:08:59
can't fire people anymore because we're
00:09:01
going to give them tons of protections
00:09:02
so for and so the last 40 years it just
00:09:04
got dumber and then 10 years ago you
00:09:06
started doing all the virtue signaling
00:09:07
and and you know hiring based on your
00:09:10
identity versus based on anything else
00:09:12
so so it's got even worse and so now
00:09:13
it's so broken that yes we're letting
00:09:15
tens of billions of dollars on Fire And
00:09:17
so there's really two things here one
00:09:18
you got to take a chainsaw as Elona said
00:09:21
you got to take a chainsaw and just like
00:09:22
cut a ton of broken stuff but then day
00:09:24
two is you got to say how do we make
00:09:26
this not stupid in the future and
00:09:28
there's things like maybe you bring back
00:09:29
tests maybe you bring accountability
00:09:31
thing I the thing I'm most passionate
00:09:33
about is you know right now there's over
00:09:35
a million rules to the federal level
00:09:36
it's stuff that everyone disagrees on
00:09:38
you can be the most left person trying
00:09:40
to build like solar or wind or whatever
00:09:42
and you're like what I have to do what
00:09:43
study over how many years this makes no
00:09:45
sense so so we gota we gotta we gotta
00:09:47
take these regulations and not only cut
00:09:48
them but we got to make a data driven
00:09:50
system that forces regulations to defend
00:09:52
themselves and that way instead of
00:09:53
having a cancer that just grows forever
00:09:55
out of control you could actually have a
00:09:57
process that like naturally things
00:09:59
naturally makes things fight for himself
00:10:01
and that that way doesn't get as dumb
00:10:03
ever again I would why not just make
00:10:05
them automatically sunsetting after five
00:10:07
years if they're not renewed exactly but
00:10:09
make the process to renew them difficult
00:10:10
and data driven exactly so they yeah
00:10:14
yeah exactly so if you renewing them
00:10:17
will take so much time that it will slow
00:10:19
it down I wonder where we got to the
00:10:21
place freedberg I will bring you in on
00:10:22
this since really I think it was maybe
00:10:24
two or three years ago you started to
00:10:27
point out exactly the death spir going
00:10:29
to get in I want to give you a flowers
00:10:30
here on the Pod because from this pod
00:10:33
and your obsession uh and you're harping
00:10:36
on and on about this national debt
00:10:39
problem you saw it early you talked
00:10:42
about it constantly you brought a lot of
00:10:44
consensus on board and now we're seeing
00:10:45
it as the issue of the transition is the
00:10:48
debt and we were sitting here for the
00:10:50
past year or two saying when are
00:10:52
politicians going to even pay attention
00:10:53
to this and now they are paying
00:10:55
attention to it so first here's your
00:10:57
virtual flowers
00:11:00
how are you feeling about the transition
00:11:02
to date well I mean I think first of all
00:11:04
there's like three layers of the problem
00:11:06
which I've been trying to Har on for
00:11:07
almost four years number one is the
00:11:09
inefficiency and lack of accountability
00:11:11
which Joe and Gavin are obviously
00:11:12
talking about and that leads to
00:11:14
excessive spending and that leads to the
00:11:17
debt problem and the debt problem
00:11:19
creates this kind of arithmetic debt
00:11:21
death spiral which is um something you
00:11:23
don't want to find yourself in so that's
00:11:25
the inevitability that I've been kind of
00:11:27
really worried about and I think that
00:11:28
you know look there's uh the first
00:11:30
derivative and second derivative can
00:11:32
kind of get addressed and then hopefully
00:11:33
you don't get to the absolute point
00:11:35
where you have this breaking breaking
00:11:36
point I I think right now everyone's
00:11:38
banking number one on can we deregulate
00:11:41
in a way that can unlock growth and by
00:11:43
unlocking growth the right the the the
00:11:46
the kind of arithmetic argument is grow
00:11:49
GDP because you're not going to be able
00:11:50
to shrink debt super fast uh in order to
00:11:53
grow GDP there there needs to be some
00:11:55
unleashing happening and so if you can
00:11:57
get GDP to grow four five
00:11:59
% and you can minimize the excess
00:12:03
spending meaning you can minimize the
00:12:04
deficit the federal deficit and
00:12:06
therefore minimize the increment in in
00:12:08
the debt level the debt to GDP ratio
00:12:11
becomes more manageable because
00:12:12
ultimately you can only tax so much of
00:12:14
GDP so yeah I feel like this is
00:12:17
important in both senses one is just cut
00:12:19
the inefficient wasteful spending get
00:12:20
rid of the regulations and that'll
00:12:23
unleash the necessary growth one of the
00:12:26
proxies I Look to and I I think that
00:12:28
this is going to be kind of
00:12:29
the critical motivating factor for the
00:12:31
United States whether it's this
00:12:32
Administration or the next or on a
00:12:35
continuous basis there's going to be
00:12:36
this kind of moment where we're going to
00:12:38
look across the water at what China has
00:12:42
and you can see what China's getting
00:12:44
what China is making what China is doing
00:12:47
I've talked about this a lot as well and
00:12:48
I do think this is the most critical
00:12:50
metric that no one talks about as much
00:12:52
as I think we should which is the
00:12:54
increment in electricity production
00:12:57
capacity in China compared to the United
00:12:59
States and so we are going from one
00:13:01
terawatt to two terawatts they're going
00:13:03
from I think two to eight over the same
00:13:06
time period and they're doing so
00:13:07
important for the people who are
00:13:08
listening who think you're saying I'll
00:13:11
let Joe I'll let Joe answer that Joe go
00:13:13
ahead just unpack it for the audience I
00:13:14
mean I obviously understand why this is
00:13:16
important but I want you to the number
00:13:18
one thing I'd say is it's usually
00:13:19
correlated to how well the working class
00:13:22
is doing in your country to per capita
00:13:24
GDP to cost of goods I mean obviously I
00:13:26
care about it because we want to scale
00:13:28
AI for all the things we're doing and we
00:13:30
want to be able to do it effectively but
00:13:31
if you just look over time it's really
00:13:33
clear the relationship between how well
00:13:35
is your average working middle class
00:13:37
person doing and their their quality of
00:13:38
life their cost of life and the cost of
00:13:40
electricity and cost of power and it's
00:13:42
it's crazy we're not just like ramping
00:13:43
up and cutting things to do this more
00:13:45
electricity means more automation means
00:13:48
more AI means more things are being done
00:13:50
for the for every person that are being
00:13:53
done in factories or being done by
00:13:54
machines and that
00:13:56
unlocks um a new kind of level of living
00:14:00
and that's been kind of a continuous
00:14:01
process for humans there's this guy that
00:14:03
writes these books that Bill Gates
00:14:05
always talks about V Smith smear name SM
00:14:10
SM he's got all these books on the
00:14:12
history of the relationship between
00:14:13
energy and kind of prosperity there's
00:14:16
definitely correlation there and there's
00:14:18
definitely causation there Gavin and I
00:14:21
there's even like really simple ones I
00:14:22
don't know if you've ever seen Lewan Yu
00:14:24
from you know the founder of Singapore
00:14:26
essentially talk about just how air
00:14:29
conditioning change the country's fate
00:14:33
it raises average IQ by two or three
00:14:35
points it really matters but look I
00:14:37
meend day at the end of the day my my
00:14:40
key point was we have to make sure that
00:14:43
this Administration and Joe can hear me
00:14:44
on this and I know he he believes this
00:14:47
but for me it has to be such a priority
00:14:49
that we um accelerate a nuclear uh
00:14:52
energy roll out in the United States
00:14:54
because that's what China's doing they
00:14:56
have dozens of Gen 4 reactors that are
00:14:58
going to get built built out Each of
00:14:59
which has a gwatt of production capacity
00:15:02
and they're doing it at a cost that we
00:15:04
can't compete with today but
00:15:06
fundamentally they can do it in the US
00:15:08
we can't even do it and regulations it
00:15:11
goes back to Joe yeah and so so regul
00:15:13
the regulatory structure prohibits our
00:15:14
ability to actually expand energy
00:15:16
capacity or electricity production
00:15:17
capacity which is a critical difference
00:15:19
and that ultimately leads to a situation
00:15:22
where in 10 years we're going to be
00:15:23
looking across the water at you know a
00:15:27
competitive country that has 3x 4X the
00:15:31
electricity production capacity of our
00:15:33
country and everything is cheaper
00:15:35
everything is faster it gives them
00:15:37
superiority in a lot of functions that
00:15:38
we can't this this is a key national
00:15:40
security thing too because we need
00:15:41
manufacturing here to be affordable and
00:15:43
competitive in order to for National
00:15:45
Security you know David when you when
00:15:46
you texted me last minute to join this I
00:15:47
was actually in a meeting with our
00:15:48
friend from Founders fund who's building
00:15:50
nuclear fuel again for the first time in
00:15:52
the US that's actually gone away too and
00:15:54
so so that needs to be approved but you
00:15:55
know I know Chris Wright he's a nominee
00:15:57
for the new you know energy a Secretary
00:16:00
of Energy and he you know he's super pro
00:16:02
Liberty guy but super pro cheap cheap
00:16:04
energy guy and he's Pro Nuclear So I
00:16:05
think we have some really great people
00:16:06
who care about this we're gonna be
00:16:07
fighting hard for it what's what's your
00:16:09
read on it yeah what's your read on
00:16:11
energy and the blockers yeah I mean more
00:16:14
nuclear more better like I
00:16:16
mean I agree with everything Joe said
00:16:18
and the only thing I would just add is
00:16:20
it is the most environmentally friendly
00:16:23
kind of energy source some way it's even
00:16:27
more like I I think think it is highly
00:16:29
likely that in my lifetime the world
00:16:31
just runs on solar like if you just you
00:16:35
know we all know compound interest is
00:16:37
the greatest force on Earth but if you
00:16:39
just look at the rate at which
00:16:40
photovoltaic cell efficiency is
00:16:43
compounding battery efficiency is
00:16:45
compounding and people make these
00:16:46
balance of system
00:16:48
arguments but it it it will it will
00:16:50
never be as cheap as nuclear but it will
00:16:52
likely approach coal and I think a lot
00:16:55
of the world will run on solar but
00:16:57
that's going to take 50 years um nuclear
00:17:00
is arguably just as environmentally
00:17:02
friendly done right and carefully and it
00:17:04
is here now and so I
00:17:07
just yeah I mean that's yeah I I don't
00:17:10
want I it is unbelievable to watch not
00:17:14
exactly Mor's law but this precipitative
00:17:16
drop in the cost of just solar panels
00:17:20
solar solar's very solar's very
00:17:22
impressive but in some ways to me it's a
00:17:23
little dystopian to think about all
00:17:24
these forests just covered with this
00:17:26
stuff I it's rate limiting it's rate
00:17:28
limitting you have you have to have many
00:17:31
many acres rolled out and with nuclear
00:17:34
you can have a building that can you
00:17:36
know power the equivalent of many acres
00:17:38
so the answer's both the answer is both
00:17:39
but it is like Elon has tweeted many
00:17:41
times about the tiny
00:17:44
fraction of you know deserted um desert
00:17:48
areas of America that need to be covered
00:17:50
with panels and then you put in
00:17:52
batteries there's still a space limiting
00:17:54
like let's just fast forward a 100 years
00:17:56
on planet Earth and if you look at the
00:17:58
past 100 years and the 100 years before
00:18:00
that like energy demand uh even on a on
00:18:04
a per capita basis has this nonlinear
00:18:07
kind of scaling problem and that means
00:18:09
that land consumption will scale
00:18:11
nonlinearly now we have it if we're
00:18:13
gonna say 100 years though I mean so
00:18:14
interrupt you're gonna say 100 years
00:18:16
technology changes it's very clear you
00:18:17
could probably do it from space if you
00:18:18
really wanted to I mean this sounds
00:18:20
crazy but you probably could have like
00:18:21
tons of these in space 100 years is far
00:18:24
I mean we have we have enough uranium in
00:18:26
just the crust of the you know like
00:18:28
Norther hemisphere to in other words
00:18:30
power to power everything we on thege of
00:18:33
unlimited energy for all time just to
00:18:36
wrap this up and move us on if we can
00:18:38
get out of our own way get out of our
00:18:39
own way with a bunch of midwit who have
00:18:43
put so much regulation in place and who
00:18:45
are working from home for four hours a
00:18:47
week if we can just get them out of the
00:18:49
way then maybe we could approve some
00:18:51
nuclear reactors and a little more solar
00:18:53
batter did you just call did you just
00:18:54
call government workers midwidth is that
00:18:57
just the ones who are block
00:18:59
if you're not blocking and you're
00:19:00
working 50 hours a week God bless a lot
00:19:03
of people working hard in the government
00:19:04
I that's exactly my point but there's
00:19:06
some number of them who are just
00:19:08
obviously do not see the forest through
00:19:09
the trees by the way on that point Jal
00:19:11
it is interesting Texas is the number
00:19:13
one solar producer in the country and
00:19:15
it's not because everybody there is more
00:19:18
environmentally conscious no we're not
00:19:20
that liberal subies it's just it's easy
00:19:24
to build stuff in Texas build solar
00:19:26
power plants this is I mean for the Libs
00:19:30
in cities who are so Pro solar and anti-
00:19:32
natural gas or whatever yeah look at
00:19:34
Texas they're just build as much as you
00:19:36
want and where Joe and I live in Austin
00:19:39
home prices have gone down two years in
00:19:41
a row rent has gone down two years in a
00:19:44
row they are building like lunatics
00:19:46
because you don't need to beg and and
00:19:50
bribe people to build stuff Joe maybe
00:19:52
some thoughts on what we've seen in
00:19:55
regulation Austin Austin city council is
00:19:57
not perfect Austin they called the
00:19:58
blueberry and the tomato soup so there
00:20:00
may be a tiny bit of begging going on
00:20:01
but despite that you're able to build
00:20:02
there and everywhere else around You're
00:20:04
Building like M so it works there's
00:20:06
there's there's plenty of Supply I agree
00:20:07
speaking of Regulation Gary Gensler is
00:20:10
out Paul Atkins is in uh and Bitcoin
00:20:13
just cracked a 100 this is all obviously
00:20:15
related Atkins was previously SE
00:20:17
commissioner under Bush 2 in the early
00:20:19
2000s in the 90s he worked for both Bush
00:20:21
one and Clinton at the SEC according to
00:20:24
the New York Times Atkins is admired
00:20:27
among DC Legal C circles and Regulators
00:20:29
he's Pro crypto and he's been helping
00:20:32
draft some best practices for the crypto
00:20:34
trading platforms as you know Gary
00:20:37
gensler's approach to crypto was there's
00:20:41
a rule set follow the rule set we're not
00:20:43
here to change the rules we're here to
00:20:45
enforce them good luck and in some cases
00:20:49
I think maybe he was right with icos and
00:20:51
a bunch of scams but he also gave good
00:20:54
actors no path to go forward anybody
00:20:57
have strong feelings on this I love Paul
00:20:59
J keelly I think he's I think he's gonna
00:21:00
he's a fair guy yeah I've met him
00:21:02
several times at conferences and groups
00:21:04
and he's really smart guy cares about
00:21:06
the rules cares about helping innovators
00:21:08
I the thing that really pissed off you
00:21:09
wenter off with Gary I mean I love
00:21:10
you've probably seen like I think I
00:21:12
think uh I think both coinbase with
00:21:14
Brian and the winlos twins that put
00:21:16
something out they won't even hire
00:21:17
anyone who worked with Gary on any of
00:21:19
this stuff they were doing and the
00:21:20
reason they're that angry at them is
00:21:22
they were purposely not defining the
00:21:24
rules in certain cases and then going
00:21:26
after people after not having to it in
00:21:29
order to kind of like play a Goa game
00:21:30
that was very dishonorable and Paul sees
00:21:32
all that and there's no way he's going
00:21:33
to allow that Gavin any thoughts here on
00:21:36
how this might change regulations in our
00:21:38
business also the fund business Joe is a
00:21:41
venture capitalist hvc uh you in public
00:21:44
markets and funds I have I do preedee so
00:21:47
yeah what do you think in terms of funds
00:21:49
and regulations there and then crypto
00:21:51
and and you know the SEC may be becoming
00:21:54
Innovative as opposed to punitive and
00:21:58
you know
00:21:59
what's the word for freed BG what's the
00:22:02
word I'm looking for here adversarial
00:22:05
well they are a regulator they are a
00:22:06
regulator but they don't also seem they
00:22:09
seem to have gone beyond just regulating
00:22:11
they seem to have been aggressively
00:22:13
adversarial yeah I mean they're
00:22:15
regulatory enforcement agency that's
00:22:16
their job so I I don't know like you
00:22:19
know I mean but not giving a path or not
00:22:21
even meeting with people I think that
00:22:23
was the thing that I felt was kind of
00:22:24
weird like they meet with you if you
00:22:26
were a big donor to the left they
00:22:27
wouldn't meet with you otherwise that
00:22:28
was part of it too that was sketchy that
00:22:29
was why SPF got tons of meetings but
00:22:31
Brian got none right I mean it's
00:22:32
nonsense like that ah go I do think
00:22:35
cryptocurrencies yeah go ahead at some
00:22:37
level fundamentally reduce the power of
00:22:40
nation states and that is something that
00:22:43
is of professed by the True Believers
00:22:47
but it is true and so if you um are on
00:22:51
the left and you know you are um a
00:22:54
devout believer in the power of the
00:22:55
state to do good things I get why you
00:22:58
not like
00:22:59
cryptocurrencies you know at the same
00:23:01
time we're very early in crypto I do
00:23:03
think I read with interest all of the
00:23:06
post that David Marcus and his peers at
00:23:09
Libra made um made about what happened
00:23:12
to them Libra was a Facebook project to
00:23:15
embrace cryptocurrency at a very
00:23:18
intrinsic level onto the sort of
00:23:21
identity level of every one of their
00:23:23
billions of users and I think you can
00:23:25
argue it would have been really really
00:23:26
good for not just America but the ENT
00:23:28
World um you know there are a lot of
00:23:31
immigrants in American in America who
00:23:33
send remissions back to their home
00:23:35
countries at extremely high um predatory
00:23:39
rates yes yeah predatory rates it would
00:23:42
have made that free and that would have
00:23:43
been amazing for a lot of really
00:23:46
hardworking um people all over the world
00:23:49
it would have you know Visa Mastercard
00:23:51
they do charge big fees I mean not Visa
00:23:54
Mastercard do not charge big fees but
00:23:55
the um credit card complex and aggregate
00:23:58
is a is a reasonably big fee I mean
00:24:00
everybody oh it's just you know two two
00:24:02
and a half percent to make it you know
00:24:03
perfect and safe um I think Facebook had
00:24:06
a sound argument that they could have
00:24:08
done that cheaper and that would have
00:24:09
been an efficiency game for America and
00:24:12
it really did bum me out you know to
00:24:14
read some of the
00:24:16
letters that they sent the way they
00:24:19
killed Libra if anyone doesn't know and
00:24:21
maybe we can you can pull up David
00:24:22
Marcus's post they just all these
00:24:25
politicians sent letters to participants
00:24:28
and financial markets saying we don't
00:24:30
know if they are doing anything wrong
00:24:33
but we think they probably are and we're
00:24:36
going to look at this very closely and
00:24:37
we want to discourage you from
00:24:40
participating it was wor it was worse
00:24:42
than that it was like a mafia letter it
00:24:44
was like if you are to support this and
00:24:46
help with this we are going to look into
00:24:47
everything else you are doing and we may
00:24:49
then find some issues it was it was like
00:24:50
a mafia threat we can't stop you legally
00:24:53
but watch out like it was really sketchy
00:24:55
it was shered brown who who now is out
00:24:57
of office who led this senator from Ohio
00:24:59
but it was it was really bad it was
00:25:00
really bad what they were doing it was
00:25:02
really bad I found it upsetting as an
00:25:04
American and and it's you you kind of
00:25:07
hinted at this Gavin the fear of
00:25:10
governments is that they will lose
00:25:12
control of money supply and monetary
00:25:14
policy maybe unpack that a bit and then
00:25:16
freeberg will go to you on the same sort
00:25:18
of uh thread yeah I mean look it's a it
00:25:21
is a rational fear I mean controlling
00:25:23
monetary Supply like at some level the
00:25:26
you know the greatest you know Powers we
00:25:29
give the state are a monopoly on
00:25:30
violence to keep us safe um and you know
00:25:34
control over the money supply has you
00:25:36
know a means of exchange a unit of
00:25:38
account um and those are great Powers
00:25:41
particularly if you're America and your
00:25:42
reserve currency the only thing I would
00:25:44
just say to balance this out um and I do
00:25:47
think people are are very positive on
00:25:49
Paul Atkins I've never met him but the
00:25:52
reaction from people who know him like
00:25:55
Joe and who I respect has been very
00:25:57
positive is important to remember we
00:26:00
have the best Capital markets in the
00:26:02
world you know the US equity and fixed
00:26:05
income markets are the most trusted
00:26:08
places on Earth and we can always make
00:26:09
them better but just it is very you know
00:26:13
you you want to be very Vigilant about
00:26:15
keeping them fair and keeping out things
00:26:18
like inside information which makes
00:26:20
people feel comfortable doing business
00:26:22
here you know making you know having
00:26:24
investors have confidence in a company's
00:26:26
financial statements and those fincial
00:26:28
markets are one reason America is such a
00:26:30
great country and Gavin to put this in
00:26:32
context this was during a time period
00:26:36
this David Marcus Libra process when
00:26:39
they felt certain people in government
00:26:42
and I'm not saying I endorse this but
00:26:44
this is their position Zuckerberg had
00:26:46
too much power Zuckerberg was censoring
00:26:49
people the right felt they were being
00:26:50
censored the left felt that Cambridge
00:26:54
analytica and people were using
00:26:56
targeting the first Le the general Vibe
00:27:00
and even JD Vance is kind of was in on
00:27:03
this as well was too much power
00:27:06
specifically at meta too much influence
00:27:09
now they've got this many people this
00:27:11
penetration and the algorithm plus we're
00:27:13
going to give them money and they're
00:27:15
going to have power over people's wallet
00:27:17
and then what does the government have
00:27:18
they can't censor people they can't
00:27:21
control the message and they can't
00:27:22
control the pur strings that that is the
00:27:24
time period we're talking about here oh
00:27:26
no no I understand 100%
00:27:28
why they did it I just think as an
00:27:30
American at a minimum the way that they
00:27:33
did it yeah was to you know use one of
00:27:36
Joe's phrases dishonorable you know if
00:27:38
you if you want to say that we're going
00:27:40
to kill this then just like let's have a
00:27:42
debate as a nation absolutely don't do
00:27:45
it in Shadows yes rule of law rule of
00:27:48
law is also another reason America is a
00:27:51
great country yes freberg your thoughts
00:27:54
on Dave marcus' comments
00:27:58
and this sec pick and and just generally
00:28:00
a more Pro less regulation kind of
00:28:04
situation I think the is like one of the
00:28:07
most important agencies we have and I
00:28:10
think they're one of the best federal
00:28:12
agencies so I've worked with them and
00:28:14
I've worked with any other agencies and
00:28:16
there's all these issues with the SEC
00:28:18
but they play a very important job and
00:28:20
if you've worked in in foreign markets
00:28:23
and you've dealt with foreign Securities
00:28:25
Regulators you're going to be like thank
00:28:26
God for the SEC like can't wait to go
00:28:29
back this whole crypto thing I there
00:28:31
there's there's a distinction I think
00:28:33
between Bitcoin and cryptocurrencies as
00:28:36
speculative kind of assets speculative
00:28:39
trading what do you see those
00:28:40
differences as well I I definitely
00:28:42
concur with Gavin I think Bitcoin
00:28:44
fundamentally is meant to be supposed to
00:28:46
be ultimately will become a real threat
00:28:48
to the US dollar and it's kind of ironic
00:28:52
that Trump had this declaration this
00:28:53
week yeah that he's going to put 100%
00:28:56
tariff on All These Bricks Nations that
00:28:58
try to participate in an alternative
00:29:00
currency to the US dollar the greatest
00:29:02
currency on earth when he literally
00:29:03
turns around and then says we're going
00:29:05
to support Bitcoin it felt like the
00:29:07
biggest irony of the week to me because
00:29:09
I do think Bitcoin is the big threat to
00:29:11
the US dollar and I do think that at
00:29:14
some point whether it's this
00:29:15
Administration or the next they're going
00:29:16
to wake up to that fact and maybe the
00:29:19
the Bitcoin does you know the network
00:29:20
State concept does emerge and that's
00:29:23
where we end up but I do think we want
00:29:24
to have and are going to have a strong
00:29:26
federal government in the United States
00:29:27
for quite some time
00:29:28
that's going to play an important role
00:29:29
in everyone's lives here and I don't
00:29:31
know if you can really just say let the
00:29:33
dollar you know be supplanted by Bitcoin
00:29:36
Bitcoin seems to be a more of a safe
00:29:37
haven asset and that seems to be the
00:29:39
trade that it's store value should be
00:29:41
kind of store value and alternative it's
00:29:43
just gonna take over gold what do you
00:29:45
think you think um the state has
00:29:48
reasonable concerns about crypto
00:29:50
competing with it and then maybe
00:29:52
specifically when you saw Trump talking
00:29:54
about hey congratulations on your
00:29:56
Bitcoin I did this for you on 100K and
00:29:59
taking credit for it at the same which
00:30:01
he should take credit for it he did it
00:30:02
he did that last $40,000 per coin and
00:30:05
then you see him talking about the
00:30:07
bricks and then we have the US currency
00:30:10
so which is a bigger threat to
00:30:12
American uh exceptionalism and Supremacy
00:30:16
on planet Earth Bitcoin or bricks so one
00:30:19
represents a move towards Liberty and
00:30:21
what one represents a move towards
00:30:22
authoritarianism if bricks were to
00:30:24
become dominant if China and Russia were
00:30:27
able to control global currency together
00:30:28
and and other players that is terrible
00:30:31
it's bad for the US for so many reasons
00:30:33
bad for us consumer for so many reasons
00:30:34
he's right to fight it at the same time
00:30:37
having you know to channel biology
00:30:39
having this like Pro Liberty Network
00:30:40
State like like forces in that direction
00:30:42
like for example I think rather than
00:30:44
just you know Hong Kong and Singapore
00:30:46
Hong Kong's been lost we should have
00:30:47
like more Hong Kong and singapor in the
00:30:48
west to compete with the US that'd be
00:30:50
good for all of us who are on the pro
00:30:51
Freedom side of the US and make the US
00:30:53
wealthier would show examples of of new
00:30:55
experiments that create great wealth and
00:30:57
then that's kind of the side of Bitcoin
00:30:58
is you want more experiments and
00:30:59
competition from the Liberty distributed
00:31:01
side you don't want competition from the
00:31:03
authoritarian side so to me it was very
00:31:05
consistent yeah well I would just say
00:31:06
Texas is the Singapore of America you
00:31:09
know and it is putting pressure on the
00:31:10
rest of America and I don't say that
00:31:12
just because I grew up in Texas um it's
00:31:14
just a fact I do think longterm Bitcoin
00:31:18
I do not think the bricks will ever be
00:31:20
able to replace the
00:31:22
dollar just rule of law even if it has
00:31:26
occasionally been corrupted in America
00:31:28
is very very powerful yes as opposed to
00:31:32
rule by law and but I do think
00:31:35
Bitcoin will at some point be a serious
00:31:38
threat to the US
00:31:40
dollar and that just is what it is and
00:31:43
we'll see how different administrations
00:31:45
react would that it's a check on it's a
00:31:47
check on the most aggressive mistakes
00:31:49
first of all if the debt is doesn't get
00:31:50
under control if our deficit is
00:31:51
ridiculous Bitcoin is a wonderful check
00:31:53
on that you actually want healthy
00:31:55
competition from something good because
00:31:56
you want to check the excess is and the
00:31:58
craziness you want you you need someone
00:32:00
coming from the outside to say no don't
00:32:02
do that let's just say at some point an
00:32:03
AOC like person gets in charge it could
00:32:05
happen in the next 20 years you need
00:32:07
some kind of check on the dollar and
00:32:08
it's much better to come from the
00:32:09
Liberty side than from the other side I
00:32:11
agree with all that I will just say on
00:32:13
AOC I thought it was very interesting I
00:32:15
think if you're a Democrat I think
00:32:17
you're probably largely heartened by the
00:32:20
kind of intellectual leaders of that
00:32:21
party their reaction to losing you know
00:32:24
it's focused around hey we do need to D
00:32:26
regulate it is too hard build Josh
00:32:28
Shapiro's been tweeting every three days
00:32:30
about how he's making it easier you know
00:32:32
to do business in Pennsylvania it used
00:32:34
to take 20 days to become get a
00:32:36
hairdresser license now it takes an hour
00:32:38
all that is good um but I actually
00:32:40
thought AOC who is a very talented
00:32:43
politician she's an extraordinary
00:32:45
Communicator she's an extraordinary
00:32:46
Communicator her reaction was like she
00:32:49
posted this on Instagram if you voted
00:32:52
for Trump I want to hear why I want to
00:32:55
hear the things you listen to that
00:32:57
convinced you
00:32:58
like I don't say this out of anything
00:33:00
other than genuine curiosity basically
00:33:02
clearly I am missing something and I you
00:33:05
know I want to listen and I just thought
00:33:07
that was a very interesting reaction
00:33:09
that is the proper reaction yeah
00:33:11
absolutely yeah I want to point out one
00:33:13
thing I did I was doing some research uh
00:33:15
before the show and I found this SEC
00:33:18
speech this is uh really fascinating
00:33:21
this is from 2007 from Paul and he was
00:33:25
kind of giving his State of the Union
00:33:26
here and he talking about doing some
00:33:30
self-reflection and he's talking about
00:33:32
accreditation rules in 2007 before the
00:33:34
great financial
00:33:36
crises the concept of economic risk and
00:33:38
return also affect a different proposal
00:33:41
of the commission the
00:33:42
SEC relating to private investment funds
00:33:45
specifically part of The commission's
00:33:46
Proposal proposal would add an
00:33:47
additional requirement for any natural
00:33:50
accredited person to have at least 2.5
00:33:52
million investments before he or she
00:33:54
could invest in private investment F in
00:33:56
a private investment fund like AE fund
00:33:58
or private Equity Fund other than a
00:34:00
venture capital fund the underlying
00:34:01
premise for the commission's proposal is
00:34:04
that these types of Investments are too
00:34:05
risky for individuals other than the
00:34:06
very rich therefore we would have to
00:34:08
presume that the non-rich are either
00:34:10
unsophisticated or lack access to
00:34:12
sophistication and it is simply not
00:34:14
tolerable to have these types of people
00:34:16
at risk of losing their money on a hedge
00:34:18
fund assuming that these premises are
00:34:20
true however what Evans does the
00:34:21
commission have to support the
00:34:23
conclusion that private investment funds
00:34:25
are the most risky what makes a hedge
00:34:27
fund fund or a private Equity Fund more
00:34:29
risky than a venture capital fund great
00:34:31
question and how does the risk profile
00:34:34
of a pulled investment compare with the
00:34:35
risks of investing in Securities of a
00:34:37
single issue for which this new 2.5
00:34:39
million standard is this is where it
00:34:41
gets super interesting many public
00:34:43
comment letters Express indignation at
00:34:45
the commission's proposal one
00:34:46
commentator wrote stay out of my wallet
00:34:48
stop trying to protect me from myself
00:34:50
stop presuming to know more than I do
00:34:53
about my own life risk tolerance and
00:34:55
financial sophistication the
00:34:56
commission's pr me very well prevent the
00:34:58
non-rich from losing their money in
00:35:00
private investment funds but it also
00:35:02
certainly will prevent the non-rich from
00:35:04
participating in any upside profits and
00:35:06
gains on these funds does this mean the
00:35:09
rich get richer while the non-rich
00:35:10
should be content to just hold their
00:35:12
place on the economic ladder this to me
00:35:15
when I saw this I was like you know what
00:35:18
I am feeling absolutely fantastic about
00:35:20
Trump if he stops these wars or stops
00:35:23
one out of two and keeps us out of you
00:35:25
know participating even though we're not
00:35:27
on the around with the other any new
00:35:29
ones and he removes regulations and we
00:35:32
have people in power who understand that
00:35:35
moving up the socioeconomic ladder is as
00:35:37
important as protecting the downside
00:35:39
risk especially when we see wealth
00:35:42
polarization this is a great pick now
00:35:45
the other picks there's some wackpack
00:35:47
picks in there I'll be totally honest I
00:35:49
don't know what the strategy is Joe I'll
00:35:50
ask you that in a minute but what do we
00:35:53
think um Gavin of this sort of approach
00:35:56
here which is really thinking
00:35:58
thoughtfully about what is sophisticated
00:36:01
and and what are we doing here what is
00:36:03
the outcome we're looking for yeah no I
00:36:06
think it's great and you could just I
00:36:07
mean the reality is a hedge fund that
00:36:09
runs with a lot of Leverage probably is
00:36:11
more risky I don't know that it's more
00:36:13
risky than a single security but
00:36:16
um like at the end of the day since he
00:36:19
wrote that letter it was an incredible
00:36:21
15-year run for private equity and you
00:36:25
know now all the big private Equity
00:36:27
firms are making a huge effort now um
00:36:31
when probably the business is more
00:36:32
mature and maybe the return
00:36:34
opportunities aren't what they were to
00:36:36
appeal to Main Street America like it
00:36:38
would have been cool if you know
00:36:40
Blackstone or KKR or whoever in 070809
00:36:45
2010 you know could have had their fund
00:36:47
up on like the Fidelity Marketplace for
00:36:50
subscription that probably would have
00:36:52
been good for America so I think his
00:36:54
comments are well taken imagine if Elon
00:36:56
could raise for SpaceX like you know
00:36:58
from regular Americans I'm sure he would
00:36:59
have loved to do that if it wasn't crazy
00:37:01
risky with the SEC right so there's all
00:37:02
these people they're blocking us from
00:37:04
letting the average person be part of it
00:37:05
it does keep them from climbing the well
00:37:07
flatter I think it's crazy yeah freeberg
00:37:09
any thoughts here I mean this has been
00:37:11
my pet peeve for a long time is letting
00:37:14
people do what they want with their
00:37:15
money if you don't allow people to take
00:37:17
risk with their money and move up from
00:37:19
poor to middle class from middle class
00:37:21
to upper middle class and maybe
00:37:22
eventually becoming affluent Jason
00:37:24
there's a lot of places people can
00:37:25
invest their money they can buy public
00:37:26
stocks there there's $20 trillion of
00:37:28
public stocks they can buy I don't think
00:37:30
that you're prohibiting people from
00:37:31
transitioning their wealth like banss by
00:37:34
not being able to buy private stocks in
00:37:35
fact I think it's more likely than not
00:37:37
that people are going to go market
00:37:39
Securities in private markets
00:37:41
and rip poor people off even worse and
00:37:43
that's why there are these regulatory
00:37:45
kind of barriers and I don't think that
00:37:47
it's necess look everyone says let's
00:37:49
make it everything free and everything
00:37:51
libertarian seems like a good idea until
00:37:52
someone gets punched in the face gets
00:37:54
ripped off someone dies from a drug you
00:37:56
know that's not properly tested and then
00:37:58
we're all like where were the Regulators
00:38:00
where were the agencies to protect the
00:38:02
individuals and I think that's the role
00:38:04
that these agencies are are kind of
00:38:05
providing and that's the reason these
00:38:07
rules are in place I don't think that
00:38:08
this is like one of these things where
00:38:10
oh Liberty's being denied I think it's
00:38:11
you know there's a there's a care
00:38:13
there's a careful line to
00:38:16
walk I'm generally pretty CC in most
00:38:19
things so I agree with what a lot of
00:38:20
David said and I do think if you were to
00:38:23
allow ordinary Americans to buy private
00:38:26
companies that are held to a lower
00:38:28
standard of disclosure and Reporting
00:38:30
than public companies like something
00:38:32
would have to change like you know just
00:38:34
hey if if a private company wants
00:38:37
to like public companies are you know
00:38:40
are held to certain standards for reason
00:38:41
TW the numbers and non financials and
00:38:45
yeah and there is a lot of fraud in
00:38:48
Venture there are unethical people there
00:38:51
are and smart diligent people can't even
00:38:54
find it all you know that's it's like
00:38:56
the smartest most diligent people are
00:38:57
still getting ripped off yeah I mean
00:38:59
look at FTX I mean I mean FTX ripped off
00:39:02
some very very intelligent friends of
00:39:04
ours did did any diligence at FTX yeah
00:39:08
but well which is an important part of
00:39:10
this but I mean if you had a
00:39:11
sophistication I mean there's such an
00:39:12
easy solution to this you just do a
00:39:14
sophistication test people take a 5 hour
00:39:17
course and they answer 50 questions the
00:39:18
end like a driver's license well I do
00:39:20
think it is a little that's why I talked
00:39:23
about private Equity these are extremely
00:39:25
sophisticated institutions who are
00:39:26
always in vesting alongside their
00:39:28
clients y um and they're buying
00:39:30
established businesses they're putting
00:39:32
leverage on them but like I think
00:39:35
private Equity is kind of a middle
00:39:36
ground and a on a pooled basis I think
00:39:40
you could argue and you know maybe those
00:39:41
funds would need to change their
00:39:43
reporting and their disclosures to deal
00:39:45
with kind of the average American but um
00:39:48
and by that I mean strengthen
00:39:51
it but I think private
00:39:54
Equity his comments were well taken
00:39:56
that's what I would say Joe what are
00:39:57
your thoughts yeah I mean I think the AL
00:40:00
here the alha of the room that's really
00:40:01
funny here is that you have the people
00:40:03
on the left saying the only people with
00:40:05
lots of money are smart enough to be
00:40:06
allowed to do this which I think is just
00:40:07
a very funny position to take if you
00:40:09
just like you have to have a few million
00:40:11
dollars to be sophisticated enough to be
00:40:13
allowed do this I like your idea JC of a
00:40:15
test like like there should be there
00:40:17
should be some other way of accessing
00:40:18
this if you really want to I me I agree
00:40:20
I agree listen I don't want to live in a
00:40:21
world where we're all constantly being
00:40:23
spammed to the average person by like
00:40:25
stupid Financial stuff that's just like
00:40:27
we would be scammed all the time there
00:40:28
probably should be some rules like I
00:40:30
agree I I'm not like a total libertarian
00:40:32
on this I think it'd be really annoying
00:40:33
but yeah but just to totally block
00:40:35
people from participating to me that
00:40:36
sounds
00:40:37
crazy like I mean a finance college
00:40:41
professor you know might um unless they
00:40:44
had another job might not you know be
00:40:47
able to
00:40:47
do
00:40:49
$250,000 let's talk let's talk about a
00:40:52
very important Public Market set of
00:40:56
transactions G I'd like your point of
00:40:58
view on Michael sailor's convertible
00:40:59
note issuances being used to bitcoin the
00:41:04
to which this morning this morning
00:41:06
Bloomberg reported is the hottest trade
00:41:09
in hedge funds right now Nick if you
00:41:11
could pull up the article and then jel I
00:41:13
know you've been an
00:41:14
outspoken opinion Setter on Michael
00:41:17
sailor's promotion of his Securities
00:41:20
actions we'd love your point of view
00:41:22
Gavin just say at some point this does
00:41:24
get too big for a while when it was
00:41:26
smaller
00:41:27
you know you could support the debt
00:41:29
sorry could you explain it could you
00:41:30
explain it to everyone Kevin yeah so
00:41:32
what he is doing is issuing debt and
00:41:34
buying Bitcoin with the premise that
00:41:36
Bitcoin is always going to go up and he
00:41:38
you know has made eloquent arguments why
00:41:41
why that is the
00:41:44
case no trees grow to the sky and I
00:41:49
think the interest expense on his
00:41:51
convertible notes is 75 million off the
00:41:55
top of my head and by the way I am I
00:41:57
could care less about micro strategy
00:41:59
like I'm not close to it I'm not
00:42:02
involved yeah I don't know any hedge
00:42:04
funds who own it no horse in the race
00:42:07
yeah yeah I I think a lot of hedge funds
00:42:10
are short micro strategy but I have no
00:42:11
horse in the race but the but his the
00:42:14
underlying
00:42:16
business that you know pays the interest
00:42:20
expense on the
00:42:21
debt only does $400 million a year in
00:42:24
revenue and it's you know high gross
00:42:25
margin Revenue
00:42:28
but I just
00:42:30
unless debt
00:42:32
investors have absolute confidence and
00:42:35
Bitcoin has collateral and I don't think
00:42:38
that's where fixed income markets are
00:42:42
yet he it will get to a point where it
00:42:45
is it is too big for the size of his
00:42:48
company and then yeah maybe he can over
00:42:50
collateralize it and you know have $10
00:42:53
in Bitcoin for every dollar of
00:42:55
debt but then like the the magic money
00:42:58
creation machine that you know I see
00:43:01
discussed on X breaks down because
00:43:04
that's like you know that's
00:43:06
[Music]
00:43:07
um that's very very different than what
00:43:10
is being discussed today Joe do you have
00:43:12
a point of you because Joe's got a run
00:43:14
by the way so Joe you want to close this
00:43:16
out defense people down here in La
00:43:18
listen listen I am very bullish Bitcoin
00:43:20
I love the all energy of our society
00:43:22
around it I I agree with what you said
00:43:25
Dave that there's like there needs to be
00:43:26
some barriers for the public taking
00:43:28
crazy risks and I think the risks around
00:43:30
how he's accessing this is actually very
00:43:32
unusual and does scare me a bit and
00:43:34
people need to do their research so they
00:43:35
shouldn't just like throw money without
00:43:37
studying it it it does scare me a little
00:43:38
bit how blly people are throwing money
00:43:40
at this without knowing the kind of
00:43:42
Leverage he's taking you know yeah Joe
00:43:44
anything else you want to share that
00:43:45
you're up to that you're excited about
00:43:47
before you head out uh well you know uh
00:43:49
I this this weekend is the annual
00:43:51
Defense form at the Reagan Library we on
00:43:53
the board it's the biggest biggest
00:43:54
defense event of the Year everyone's
00:43:56
coming and I I guess I guess I'm really
00:43:58
excited that America has like woken up
00:44:00
in assuming we can bring back more
00:44:01
advanced manufacturing here I think
00:44:03
there's enough top companies now with
00:44:05
andrel with seric we're gonna be
00:44:06
building thousands of these vessels for
00:44:08
the Navy with AI uh with epis we're like
00:44:10
turning things off you know 10 miles
00:44:12
away whatever fairly far away with
00:44:13
microwave radiation there some really
00:44:14
cool technology coming that actually is
00:44:16
going to make us be able to deter our
00:44:17
enemies so if you ask me six seven years
00:44:19
ago I was panicked we we're gonna like
00:44:21
get way behind China I'm feeling really
00:44:22
good about it now I think and and I
00:44:24
think I think you know just this the
00:44:25
Pete Heth pick uh you know is actually
00:44:28
someone I'm quite bullish on from
00:44:29
everything I'm hearing so I I think
00:44:30
things are going the right direction is
00:44:32
there a wholesale upgrade happening in
00:44:34
defense in the United States are all
00:44:36
systems and all strategies being
00:44:39
rethought right now uh using technology
00:44:41
and Innovation and kind of a
00:44:43
performance-based product mindset
00:44:46
leading uh kind of a reinvention of
00:44:49
Everything Is that what's going to
00:44:50
happen in the next four years Warfare is
00:44:51
fundamentally totally shifted we're
00:44:53
seeing some of this in Ukraine there's
00:44:55
all sorts of new ways you want to swarm
00:44:56
things on the LAN land swarm things in
00:44:57
the water swarm things in the air how do
00:44:59
they coordinate and how do they work
00:45:01
together how do you manufacture enough
00:45:02
of these things and and how do you use
00:45:04
electronic warfare and new ways to you
00:45:06
know it's basically created by this and
00:45:07
so that it's just a whole new way of
00:45:09
doing things and we are going that
00:45:11
direction too much of the money David
00:45:13
like 95% of the money is still going
00:45:15
towards like frankly like wasteful
00:45:17
Legacy like mostly things we don't need
00:45:19
plus
00:45:20
main but enough is Shifting and there's
00:45:23
enough good people fighting and as long
00:45:24
as we keep just allowing open
00:45:26
competition to say okay which is better
00:45:28
and just let the best things win as long
00:45:30
as we keep doing that I think it's going
00:45:31
the right way I'm feeling very good
00:45:32
about it is the Chinese position shift
00:45:36
in their technology strategy and their
00:45:38
system strategy gonna motivate a shift
00:45:40
here do you think shift underway already
00:45:43
I mean it definitely already has so we
00:45:44
don't we don't need big aircraft
00:45:45
carriers we don't need f35s we need
00:45:47
drones we need lasers I there's a ro I
00:45:50
think there's a role for carriers and
00:45:51
force projection I don't think we need
00:45:53
like incrementally a ton more of them I
00:45:54
don't I'm not I'm not like this radical
00:45:57
where you get rid of all this but yeah
00:45:58
on the margin I'd much rather have
00:46:00
10,000 more smart drones above and below
00:46:03
the water than like an in incremental
00:46:05
carrier right so there's there's on the
00:46:06
margin there's all these things that are
00:46:07
better uses of money and I think we're
00:46:09
pushing that way and is there a huge
00:46:11
tidal wave of venture money I was at a
00:46:13
dinner last night where there was this
00:46:14
conversation about defense Tech used to
00:46:16
be off limits and a lot of LPA say you
00:46:18
can't invest in defense companies that's
00:46:20
now changed or is changing and
00:46:21
everyone's kind of coming up with their
00:46:23
own defense Tech strategy do you think I
00:46:25
mean you were obviously in this cabin I
00:46:27
don't know if you're an investor in the
00:46:28
space but are you guys seeing a big
00:46:30
shift yeah I I obviously you know listen
00:46:32
there's only been nine unicorns I think
00:46:34
still at this point and I started three
00:46:35
of them and invested three of them the
00:46:36
first round so I'm obviously pretty
00:46:38
involved in in the space and I think
00:46:40
listen it helps me if there's more money
00:46:42
for my companies which is really great
00:46:45
um right I think there's probably the
00:46:46
right answer for the US is not going to
00:46:48
be like a thousand businesses or 100
00:46:49
businesses it's gonna be it's gonna be
00:46:51
like seven to 10 new primes one of
00:46:53
them's obviously andal I think one is
00:46:54
probably cic and epis we'll see but like
00:46:56
there's going to be 7even to 10 new
00:46:57
primes and that's what it's going to be
00:46:59
so there's going to be a lot of zeros
00:47:00
and a lot of bad Investments but yes
00:47:02
these seven to 10 new primes are going
00:47:03
to be huge and if you can get access to
00:47:04
them you're gonna do really well Gavin
00:47:07
how do you look at that market you agree
00:47:08
I I agree with everything um Joe said
00:47:11
the only thing I would just add on China
00:47:13
what we are doing by restricting their
00:47:16
access to Advanced um compute and
00:47:19
advanced networking if you have read or
00:47:22
watched the three three body problem
00:47:24
America is unfolding a sofon over China
00:47:27
yeah yeah that's great way I say it I
00:47:30
have been really impressed with um some
00:47:32
of the Chinese models that have come out
00:47:34
and I think the risk to this strategy is
00:47:36
necessity is the mother of invention and
00:47:39
despite this handicap they're managing
00:47:41
to stay just behind the Leading Edge of
00:47:44
America which is
00:47:45
amazing but you know nvidia's Blackwell
00:47:48
chip comes out next year you're gonna
00:47:49
have new chips from AMD new A6 from
00:47:53
broadcom and I I think at that point it
00:47:56
it is not going to be possible for them
00:47:58
to keep up
00:48:00
anymore so that's actually positive
00:48:03
regulation and great in your mind
00:48:07
foreign policy
00:48:10
um it is very aggressive foreign policy
00:48:14
yeah clearly you know that could have
00:48:17
lots of unforeseen consequences yeah
00:48:19
what do you think about the the rare
00:48:20
earth trade restrictions coming to us
00:48:23
and is that going to actually affect
00:48:25
this Supply we have lots of rare earth
00:48:27
here in America like in America we have
00:48:29
everything in America like we you know
00:48:31
and I think there's a project underway
00:48:32
to restart Rare Earth production there
00:48:35
ever ever to be a conflict all this
00:48:37
stuff would go away it's it's it's a
00:48:39
cost and it's allowing us to do the
00:48:40
refining here so refining some of this
00:48:42
stuff like we desperately need gallium
00:48:43
gallium nitride for things I'm doing
00:48:45
right with with like shooting the
00:48:46
microwave radiation the problem is the
00:48:48
refining is really messy if you let it
00:48:51
happen in the US it will still be messy
00:48:52
but it will be cleaner but we're not
00:48:54
letting it happen so there's things like
00:48:55
this ulation and obviously the cost here
00:48:58
is different we have a different cost
00:48:59
structure Joe you got to go meet 20
00:49:01
Senators so thank you we
00:49:04
appreciate good times jcal do you want
00:49:07
do you want to talk about AI with Gavin
00:49:09
and I think that would be like a a great
00:49:11
next place to go would be to talk about
00:49:14
the supercomputer being built by friend
00:49:16
of the Pod Elon he's now got the world's
00:49:20
largest super computer and he's going to
00:49:22
10x it according to
00:49:24
reports yeah and I would just say this
00:49:27
is I think a very important moment
00:49:30
for
00:49:32
AI you know for this entire AI trade in
00:49:35
public and private
00:49:37
markets you know everybody I'm sure who
00:49:40
watches your podcast is very aware of
00:49:41
scaling
00:49:42
loss and we have not had scaling loss
00:49:45
for training or if you 10x the amount of
00:49:47
compute used to train a model you
00:49:48
significantly improve the intelligence
00:49:50
and capability of that model and often
00:49:53
there are these um you know kind of
00:49:55
emergent properties that emerge
00:49:57
alongside that that higher IQ no one
00:50:01
thought it was possible to make more
00:50:04
than
00:50:05
25,000 maybe 30,000 in 32,000 pick a
00:50:08
number Nvidia Hoppers
00:50:10
coherent and what coherent means is in a
00:50:14
training cluster that each
00:50:16
GPU to kind of simplify it knows what
00:50:19
every other GPU is thinking so every GPU
00:50:23
in that 30,000 cluster knows what the
00:50:25
other 29,9 199 are thinking and you need
00:50:28
a lot of networking to make that happen
00:50:31
enabled by infiniband right uh
00:50:33
infiniband and and I think even more
00:50:35
importantly in vlink although a lot of
00:50:38
ethernet is um you know never bet
00:50:41
against the internet NE never bet
00:50:43
against ethernet um like if you read the
00:50:45
Llama 3.1 technical paper you know got a
00:50:48
lot of people excited about skinny link
00:50:49
ethernet but um just to slow down for
00:50:52
the audience here Gavin maybe explain
00:50:55
why trans sporting information between
00:50:58
the gpus is important and we're talking
00:51:00
we we're in the weeds here a little bit
00:51:02
everybody's heard of ethernet but some
00:51:03
of the other protocols and ways of
00:51:05
moving stuff around large amounts of
00:51:08
data that's what these h20s h100s do
00:51:11
particularly well they'll move a couple
00:51:12
of terabytes a second from One processor
00:51:15
to the next processor yeah so um you
00:51:20
know
00:51:21
picture a
00:51:23
server in the case of
00:51:25
GPU it looks like maybe three pizza
00:51:28
boxes stacked on top of each other and
00:51:30
it has eight gpus
00:51:33
together and those H gpus are connected
00:51:36
today with something called NV link
00:51:39
probably you can think of the speed of
00:51:41
communication on chip is the fastest tip
00:51:46
to memory next fastest you know um chip
00:51:49
to chip within a server next
00:51:52
fastest and so you take those units of
00:51:55
servers
00:51:57
which are connected the gpus are
00:51:58
connected on the server with a
00:51:59
technology called NV switch and you
00:52:01
stitch them together with either
00:52:03
infiniband
00:52:05
or ethernet into a giant cluster and
00:52:10
each GPU has to be connected to every
00:52:13
other GPU and know what they're thinking
00:52:15
they need to be coherent they need to
00:52:17
kind of share memory for the compute to
00:52:20
work the gpus need to work together for
00:52:25
AI
00:52:27
and no one thought it was possible to
00:52:30
connect more than 30,000 of these with
00:52:32
today's
00:52:34
technology from public reports Elon has
00:52:37
he so often
00:52:39
does Focus deeply on this thought about
00:52:42
it from first
00:52:44
principles how long it should take the
00:52:46
way it should be done and he came up
00:52:48
with a very very different
00:52:50
way of Designing a Data Center and he
00:52:53
was able to make over 100,000 gpus
00:52:58
coherent no one thought it was
00:53:02
possible if uh I was I was a last minut
00:53:05
ad for this but I would have said there
00:53:07
were all these articles that were being
00:53:09
published in the summer saying that no
00:53:12
one believed he was going to be able to
00:53:14
do it it was hype it was you know
00:53:17
Ridiculousness and that was coming the
00:53:20
reason the reporters felt comfortable
00:53:22
writing those silly stories is because
00:53:25
engineers at meta and Google and other
00:53:27
firms were saying we can't do it there's
00:53:30
no way he can do
00:53:32
it he did it and I think the world
00:53:35
really only believed it when you know
00:53:38
Jensen did that podcast I think with uh
00:53:41
was it with gersner am I with gers yeah
00:53:44
it was with kerser and said um what Elon
00:53:47
did was superhuman no one else could
00:53:48
have done it and I actually think you
00:53:50
can argue that Elon doing that in a lot
00:53:54
of ways kind of Saved Nidia from a tough
00:53:57
six-month period when Blackwell was
00:53:59
delayed because everyone who was waiting
00:54:01
for Blackwell and thought it was
00:54:02
impossible to make a 100,000 Hoppers
00:54:04
coherent rushed out and bought a lot of
00:54:07
Hoppers to try and do it themselves now
00:54:09
we will see if someone else is able to
00:54:11
do it it was really really
00:54:14
hard no one else thought it was possible
00:54:19
and as a result of that gr 3 is in
00:54:22
trading now on this giant Colossus
00:54:25
supercomputer the biggest in the world
00:54:27
100,000 gpus Memphis the old Electrolux
00:54:31
Factory and they're they're putting a
00:54:33
lot of energy in there a lot of natural
00:54:34
gas a lot of yeah a dingy Electrolux
00:54:38
Factory yeah with a lot of Mega packs
00:54:41
around it and the City of Memphis is all
00:54:43
in unsupportive this yeah which is
00:54:46
obviously smart for them but you have
00:54:48
not had a real test of scaling laws for
00:54:50
training arguably since gp4 and this
00:54:53
will be the first test and if scaling
00:54:55
lws for training hold grock 3 should be
00:54:59
a significant advance in the
00:55:00
state-ofthe-art
00:55:02
that is an immensely you know from like
00:55:06
a basian way to look at the world that
00:55:08
is like an immensely important data
00:55:11
point but it if that card doesn't work
00:55:15
and I think it is going to work I think
00:55:17
rock rock 3 is going to be really good I
00:55:20
should note that I am with consumers
00:55:22
yeah you're you're involved my firm is
00:55:23
an investor in X got it yeah they
00:55:26
they've raised a tremendous amount of
00:55:27
capital A lot of it from the Middle East
00:55:29
and they're supposedly going to build
00:55:30
Colossus to a million gpus is the state
00:55:34
of goal 10 times bigger than it is
00:55:36
currently there's been some debate back
00:55:38
and forth freeberg about hey are we
00:55:41
hitting a wall here maybe you could
00:55:42
explain the wall either of you to the
00:55:45
audience yeah David well I'll I'll let
00:55:47
Gavin speak to the wall I I mean Gavin I
00:55:49
think one of the questions also is you
00:55:52
know do we
00:55:53
see an evolution
00:55:56
if if the kind of increment in
00:55:58
performance relative to the investment
00:56:02
in net kind of training compute
00:56:05
resources
00:56:07
declines do we start to see a shift in
00:56:10
um how the architecture of the systems
00:56:12
are run meaning like do we start to
00:56:14
build models of models and that starts
00:56:16
to resolve a higher level architecture
00:56:19
that unlocks new performative
00:56:22
capabilities so I would just say we're
00:56:24
already building models of models you
00:56:25
know almost every application startup
00:56:28
I'm aware of is chaining models you know
00:56:29
you start with a cheap model you check
00:56:31
the cheap models work with a more
00:56:33
expensive model you know lots of very
00:56:35
clever things are being done you know
00:56:36
every every AI application company has
00:56:39
what's called a router so they can you
00:56:41
know swap out the underlying model if
00:56:42
another one is better for the task at
00:56:44
hand
00:56:45
[Music]
00:56:46
um as far as what the wall is there's
00:56:49
been a big debate that we were hitting a
00:56:51
wall on these scaling laws and the
00:56:54
scaling laws were breaking down and I
00:56:57
just thought that was deeply silly
00:56:59
because no one had built a cluster
00:57:01
bigger than you know 32,000
00:57:04
h100s and nobody knew it was it was a
00:57:08
ridiculous debate and there were you
00:57:11
know really smart people on both sides
00:57:13
but there is there there's no new data
00:57:15
grock 3 is the First new data point to
00:57:19
support whether or not scaling laws are
00:57:20
breaking or holding because no one else
00:57:23
thought you could make a 100,000 Hoppers
00:57:25
coherent
00:57:26
and I think based on public reports
00:57:28
they're going to 200,000 um Hoppers and
00:57:31
then the next chck is a million it was
00:57:33
reported they're going to be first in
00:57:35
line for Blackwell but grock 3 is a big
00:57:37
card and will
00:57:39
resolve this question of whether or not
00:57:42
we're hitting a wall the other question
00:57:44
you raise David is very interesting and
00:57:46
by the way we should note there is now a
00:57:48
new axis of scaling some people call it
00:57:52
test time compute some people call it
00:57:53
inference scaling and basically the way
00:57:55
this works you just think of these
00:57:56
models as human the more you speak to
00:57:58
one of these models the way You' speak
00:58:00
to your like 17-year-old going off to
00:58:02
take the SAT the better it will do for
00:58:03
you as a human you know if I ask you
00:58:06
David what's two plus two four flashes
00:58:08
in your mind right away if I ask you to
00:58:10
you know unify a grand unified theory of
00:58:13
physics that accounts for both quantum
00:58:14
mechanics and relativistic physics you
00:58:16
will think for a lot longer we have been
00:58:20
yeah nobody knows we have been giving
00:58:23
these models the same amount of time to
00:58:25
think no matter how complicated the
00:58:27
question was what we now learned is if
00:58:29
you let them think for longer about more
00:58:31
complex questions test time compute you
00:58:34
can dramatically improve their IQ so
00:58:36
we're just at the beginning of this new
00:58:38
scaling law but I think the question you
00:58:41
rais on Roi is very
00:58:44
good and I'm happy to address and
00:58:47
there's a context
00:58:48
window oh shift under way as well which
00:58:51
also creates a new kind of scaling
00:58:53
access arguably in terms of potential
00:58:56
set of applications so networks of
00:58:59
models think time context window there
00:59:02
are multiple Dimensions upon which these
00:59:05
these tools ultimately kind of resolve
00:59:08
to better performance oh yeah we have
00:59:10
even if scaling laws for training break
00:59:13
we have another decade of innovation
00:59:14
ahead of us exactly and and and as my
00:59:17
understanding from speaking to folks I'm
00:59:19
certainly not as deep in and well versed
00:59:20
as you but there's a lot of effort and
00:59:22
research going on in re-engineering
00:59:25
various parts of the uh the stack to
00:59:28
reduce energy to reduce every resource
00:59:31
that effectively drives model
00:59:32
performance to basically re-engineer
00:59:34
architecture it was all like very Brute
00:59:36
Force for a period of time and it was
00:59:38
like push push push but now as we go
00:59:39
back and we start to re-engineer and
00:59:41
architect things in perhaps a more
00:59:42
designed way we get better performance
00:59:44
and there's a lot of work to do there
00:59:46
still absolutely this is one of the
00:59:48
great things about capitalism and a
00:59:50
functioning Capital Market is you've got
00:59:52
people working just on the context
00:59:54
window for people who don't know what
00:59:55
that is is
00:59:56
that's the number of tokens a token is
00:59:58
essentially a word you can think of it a
01:00:00
piece of
01:00:01
information the number of tokens you can
01:00:04
put into a conversation with a large
01:00:06
language model some people have really
01:00:08
large context Windows some people have
01:00:10
smaller ones but you can basically put
01:00:11
an entire book in the context window and
01:00:13
start asking questions against the model
01:00:15
and the speed of those is critically
01:00:17
important as well because if you put the
01:00:18
book in there and it takes you 10
01:00:20
minutes to get an answer that's not
01:00:21
functional right Gavin are you an
01:00:23
investor in open AI oh I absolutely not
01:00:26
yeah can you kind of theorize on what
01:00:29
the buildout that's being done with
01:00:31
Colossus does to the advantage that open
01:00:33
AI has today how long till we kind of
01:00:36
catch up there with xai and you know how
01:00:39
much is going to be disrupted and how
01:00:41
quickly here well if scaling l
01:00:46
Hold the best information I have is the
01:00:49
largest cluster Microsoft has after
01:00:52
panicking is still smaller than XI
01:00:55
cluster in
01:00:57
if you didn't believe it it was possible
01:00:58
you you weren't even working on
01:01:01
it grock 3 should takes the lead if
01:01:05
scaling laws
01:01:06
hold in January or
01:01:10
February I do think a lot of talent has
01:01:12
left open
01:01:14
AI I thought it
01:01:17
was a really shocking statement from
01:01:20
Mira moradi that she um resigned during
01:01:24
a fund raise that's the only way she can
01:01:27
express disapproval of what is going on
01:01:28
there and still probably get her money
01:01:32
right so I
01:01:35
think I think there's there's a lot of
01:01:37
reasons if scaling lost hold to be to to
01:01:40
be optimistic about Croc
01:01:43
3 but I think um and and then by the way
01:01:46
on the power question and and they are
01:01:48
you know 20 23 and 24 it was just a
01:01:51
panic to get gpus and get them plugged
01:01:53
in now we're you know trying to make
01:01:56
them efficient and thoughtful into your
01:01:57
point re architecting them and the h20s
01:02:00
now are 50% less power and either 50%
01:02:03
more twice as much compute depending on
01:02:06
the task they they have a little more
01:02:08
compute and a lot more memory which
01:02:09
really matters um so per kind of unitive
01:02:12
effective compute they're a lot more
01:02:14
power
01:02:15
efficient they two or three times you
01:02:17
think or no the h20s probably not
01:02:21
probably not 2x but a good increment and
01:02:23
the h100 was a great chip and then
01:02:26
yeah yeah black well is just around the
01:02:27
corner and that's an entirely new
01:02:29
architecture with an entirely new set of
01:02:32
networking technology what would
01:02:33
consumers if we had to sort of speculate
01:02:35
here what would consumers how would
01:02:37
consumers experiences change in using
01:02:41
forward-facing language models and then
01:02:43
maybe what are developers going to see
01:02:44
on the back end you know in terms of
01:02:46
what they're going to be able to build
01:02:48
if if this hands out you know in the
01:02:49
next you know right now two years yeah
01:02:52
right now you have like a friend in your
01:02:54
pocket who has an IQ of
01:02:58
115 110 maybe but has all of the world's
01:03:02
knowledge yeah accessible to it and
01:03:04
that's what makes it amazing I think
01:03:06
this will be like you have a friend in
01:03:08
your pocket but and they sometimes make
01:03:10
things up again they're very human and a
01:03:13
lot of humans when they don't know the
01:03:14
answer they dis symble yeah these AIS do
01:03:17
it too so you will have a friend in your
01:03:21
pocket with an IQ of what maybe 130 that
01:03:24
knows everything has more upto-date
01:03:26
knowledge of the world
01:03:30
and is more grounded in factual
01:03:34
accuracy and it is interesting for any
01:03:36
question involving realtime information
01:03:39
mostly Sports and finance you know I
01:03:41
always you know if there's a stock down
01:03:43
25% ask every AI why is the stock down
01:03:46
25% generally grock is the one that
01:03:48
knows um
01:03:50
but yeah no Gro grock because of the
01:03:53
Twitter data set exactly knows what is
01:03:57
happening at the moment in the world
01:03:59
today all right and then you know as we
01:04:01
sort of wrap up here on the AI
01:04:04
issue what about the ROI here that David
01:04:07
was mentioning yeah so the I find these
01:04:10
these debates also very funny you know
01:04:12
there have been articles written about
01:04:14
multi hundred billion dollar Roi
01:04:16
questions those are very strange to me
01:04:20
because yeah the biggest Spenders on
01:04:23
gpus are public companies and they
01:04:25
report Financial results every quarter
01:04:28
and you can calculate a metric called
01:04:30
return on invested Capital yes and roic
01:04:34
Roi has gone vertical since they ramped
01:04:37
their capex on gpus it actually just
01:04:40
started to level out in this latest
01:04:42
quarter so the ROI on AI has been very
01:04:44
positive thus far just a fact it's a
01:04:48
really good question will it continue
01:04:51
particularly if you know it's going to
01:04:52
cost a hundred billion dollars to train
01:04:54
a model in two or three years um which I
01:04:56
think is a realistic estimate um however
01:05:00
gu the counter to that isn't the counter
01:05:02
to that that maybe there's a little bit
01:05:04
of hype that you know maybe there people
01:05:07
are trying to determine the ROI and
01:05:09
correlate it more
01:05:11
precisely and I guess that's the
01:05:13
challenge you know meta doing AI across
01:05:16
its entire Enterprise you might see and
01:05:17
Google you might see
01:05:19
it directly making ads more effective as
01:05:22
an example 100% yeah go
01:05:26
Oni yeah but then for other folks like
01:05:29
you know is it actually happening or is
01:05:31
it a toy I guess is the criticism I hear
01:05:34
I'm not saying that's my position but
01:05:35
that's the criticism I hear is like are
01:05:36
people actually getting money from the
01:05:38
co-pilot or maybe this is just product
01:05:40
Market fit Discovery process because the
01:05:43
the AI
01:05:44
laptops AI Intelligence on Apple and
01:05:48
let's say some general llms people feel
01:05:51
maybe aren't worth the money or co-pilot
01:05:53
for Microsoft maybe not worth the money
01:05:55
yeah I mean I personally have not had
01:05:57
good experiences with co-pilot but I
01:05:59
would
01:06:01
say that and I'm sure both of you have
01:06:05
come across these there are lots of
01:06:07
companies that are just these thin
01:06:09
rappers over a foundation model and they
01:06:13
go from zero to 40 million
01:06:15
instantaneously yeah and they're
01:06:17
profitable and for their customers
01:06:20
they're replacing labor budgets I think
01:06:23
I'm sure you guys are noticing this too
01:06:25
but startups today at a given size are
01:06:28
employing fewer people than they would
01:06:29
have three years ago and just like you
01:06:31
know it's funny people were very SK I
01:06:33
would say 50% less yeah and that's the
01:06:35
ROI on AI and like in you know I went to
01:06:38
the first AWS reinvent conference and no
01:06:40
big companies were using cloud computing
01:06:42
it was all startups startups always
01:06:44
adopt Technologies first so outside of
01:06:47
the ROI on AI that you're seeing in
01:06:49
Google and meta from you know using this
01:06:52
across their businesses you seeing real
01:06:54
Roi on I from startups the same way they
01:06:57
saw real Roi from cloud computing before
01:06:59
anyone else it's crazy but I don't think
01:07:03
these companies are in a classic
01:07:04
prisoners dilemma they all believe to
01:07:07
varying degrees that whoever gets there
01:07:09
first to artificial super intelligence
01:07:12
is going to create tens or hundreds of
01:07:13
trillions of dollars of value and I
01:07:15
think they may be right and I if they
01:07:18
get there and they think that if they
01:07:19
lose the race their company is at Mortal
01:07:23
risk so as long as one person is
01:07:26
spending I think they will all spend
01:07:29
even if the ROI decelerates it is a
01:07:31
classic prisoners dilemma competition is
01:07:33
amazing it really is when you have a
01:07:35
free market with competition we're
01:07:37
sitting here two years ago three years
01:07:38
ago on this pod Gavin just lenting like
01:07:42
oh my God what could happen and China
01:07:43
could just roll over us they've got
01:07:45
everything dialed in we're a disaster
01:07:46
and now here we are China's a disaster
01:07:49
they've got all kinds of challenges and
01:07:50
the free market is even with weapon
01:07:53
systems we're seeing capitalism applied
01:07:54
in competition applied there all right I
01:07:56
think we should just wrap up on this uh
01:07:58
Brian Thompson Story the CEO of United
01:08:00
Healthcare was shot and killed in
01:08:01
outside a Manhattan hotel on Wednesday
01:08:03
morning United Healthcare is an
01:08:05
insurance subsidiary of United Health
01:08:07
Group and uh they employ over 140,000
01:08:11
people they provide coverage to
01:08:13
millions and this
01:08:16
was I don't know if you've seen them
01:08:18
assuming you guys have seen the video on
01:08:19
social media or go by on
01:08:21
X it there's been a big debate was this
01:08:24
like a really well-trained killer like
01:08:27
an assassin like a Hired Gun was it
01:08:29
something personal maybe and this
01:08:31
person's a hack and they're not really
01:08:32
good at doing a hit like this and most
01:08:35
people are sort of coming somewhere in
01:08:37
between ABC has reported and this is
01:08:40
where this thing is taken like a crazy
01:08:43
turn that the words deny defend and
01:08:46
depose were discovered on the bullet
01:08:48
casings at the
01:08:49
scene and those are the
01:08:52
terms two of the three are in a book
01:08:54
about how health insurers uh reject many
01:08:58
of the claims every year they deny it
01:08:59
they defend it and then they will depose
01:09:01
people to harass them
01:09:03
essentially this is a crazy crazy story
01:09:09
and it's breaking here I don't know if
01:09:10
anybody has any insights on it but I
01:09:12
thought I would bring it up here to just
01:09:15
maybe intelligently speculate about what
01:09:17
we've seen I mean it seems like the most
01:09:19
likely case is someone's loved one died
01:09:22
for they were denied coverage and what
01:09:25
whether this person was hired or they're
01:09:27
a victim or related to a
01:09:29
victim I don't know if it matters I I
01:09:31
think
01:09:34
there's the observation I'll make is a
01:09:37
question which
01:09:39
is should CEOs be personally responsible
01:09:43
for corporate actions generally speaking
01:09:46
so there's a difference between a CEO
01:09:49
committing fraud or being negligent
01:09:55
but if you don't get a good service or
01:09:59
good quality of service or the product
01:10:00
you expect even if it is something you
01:10:03
depend on
01:10:05
for healthare for example let's say you
01:10:08
take a drug and the drug causes a side
01:10:10
effect that causes some permanent damage
01:10:12
should the CEO be individually held
01:10:15
accountable and if that were the case
01:10:17
would anyone want to be a CEO of a
01:10:19
company that sets out provide services
01:10:22
that are critical like this it's a very
01:10:24
like I think challenging question to
01:10:26
think about because certainly you could
01:10:29
feel like you want to hold someone
01:10:30
responsible because a loved one died
01:10:33
because that CEO's job was to make more
01:10:35
money for their shareholders and
01:10:37
therefore deny claims and therefore the
01:10:40
way that they run that business is
01:10:42
wrong I think ultimately we've got to
01:10:45
kind
01:10:46
of have this distinction between
01:10:49
negligence fraud and acting on the
01:10:52
corporate behalf there's a there's a for
01:10:54
a period time all of these um
01:10:56
documentaries and this movement against
01:10:59
the idea of the corporation generally
01:11:01
speaking you guys remember this was like
01:11:03
a decade ago or 15 an capitalism
01:11:06
movements yeah and it's like this the
01:11:08
corporation Shields
01:11:10
individuals and the corporation creates
01:11:13
a shield for individuals to do harm is
01:11:16
kind of the argument that's made and so
01:11:18
there's a lot of people in this camp
01:11:20
that think that these all CEOs of
01:11:22
companies that let people down are evil
01:11:24
should be killed should be put in jail
01:11:25
whatever the awful kind of
01:11:27
contextualization of that is and I do
01:11:30
think like it's really important to
01:11:31
think about well if no one were to be
01:11:33
the CEO because they face that threat
01:11:35
and those companies can't make money as
01:11:37
a business then those Services go away
01:11:39
entirely that's kind of the end state of
01:11:41
where this goes there's going to be
01:11:42
these difficult situations if a CEO does
01:11:45
something negligent fraudulent
01:11:48
wrong there's a court and there's a
01:11:50
system and there should be kind of laws
01:11:52
that protect people I I don't know man
01:11:54
the whole thing is pretty depress
01:11:55
pressing to think that a guy who's the
01:11:57
CEO he's you know from from what I heard
01:11:59
guy got a family yeah I mean he's got a
01:12:01
family he's got a wife you people that
01:12:03
have known him said he was like a nice
01:12:04
guy and a good guy and he runs a tough
01:12:07
business you know insurance is a very
01:12:09
tough business well and to your point
01:12:10
about this seems like it's very
01:12:12
personalized with the casings if that is
01:12:14
in fact true again this is breaking news
01:12:16
so we're speculating here you hopefully
01:12:18
informed this chart has been the one
01:12:20
that's been circulating on social media
01:12:22
now this has now become Gavin like a raw
01:12:24
shock test about how you feel about
01:12:27
Corporate America Healthcare Etc but
01:12:30
United Healthcare at least according to
01:12:32
these charts that are speculating deny
01:12:35
claims at a rate at a multiple two or 3x
01:12:39
other people in the industry and that
01:12:41
they are the most hated not that any of
01:12:43
this obviously results in somebody
01:12:46
deserving to die I mean I can't believe
01:12:49
I'm even saying this but tellor
01:12:52
Len was
01:12:55
went
01:12:56
viral with a series of tweets not tweets
01:12:59
I think she's on whatever the blue thing
01:13:01
is sky blue blue
01:13:04
sky where she uh wrote some quotes here
01:13:08
and people wonder why we want these
01:13:10
Executives dead Loren wrote on Blue Sky
01:13:13
a microblog social media side alongside
01:13:14
an article about how Blue Sky Blue Cross
01:13:17
Blue Shield no longer cover anesthesia
01:13:20
for the following surgeries uh that's
01:13:23
according to the New York post Gavin
01:13:26
your thoughts on this insanity and
01:13:28
tragedy yeah well first of all it's a
01:13:30
tragedy and actually I'm in New York as
01:13:32
we record this and it happened one block
01:13:34
from where I
01:13:38
am and I mean it it is Absolut it's a
01:13:42
human
01:13:43
tragedy Taylor lorens was not the only
01:13:46
person on social media who reacted that
01:13:48
way which I thought was deeply troubling
01:13:50
it was a large group of people who were
01:13:52
writing very morbid comments there was a
01:13:54
lot of anger around this issue that I
01:13:57
was unaware of I was completely unaware
01:14:00
of it well I think we all probably are
01:14:03
very privileged to have great health
01:14:04
care and able to pay our premium so
01:14:07
we're not affected by this at this stage
01:14:09
in our lives I was very affected by this
01:14:10
30 years yeah we're we're all very lucky
01:14:12
on this podcast
01:14:15
yeah and like I can't imagine how I
01:14:18
would feel if you know someone I loved
01:14:20
had been denied Medical Care and and
01:14:23
died and I felt like it
01:14:25
was unnecessary
01:14:30
and you know due to you know some
01:14:34
Corporation trying to make more
01:14:35
money but I just I cannot believe I'm
01:14:39
sure I'd be outraged I just can't
01:14:43
believe I was deeply Disturbed as a
01:14:46
person by the number of people online
01:14:49
celebrating this it's really crazy yeah
01:14:52
I thought it was it was awful
01:14:56
and I guess the last thing I would just
01:14:57
say about that chart is I think it is a
01:14:59
little misleading I think that is
01:15:00
initial denial
01:15:03
H you know maybe the you know in other
01:15:06
words maybe the company that you know
01:15:07
only denies 7% That's a final Deni you
01:15:10
know they yeah and who knows if the
01:15:12
chart's even correct I bring it up only
01:15:15
that it is the trending item and people
01:15:18
are I find in these tragedies they
01:15:20
become shock tests right like yeah
01:15:23
United healthcare's Medical loss ratio
01:15:25
is about
01:15:27
85% so
01:15:28
85 so 85 cents of every dollar they
01:15:31
collect an insurance premium they're
01:15:33
paying out in claims if you guys want to
01:15:36
look at what the most egregious
01:15:37
insurance industry in the world is it's
01:15:39
title insurance and I'll give you the
01:15:41
list of the rest travel insurance is
01:15:43
pretty bad they pay out like nothing you
01:15:45
were in Insurance business for a bit
01:15:46
there yeah yeah like I mean you know
01:15:48
health insurance is the hardest one of
01:15:49
the hardest besides auto insurance
01:15:51
businesses to be in uh you're paying out
01:15:53
constantly and uh there is a very
01:15:56
difficult kind of process of managing
01:15:59
losses because the number of claims that
01:16:00
comes in it's very easy to suddenly pay
01:16:02
everything out and then your premium
01:16:04
goes up and then people can't afford the
01:16:05
health insurance so they you're striking
01:16:07
this balance of making health insurance
01:16:08
affordable against the cost of medical
01:16:11
claims so it's a very kind of difficult
01:16:13
business to be in I think it's very
01:16:15
complicated to walk people through the
01:16:17
how they in their individual
01:16:19
circumstances aren't necessarily
01:16:21
motivated by some corporate you know
01:16:24
malic
01:16:25
it's just the way the thing has toate
01:16:28
unfortunately let me just say I I think
01:16:29
that much like we saw with
01:16:32
Hamas and the the attacks in Israel or
01:16:35
uro there were people celebrating that
01:16:38
behavior and we've seen that several
01:16:40
times since and and and people have
01:16:42
become very vocal about their
01:16:43
celebration of what they view to be uh
01:16:46
the death and harm done to those who
01:16:49
view who they view to be oppressors in
01:16:51
whatever context you want to kind of fit
01:16:53
this to and put that oppressor label on
01:16:56
an individual and that mindset seems to
01:16:58
hold true through any social Financial
01:17:01
economic political context there's an
01:17:04
oppressor group and there's an oppressed
01:17:05
group and if you're in the oppressor
01:17:07
group you deserve harm you deserve death
01:17:09
you deserve jail and this is another
01:17:12
manifestation of that mindset playing
01:17:14
out this is an individual with a family
01:17:17
who ran a business who worked very hard
01:17:19
for many years and wasn't trying to hurt
01:17:20
people and uh for him to kind of be be
01:17:23
have his life taken like and for people
01:17:25
to say that person is an oppressor I
01:17:27
think really speaks to How Deeply uh
01:17:30
people's minds have been contorted by
01:17:32
this concept that there's oppressors and
01:17:34
oppressed it's almost like a mind virus
01:17:36
and I'm not going to call it the W mind
01:17:37
virus but uh because I just think you
01:17:39
immediately shut down and won't hear
01:17:40
that because it sounds cliche but there
01:17:42
is this concept of like everyone is in
01:17:45
one of two groups you're either being
01:17:46
oppressed or you're an oppressor and if
01:17:47
you're an oppressor there is no limit to
01:17:49
what I should do or what should be done
01:17:51
to you I think it's well said it's a
01:17:52
very it's a very Stark it's a very Stark
01:17:54
and sad kind of commentary on right
01:17:56
what's going on right now sorry if
01:17:58
you're oppressed the Cory to that is if
01:17:59
you are considered oppressed you can do
01:18:01
no wrong you can do no wrong and that is
01:18:05
and obviously there are people who are
01:18:07
oppressed on this planet but they can
01:18:10
still do wrong they should still be held
01:18:12
to a moral standard yeah there there and
01:18:15
there is a moral standard and there is a
01:18:16
standard and murder obviously an
01:18:18
assassination like this is not uh
01:18:21
acceptable and it's just tragic I just
01:18:23
feel so terrible for this these kids
01:18:25
let's end on a happy note so we'll see
01:18:27
you on Saturday big party excited hard
01:18:29
turn here but uh looking forward to
01:18:31
seeing everybody on Saturday for the
01:18:33
all-in holiday Zoom is sponsoring the um
01:18:35
all-in Holiday Spectacular so if you
01:18:37
want to go to allin.com you can go sign
01:18:39
up there right Free Break yeah and then
01:18:41
you can sign up for the zoom Gavin we
01:18:44
expect you to sign up yes we have a
01:18:48
cavar budget we have to replenish over
01:18:50
here at s we do appreciate Zoom helping
01:18:52
us out for the event Zoom AI companion
01:18:54
helps set up the uh the live stream so
01:18:56
thanks to zoom for doing that everywhere
01:18:59
I have to say you know what one of the
01:19:00
great exper I I I have a couple of
01:19:02
experien with the AI that are really
01:19:03
Great Notion has a phenomenal AI built
01:19:06
into it and so
01:19:08
does Zoom with the AI summaries I love
01:19:11
getting these AI summaries of calls I
01:19:13
ask people permission to turn it on and
01:19:14
we get a summary in the bullet points
01:19:16
really well done I went to I we did a
01:19:19
hackathon in the office a couple of
01:19:20
weeks ago and we used have you guys
01:19:22
actually built applications with cursor
01:19:24
you guys have not was it fun to build it
01:19:27
it well it was great because we had so
01:19:29
many people that have never built
01:19:31
software applications in the hackathon
01:19:33
and they built tools from scratch
01:19:35
deployed them in production and are now
01:19:38
using them and I think it really
01:19:40
showcases the um kind of the impressive
01:19:42
impact that AI is having on workplace
01:19:45
productivity you don't need to buy saf
01:19:47
tools you don't need to have service
01:19:48
providers do stuff for you individuals
01:19:50
in and by the way it's only getting
01:19:51
better and you can the ability to just
01:19:54
wake up in the morning and say I want
01:19:56
this app and have it built for you it is
01:19:58
and and by the way I'll just say it's
01:19:59
it's like 70 80% there you still have to
01:20:01
debug you still have to have someone
01:20:02
come in and help kind of get things into
01:20:04
production so there's still a little bit
01:20:05
of work just pages right going back
01:20:08
going back to the going back to the
01:20:09
architecture question two months ago and
01:20:11
and they go back to 60% there now it's
01:20:14
70 to 80 and and you fast forward 12
01:20:16
months and now you've got the
01:20:17
architecture where the AI can run its
01:20:19
own QA testing and debugging and the AI
01:20:22
can run its own kind of Sal marketing
01:20:24
and it's own ux of the application and
01:20:27
it can run everything so you're
01:20:29
basically going to say I want this app
01:20:30
to do this it builds it it tests it it
01:20:33
builds the ux it tests the ux it
01:20:35
iterates the ux it does everything
01:20:36
streamlined for you and then you show up
01:20:38
a couple hours later and you're using a
01:20:40
new product that was built on the Fly
01:20:42
for you it is and and we're literally
01:20:44
like climbing this ladder Gavin like
01:20:47
very quickly that it's going to totally
01:20:49
change this entire software industry is
01:20:52
like getting re architected I could
01:20:53
guess freeberg at his company hackathon
01:20:56
what do you think I made I made a vegan
01:20:59
version of
01:21:00
Yelp only profiles all the vegan
01:21:03
restaurants I actually I actually tried
01:21:04
to make a CRM tool so that I wouldn't
01:21:06
have to pay for CRM licenses I tried
01:21:09
nine months ago to make a Benny off
01:21:12
sorry go ahead Gavin I tried nine months
01:21:14
ago to make a uh multiplayer um app for
01:21:18
Skyrim um which I was very excited about
01:21:20
Skyrim is a video game but I do think I
01:21:22
failed um maybe I should give another go
01:21:25
but the um you learned you didn't fail I
01:21:27
learned I learned I didn't fail I
01:21:28
learned yeah I have a growth mindset but
01:21:30
the um we learned we learned the limits
01:21:32
we we all T yes we tested the app but I
01:21:35
think next year the human language will
01:21:37
be the dominant programming language
01:21:40
totally totally awesome it's totally
01:21:43
yeah it's I you really made a great
01:21:46
Insight earlier Gavin with you know
01:21:49
startups is where you see these
01:21:51
Innovations happen and I I always say
01:21:54
internally
01:21:56
re resource constraints really do Drive
01:22:01
Innovation and when you only have a
01:22:03
nickel you got to try to get a dollar
01:22:05
out of it when you got a dollar and you
01:22:07
got a lot of dollars you're like it's
01:22:08
okay if I get a nickel out of a dollar I
01:22:10
got more dollars laying right over here
01:22:12
and Dylan always talked about this as
01:22:14
well like they they asked him like why
01:22:17
did Blood on the tracks like why did you
01:22:19
do that it was incredible like it's this
01:22:21
Rolling Stone interviewer was talking to
01:22:23
Dylan about it he said this was like my
01:22:25
favorite album and this is incredible
01:22:26
whatever and like what was the
01:22:27
inspiration take me behind it and he
01:22:29
said well you know I owed Columbia
01:22:30
Records and album and they had given me
01:22:31
an advance and they were going to sue me
01:22:33
and I had to give the money back and I
01:22:35
had just gone through a divorce and I
01:22:36
needed the money and I I couldn't do it
01:22:38
so I wrote the album this guy was
01:22:41
desperately was crushed that Dylan's one
01:22:44
of his best albums and pieces of Art in
01:22:47
his life was strictly a function of the
01:22:49
pressure
01:22:51
of is the mother of both there it is
01:22:55
and creative invention absolutely all
01:22:57
right everybody four thanks guys Gavin
01:23:00
Baker Joel onale David freedberg and we
01:23:04
miss you saaks uh uh taking an a a
01:23:07
victory day I it's taking a victory day
01:23:09
today and uh chth both out of the office
01:23:12
today uh I am the world's greatest
01:23:14
moderator and we'll see you next time on
01:23:16
the Allin podcast
01:23:20
byebye let your winners
01:23:22
ride Rainman
01:23:27
David and instead we open source it to
01:23:29
the fans and they've just gone crazy
01:23:31
with it love
01:23:34
[Music]
01:23:40
youen besties
01:23:42
are this my dog taking your
01:23:45
[Music]
01:23:47
driveway oh man myit will meet me at we
01:23:51
should all just get a room and just have
01:23:53
one big huge orgy cuz useless it's like
01:23:55
this like sexual tension that they just
01:23:57
need to release
01:23:58
[Music]
01:24:03
Som we need to get mer
01:24:09
[Music]
01:24:13
our going
01:24:16
[Music]