00:00:00
All right, everybody. Welcome back to
00:00:02
the number one podcast in the world, the
00:00:04
all-in podcast. With me again, the core
00:00:07
4, the original quartet, David Sachs,
00:00:10
David Friedberg, Chimath Polyhapatia.
00:00:13
I'm Jason Calakanis, and we have a very
00:00:15
full docket today. All right, topic one,
00:00:18
gentlemen. AI acceleration. It was a big
00:00:21
week for AI. New study published on
00:00:25
Monday, February 9th, in the HBR,
00:00:27
Harvard Business Review, suggesting that
00:00:30
AI tools intensify work but do not
00:00:33
reduce it. Two UC Berkeley researchers
00:00:36
spent eight months embedded at a 200
00:00:38
person tech company. So, this is one
00:00:40
company's experience. What they found,
00:00:43
employees who use AI worked at a faster
00:00:45
pace, took a broader scope of tasks, and
00:00:48
extended work into more hours of the
00:00:51
day. workers reported feeling more
00:00:53
productive, but they also felt a little
00:00:56
more stress and burnout. Saxs, your your
00:00:58
hot take here, your quick take on this
00:01:00
study. Obviously, it's just uh one
00:01:02
company, but it does track, I think,
00:01:04
some of my experiences.
00:01:06
>> All right. Well, a few points here.
00:01:08
Number one, as you may recall on the
00:01:10
prediction show for this year, my most
00:01:13
contrarian belief is that AI would
00:01:15
increase demand for knowledge workers,
00:01:17
not put them out of business. And I
00:01:18
think you see in this UC Berkeley study,
00:01:21
the reason why that might be the case is
00:01:23
because the employees who use these
00:01:25
tools, like you said, they worked
00:01:27
faster. They took on a broader scope of
00:01:29
task. They actually ended up working
00:01:31
more hours in the day. So they did more
00:01:33
work, not less, and even more effort
00:01:35
rather than less. Not because they were
00:01:37
required to, but just because they were
00:01:40
more motivated. And I think they were
00:01:42
more motivated because their work was
00:01:44
getting upleveled, right? They're kind
00:01:46
of able to offload
00:01:48
uh more menial tasks to AI and it made
00:01:52
their work more purposeful and
00:01:54
meaningful. So I think we're kind of
00:01:56
moving from what some people I think
00:01:58
maybe Jensen has called um taskbased
00:02:00
jobs to purpose-based jobs. And I think
00:02:03
a key skill of employees is going to be
00:02:05
the ability to structure work for
00:02:09
themselves and their AI agents. And the
00:02:11
employees who can do that are going to
00:02:13
be far more productive than those who
00:02:15
can't. That kind of brings me to point
00:02:17
number two, which is that I think
00:02:20
there's a tremendous opportunity this
00:02:22
year for employees who are early
00:02:25
adopters of these tools, who are, you
00:02:27
know, so-called AI natives, to
00:02:29
demonstrate their value to their
00:02:31
employers. They're going to be able to
00:02:33
get a lot more done. They're going to
00:02:35
appear to have superpowers. They're
00:02:36
going to be the people in meetings who
00:02:38
can take an assignment that would have
00:02:39
taken days before and get it done in 2
00:02:41
hours. whether it's a presentation or a
00:02:44
spreadsheet, people are going to be
00:02:45
shocked at how quickly they can get
00:02:46
these things done because they're going
00:02:47
to be fasile at working with AI. So, I
00:02:51
think there's a big opportunity there.
00:02:52
And there was an article that went viral
00:02:54
this week by Matt Schumer called
00:02:56
something big is happening where he
00:02:57
talked about this career opportunity
00:02:59
that's going to be available to kind of
00:03:00
AI early adopters. And I think that
00:03:02
brings me to my third point, which is I
00:03:04
think that you're going to see massive
00:03:07
enterprise adoption of AI, not just chat
00:03:10
bots, but agents this year. But I think
00:03:12
it's going to be driven by the bottom
00:03:14
up. It's going to be driven by these
00:03:15
early adopter employees coming in to
00:03:17
their workplaces, bringing in these kind
00:03:19
of consumerized AI tools, start using
00:03:21
them at work as opposed to top- down
00:03:24
initiatives. I think there's a lot of
00:03:26
top- down company transformation
00:03:28
initiatives that are happening in large
00:03:30
enterprises where the CEO has tasked a
00:03:32
team with figuring out how to use AI,
00:03:34
how to transform their business with AI.
00:03:36
Those initiatives are going to take
00:03:38
months. They're going to be studying
00:03:40
what tools they should use. They're
00:03:41
going to be doing RFPs. And I think it's
00:03:44
ultimately going to be very slow. And
00:03:45
while those things are trudging along, I
00:03:48
think there's going to be these early
00:03:49
adopter employees who just make the
00:03:50
transformation of Feta Comple by again
00:03:53
bringing these tools into the workplace
00:03:54
from the bottom up. So, I think in the
00:03:56
same way that you saw consumerized SAS
00:03:58
tools spread from the bottom up in
00:04:00
enterprises, I think you're going to see
00:04:02
consumerized AI tools spread from the
00:04:03
bottom up in enterprises. And I think
00:04:04
it'll ultimately be one of the big
00:04:06
themes this year.
00:04:07
>> Couldn't concur or agree more. Nick,
00:04:10
throw up that tweet I did. I I did a
00:04:11
tweet and it got 2 million views.
00:04:13
Basically, I said, listen, if you got
00:04:15
laid off by Amazon or Microsoft over the
00:04:16
last two years, just learn OpenClaw and
00:04:19
automate your previous job. Show you
00:04:21
know how to use these tools. go back to
00:04:22
your boss and say, "Hey, I want to come
00:04:24
back and automate everything or go to
00:04:26
startups." Every startup I know is
00:04:28
hiring for this position, which is
00:04:30
somebody who knows how to build and
00:04:32
manage agents. There is no job wreck for
00:04:36
this yet or a title. We should come up
00:04:37
with what this person does, but it it
00:04:39
used to be called prompt engineer. It's
00:04:41
no longer just prompt engineering. It's
00:04:43
managing and educating and offloading
00:04:46
work to an agent and then making sure
00:04:49
they're actually doing it. And right now
00:04:51
it feels like the people in my
00:04:53
organization have four of them who are
00:04:55
focused on this out of 20. I would say
00:04:57
that their leverage is between 10 and
00:04:59
20x the other 16. So now I'm going down
00:05:03
the slope of employees from you know
00:05:05
most technical you know to least and
00:05:08
trying to get each one of them to adopt
00:05:10
and create an agent for them. We'll pro
00:05:12
it will probably take six months but
00:05:14
when we do I think our leverage versus a
00:05:16
competing firm is going to be 10x. As an
00:05:19
example, in the podcasting space, Sax,
00:05:21
we now have it going through podcasts
00:05:24
looking for the best moments or you can
00:05:25
just give it a moment and it will clip
00:05:27
the clip for you and put it in the
00:05:29
Google Drive. So imagine we were all in,
00:05:32
I don't know, our little group chat and
00:05:34
you said, "Oh, from the last episode,
00:05:36
can you get me a clip of minute 3 to
00:05:39
minute 6?" And then it's just on your
00:05:41
iPhone. It's just in the group chat.
00:05:43
Boom. Nobody has to go find it. It just
00:05:45
clips it. That's the kind of work it's
00:05:47
doing. And then we have it looking at
00:05:49
our YouTube stats. We have it looking at
00:05:50
our Instagram or Tik Tok stats and then
00:05:53
trying to tell us which clips are going
00:05:55
the most viral, which ones have the most
00:05:56
comments and then giving us strategies
00:05:58
to how to make them go more viral. It's
00:06:00
really weird because it's coming up with
00:06:02
really great suggestions and taking
00:06:03
eliminating all the reporting work that
00:06:06
knowledge workers do. Chimat, you have a
00:06:08
take on this? I know you've deployed the
00:06:10
software factory which is I think um you
00:06:13
know aligned with obviously this
00:06:16
revolution happening in real time. Last
00:06:19
couple of weeks have been pretty big
00:06:20
with claude opus 4.6 coming out uh chat
00:06:23
GPT codeex coming out lot of advances
00:06:26
and obviously the open claw revolution
00:06:28
that I've now done seven podcasts on in
00:06:30
a row. What are your thoughts Jamal?
00:06:33
>> I think there are two open questions
00:06:34
that I find really interesting right
00:06:37
now. The first question
00:06:40
is, I tweeted it this morning, but is on
00:06:43
prem the new cloud,
00:06:45
which is weird to think that that could
00:06:48
even be possible, but we've spent since
00:06:51
2008
00:06:53
migrating everything to cloud because
00:06:55
there were these economies of scale. And
00:06:58
it created better margin and lower opex
00:07:02
and lower capex because you could
00:07:04
essentially share infrastructure with
00:07:05
other companies. And that's how AWS and
00:07:08
GCP have built such gargantuan
00:07:10
businesses.
00:07:12
The counterpoint to that though is that
00:07:14
in the AI revolution, companies I
00:07:16
suspect will be fighting for their
00:07:17
lives. And I think it's very much
00:07:20
unclear whether it makes sense for a
00:07:22
company to allow the natural leakage of
00:07:27
their edge and their confidential and
00:07:29
proprietary information out into the
00:07:31
wild versus the control that they would
00:07:34
get if they ran on prim. That's a really
00:07:35
important question. What do I mean by
00:07:37
all that? Once you use these tools, it
00:07:40
is very difficult for a company to be
00:07:42
able to control how their data is used
00:07:45
subsequently thereafter. Meaning if I
00:07:48
gave you Jason a PDF of some really
00:07:50
important strategy document or a
00:07:53
PowerPoint deck or a really critical
00:07:55
model and you're interrogating it with
00:07:57
one of these models.
00:07:59
If you're just using chat GPT the
00:08:03
mainline instance of it you're leaking
00:08:05
all of that prompt and response metadata
00:08:07
back to Chat GPT back to Gemini back to
00:08:11
Claude. And there's nothing a company
00:08:13
can do about that. If you're using a set
00:08:16
of agents to act on all that
00:08:17
information, all those agent traces are
00:08:19
going back to these model builders. That
00:08:22
may or may not be a problem for some,
00:08:24
but I suspect it is a deep problem for
00:08:26
others and they just haven't uncovered
00:08:28
it yet. When they realize that that is a
00:08:31
problem, the enterprise will have to
00:08:34
decide, do I just give up and keep
00:08:37
running all of this stuff in the cloud
00:08:39
in a shared experience or do I bear the
00:08:41
incremental cost of running this stuff
00:08:44
in a more coordinated manner that I
00:08:47
control on prem and that would be a
00:08:49
crazy shift just to completely go back
00:08:51
to where we were 20 and 30 years ago.
00:08:54
That's a non so obvious thing that may
00:08:56
happen. So that's number one. And then
00:08:58
number two, I also tweeted this. There
00:09:01
was this really interesting ruling
00:09:03
around what happens inside these cloud
00:09:06
environments, which was a judge saying
00:09:09
there is no attorney client privilege
00:09:11
and confirming that once you start to
00:09:13
use those tools, all of that stuff is
00:09:16
complete public domain material. If you
00:09:18
put these two things together, it
00:09:20
creates a very interesting set of
00:09:22
questions for enterprises. You will need
00:09:24
AI to survive. But if you use the tools
00:09:27
as they exist today at a public
00:09:29
endpoint, you will give up all control,
00:09:33
all security, all confidentiality that
00:09:36
you today have and the ability to follow
00:09:38
through and control what your employees
00:09:39
do with it. The only solution is to have
00:09:42
the pendulum swing all the way back and
00:09:44
have private provision networks, which
00:09:46
increases cost, but then if you save a
00:09:48
bunch of money because of AI, maybe it
00:09:51
all balances out. That is the big
00:09:52
question that I'm wrestling with right
00:09:54
now.
00:09:54
>> Good insights there. And I have some
00:09:56
thoughts on the um onrem because I'm
00:09:58
actually doing it right now. Freeberg,
00:09:59
your thoughts on this moment in time
00:10:02
when we have people uh saying it's
00:10:05
happening faster and it's become
00:10:07
recursive. Recursive obviously fancy
00:10:09
word for those in the audience who
00:10:11
haven't heard it before. Just these
00:10:12
models and these agents can go out and
00:10:15
improve their own work. So after they do
00:10:17
some work or a job for you, you can have
00:10:18
another agent say, "Hey, here's how to
00:10:20
do it better or go learn these new
00:10:22
skills. Go use this skill last 30 days
00:10:24
to go find the last seven days or 30
00:10:27
days of best practices with this tool
00:10:28
and make yourself better and do that
00:10:30
every night at 1:00 a.m. What are your
00:10:33
thoughts, Freedberg, on the moment in
00:10:35
time we are in right now?" Well, I think
00:10:37
the thinking historically was that it
00:10:40
was going to be about recursive model
00:10:42
development where we were going to
00:10:44
continuously improve the actual model
00:10:48
and we were waiting for a context window
00:10:50
where you could feed the model back to
00:10:53
itself. So, you're effectively
00:10:54
retraining the model continuously and it
00:10:56
may be the case that the output is
00:10:59
what's recursive and that turns out is
00:11:03
having the effect that everyone was
00:11:04
waiting for. So, it's kind of a
00:11:06
surprise. I saw a lot of computer
00:11:08
scientists that have worked in AI for
00:11:10
some time, I think, be a little bit
00:11:12
surprised about this moment that we're
00:11:14
in that we're seeing such incredible
00:11:15
strides in model performance just by
00:11:18
making the output recursive. So, let's
00:11:20
see how far it goes. Are you still
00:11:21
obsessed with OpenClaw, Jac? I am. We
00:11:25
have now seen that every week 5 to 10%
00:11:28
of the work we're doing inside of our
00:11:29
venture firm is being moved over to
00:11:31
OpenClaw. We call them replicants. You
00:11:33
can think of them as personas. So we now
00:11:36
have three or four of these. Uh we give
00:11:38
them a notion account, a Slack account,
00:11:41
and we give them a Google Docs account.
00:11:42
They have their own email. And I think
00:11:45
all of this technology was here all
00:11:48
along. It was really or maybe for the
00:11:50
last 6 months, let's say, really good
00:11:52
models out there. But no company would
00:11:54
give the keys to the kingdom to allow
00:11:57
these agents to actually act on your
00:11:59
behalf. Why? because they don't want to
00:12:01
be responsible if it ships your Bitcoin
00:12:04
keys or your passwords to somebody else.
00:12:07
So, in order to use these, you have to
00:12:09
trust them. And if you trust them and
00:12:11
then you are monitoring them, the
00:12:14
results are unbelievable. We uh have
00:12:17
also to your point Chamath fired up Max
00:12:19
Studios. We have Kimmy on them. We are
00:12:22
moving all of the work onto these and
00:12:24
then they'll use Kimmy for most of their
00:12:27
easy jobs, which is free. Then they will
00:12:30
use claude 4.6 opus to orchestrate
00:12:33
things. We also now that we have four of
00:12:35
them freedberg we've created openclaw
00:12:38
ultron which is one meta replicant that
00:12:44
is managing the other four and it checks
00:12:47
their work. It talks to them all day
00:12:49
long about what they're doing and then
00:12:51
summarizes it. And we're building skills
00:12:53
into each one of these. So one of the
00:12:55
skills is like doing deep research. One
00:12:57
of the skills is being able to go into
00:12:59
our sales database which is in pipe
00:13:01
drive. The gains we're getting I was
00:13:04
able to go through everything my Athena
00:13:06
assistant was doing and I was able to
00:13:08
take about and I know Chimath you have
00:13:09
an Athena assistant too. I was able to
00:13:11
take maybe 30% of the Athena assistants
00:13:14
work and give it to the replicant that
00:13:15
let the Athena assistant work on higher
00:13:17
level stuff. I would say on the average
00:13:19
investment team individual, we now have
00:13:23
probably 20% of their work being done by
00:13:25
agents in real time. And the best part
00:13:28
about it is they don't forget to do
00:13:30
work. They don't make mistakes. So once
00:13:32
you put this in, you don't need to have
00:13:34
checklists. They just do it perfectly
00:13:36
every single time.
00:13:38
>> Crazy.
00:13:39
>> And they work.
00:13:41
>> It's nuts, Jamal. So now I'm building
00:13:43
and I've been talking to Benny awful a
00:13:44
lot cuz he's got Slackbot. Claude's got
00:13:47
co-work, but none of them have the keys
00:13:49
to the kingdom. So, what I'm doing is
00:13:51
I'm upgrading to the enterprise version
00:13:53
of Slack. Shamat, you're I think
00:13:54
probably your number two investment in
00:13:56
your career. What an amazing investment
00:13:58
that was. Um, it's number four.
00:14:00
>> Number four. Okay. Listen, keep
00:14:02
grinding. Top five investment for you.
00:14:05
I'm upgrading to the highest level and
00:14:08
I'm ingesting every single Slack message
00:14:11
and then I'm upgrading and giving the
00:14:13
API key for every single email in our
00:14:15
organization to Ultron. They will know
00:14:17
everything going on in the organization.
00:14:18
It is mind-blowing how fast this is
00:14:20
going. And then finally, just a plug,
00:14:23
I'm investing in 10 startups in OpenClaw
00:14:27
space, 125K each to come to the
00:14:29
accelerator. If you're doing work on
00:14:31
this openclaw.co,
00:14:33
email me what you're doing cuz we want
00:14:35
to invest in in at least a 10 or 20 of
00:14:37
these companies right now. Uh this is
00:14:40
the 100% focus of our firm. It is
00:14:42
insane. When do you guys think
00:14:44
enterprises have a huge freakout
00:14:47
around all of this and say, "Wow, we're
00:14:49
leaking all of our most important
00:14:50
information out into the wild." But
00:14:52
Sachs, to your point, the industrious
00:14:54
person trying to get ahead all of a
00:14:56
sudden is using an open endpoint to like
00:14:59
make a deck better. and somehow all of
00:15:01
that stuff is out in the wild. They find
00:15:03
out people are going to have a freakout
00:15:05
moment here soon.
00:15:07
>> I think there's a big opportunity to
00:15:09
take something like OpenClaw and make it
00:15:11
enterprisegrade and secure and all that
00:15:13
kind of stuff. One of my partners at
00:15:15
Craft actually created a new tool called
00:15:18
Lobster Tank, which is a version of Open
00:15:20
Claw that's got some enterprise security
00:15:22
wrapped around it.
00:15:22
>> This is what I mean. On prem is back.
00:15:24
It's going to happen. It's going to
00:15:26
happen. It's cost savings plus do I want
00:15:29
to give all of the secrets in our
00:15:31
organization, every piece of
00:15:32
intellectual property to Sam Alman who's
00:15:35
got to make a billion dollars a year to
00:15:37
keep up with his spend, right? He's
00:15:38
going to build every application.
00:15:40
>> Let's not make it about Sam. Do I, if
00:15:42
I'm GEIGO,
00:15:44
>> want to have
00:15:46
all of my actuaries
00:15:49
using all of our proprietary, private,
00:15:51
and confidential data on risk pricing in
00:15:54
an open instance of an LLM? The answer
00:15:58
is no. That's obvious.
00:16:00
>> Yeah.
00:16:00
>> So now the question is how do you adapt
00:16:02
to that? How do you actually generate
00:16:04
tokens in that kind of a situation? How
00:16:06
do you reason in that kind of a
00:16:08
situation? That is a very expensive
00:16:10
technical problem. It's not necessarily
00:16:12
complicated, but it is technical. That
00:16:14
will bloat the opex because you're going
00:16:17
back to a place that you had said didn't
00:16:20
make sense anymore. It felt very
00:16:21
antiquated if you ever heard a company
00:16:23
was on prem. But AI may be the reason
00:16:25
where you can't afford to be not on
00:16:27
prem.
00:16:28
>> Yeah. And it's it's going to be on your
00:16:30
desktop, too, because one of the
00:16:32
solutions to this is just giving each
00:16:34
employee a really powerful
00:16:38
desktop that is capable of running a
00:16:41
local large language model, which right
00:16:43
now takes a Mac studio with 512 gig or
00:16:46
or daisy chaining two of them. And
00:16:50
that's what people are doing.
00:16:51
>> Remember these vax terminals? I think
00:16:53
that you could actually see a resurgence
00:16:54
of that idea. So you have a centralized
00:16:57
computer and you have a bunch of dumb
00:16:58
terminals
00:16:59
>> y
00:16:59
>> and you have a CLI and so you can
00:17:01
interact with it that way
00:17:02
>> but again it keeps everything inside the
00:17:05
>> but you could also you could also fire
00:17:07
up your own instance in the cloud and
00:17:09
just run it
00:17:09
>> too expensive at scale like for example
00:17:11
8090 is a top 20 customer bedrock
00:17:14
>> it's too expensive already as it is
00:17:17
because of all this overhead
00:17:19
>> because of their margin
00:17:20
>> because of all the nonsense that's
00:17:22
inside of AWS that you have to pay for
00:17:24
in order to just get access the bare
00:17:25
metal. So then you go to Cororeweave.
00:17:26
Okay, fine. But what does Cororeweave
00:17:28
tell you? They're an excellent business.
00:17:30
A, it's all training. B, you have the
00:17:32
situation where too much of what you
00:17:35
have has to be guaranteed into the
00:17:37
future because for them, it makes no
00:17:38
sense to price it on spot. And if you
00:17:40
buy on spot, you just get these surges.
00:17:43
You can't deal with it. So there is no
00:17:44
solution today that makes any sense.
00:17:47
>> It's absolutely correct, Jam. I'll just
00:17:48
put some numbers behind it briefly. We
00:17:50
with our agents hit $300 a day per
00:17:53
agent.
00:17:55
using the uh Claude API like instantly
00:17:58
and that was like doing maybe 10 or 20%.
00:18:01
That's a h 100,000 a year per agent.
00:18:03
We're getting to a place where we have
00:18:05
to basically now say what is the token
00:18:07
budget that we're willing to give our
00:18:08
best devs and then if you aggregate it
00:18:10
across all people you can clearly see a
00:18:13
trend where you're like well hold on a
00:18:14
second now they need to be at least 2x
00:18:16
as productive as another employee that
00:18:18
is actively happening inside my business
00:18:20
because otherwise I'll run out of money.
00:18:23
Yeah, this is a very interesting trend
00:18:25
that I you you're not going to hear
00:18:27
anybody else talk about, but when do
00:18:29
tokens outpace the salary of the
00:18:32
employee? Because you're about to hit
00:18:33
it. I'm about to hit it. I think
00:18:35
superstar developers are already there.
00:18:37
>> Yeah, I think the rank and file is
00:18:39
probably 10 20% max. More than likely,
00:18:42
they're spending a few thousand. The
00:18:44
average
00:18:46
non-technical employee is probably in
00:18:47
the hundreds to low thousands. But to
00:18:50
your point, the trend is what matters.
00:18:52
Yeah. So unless we have some gigantic
00:18:54
leap forward in generating output tokens
00:18:57
at onetenth the cost of what they are
00:18:58
today, which I suspect we will have. So
00:19:02
bear with everybody for a while because
00:19:04
I think Nvidia and Grock and Google and
00:19:07
AMD, they're all incentivized to
00:19:09
massively ramp up the energy density and
00:19:11
massively push down the token cost.
00:19:12
That's going to happen, but it doesn't
00:19:15
change the trend and it doesn't change
00:19:17
the incentives on confidentiality. Let's
00:19:20
talk about prediction markets,
00:19:21
gentlemen. They hit critical mass this
00:19:23
past weekend at the Super Bowl. More
00:19:26
than a billion bet on Kshi, 700 million
00:19:28
on poly market, almost two billion
00:19:30
dollars in wagering. The media has been
00:19:34
obsessing a bit about market
00:19:36
manipulation, insider trading, and all
00:19:38
these issues that are totally valid to
00:19:41
discuss around prediction markets, which
00:19:43
are something new in the world, at least
00:19:45
at this scale. two specific examples uh
00:19:48
from the halftime show. A day old
00:19:51
anonymous Poly Market account correctly
00:19:53
predicted 17 out of 20 halftime show
00:19:56
bets, including the special appearances
00:19:58
by Lady Gaga, Ricky Martin, but it only
00:20:00
profited 17K, a tiny amount. And then
00:20:03
another account created less than 24
00:20:04
hours before the game correctly bet on
00:20:07
Bad Bunny set list. Wall Street Journal
00:20:09
this morning with an article titled,
00:20:11
"Israeli soldiers accused of using Poly
00:20:13
Market to bet on strikes. Israel
00:20:16
arrested several people, including Army
00:20:18
Reserve, for allegedly using classified
00:20:20
information to place bets on Israeli
00:20:22
military operations." Quote, "The
00:20:25
account in question rad in more than
00:20:27
150,000 in winnings before going dormant
00:20:29
for 6 months. It resumed trading last
00:20:31
month, betting on when Israel would
00:20:32
strike Iran." Poly market data shows the
00:20:36
name of the account Rico Suave 666.
00:20:40
Roave
00:20:41
>> Rico Suave. The name of the account Rico
00:20:43
Suave 666. I think that's also the alias
00:20:45
that you were using in Vegas for a
00:20:47
little while there uh at your hotel.
00:20:49
Rico Suave 666. The platforms are
00:20:52
regulated of course by the CFTC.
00:20:56
But you know, questions here uh about
00:20:59
society getting used to this new
00:21:01
platform. Here's Kali's CEO uh talking
00:21:04
about this on CNBC.
00:21:06
>> Let's say there's a a cameraman happens
00:21:09
to be in the stadium during the
00:21:11
rehearsals. You could argue that would
00:21:13
be like somebody at a at a hotel who
00:21:16
sees a rehearsal of a CEO given a
00:21:19
presentation prior. Those guys would
00:21:22
have normally probably had to sign NDAs
00:21:24
by the company because they would be
00:21:26
worried about these issues. But in the
00:21:28
context of this, they probably wouldn't.
00:21:31
>> It's either one of two cases. Either
00:21:33
this information can be public and
00:21:36
that's okay. Or it's information that
00:21:38
cannot be public beforehand and that's
00:21:40
communicated to the staff, right? The
00:21:42
cameraman or the dancer. The reason why
00:21:44
you don't know what song is going to be
00:21:45
played first that's not public and not
00:21:47
everybody knows beforehand. It's a
00:21:48
little bit of a surprise Super Bowl.
00:21:49
>> Yeah. But but it's not nonmaterial. It's
00:21:52
not im it's not material information
00:21:53
that can't be shared. you're making it
00:21:56
that by putting it on this betting
00:21:57
platform, but they have no obligation to
00:21:59
say we're not going to tell anybody our
00:22:00
opening lineup because there might be
00:22:02
money made on this other place that's
00:22:03
now betting on this. That's not the
00:22:05
responsibility is not on them.
00:22:07
>> Your thoughts just broadly on what I
00:22:10
consider society getting used to these
00:22:11
new platforms and what they represent in
00:22:14
the marketplace of ideas.
00:22:16
>> I think the question is, is it really
00:22:19
insider trading? If you and I were
00:22:23
making a side bet and I knew something
00:22:25
about you and I had some edge or some
00:22:27
advantage and I made a bet with you, is
00:22:30
that fair? Should the government have a
00:22:32
role in regulating that? This kind of
00:22:34
goes back to securities regulation that
00:22:35
everything needs to be registered and
00:22:38
then there's this concept of insider
00:22:40
information. It's a real challenge and a
00:22:41
real question on keeping the open
00:22:45
platform
00:22:46
of opportunity for trading on anything
00:22:49
while also trying to mitigate the risk
00:22:51
of what people call insider information
00:22:53
in these trades. There's a good chart
00:22:55
that I think we talked about in our
00:22:57
group chat that shows the distribution
00:23:01
of accounts. There's a few accounts that
00:23:03
have a huge amount of money and make
00:23:05
almost all the profits and then a lot of
00:23:06
accounts that have very little amount of
00:23:08
money and they get burnt through very
00:23:10
quickly. They actually don't have an
00:23:11
edge. So the the accounts that have a
00:23:13
lot of money, they generally only trade
00:23:15
in things where they have an edge where
00:23:16
they make markets. They actually have an
00:23:18
arbitrage or yeah sharps and they eat up
00:23:21
all of the the capital. So if you're a
00:23:24
marketplace like this, you probably also
00:23:27
want to be thoughtful about the fact
00:23:28
that over time you could burn and churn
00:23:31
through all of your customers, all of
00:23:32
the users on the platform if they're
00:23:34
constantly going to be making uh trades
00:23:36
where they simply don't have an edge and
00:23:39
all the capital, all the liquidity is
00:23:40
coming from the accounts that do have an
00:23:42
edge and effectively trade off of inside
00:23:44
information. So just be that these
00:23:47
things end up eating themselves up. I
00:23:49
don't
00:23:49
>> Shabbath man, we had in trade. I'm sure
00:23:52
you remember that in I don't know if
00:23:53
that was in the early 2000s. This idea
00:23:56
has been out there but it has clicked
00:23:58
right now for some reason. What are your
00:24:00
thoughts broadly speaking on the value
00:24:03
of these platforms to society?
00:24:05
>> Let's define some terms first. So in
00:24:07
betting
00:24:08
>> there are two kinds of people. There are
00:24:10
the sharps who know what's actually
00:24:12
going to happen with a better edge and
00:24:15
then there are the squares which is
00:24:17
everybody else and they are grist for
00:24:20
the mill. And in a traditional market
00:24:23
like a sports betting market
00:24:26
there have been edge cases where you try
00:24:29
to throw a game or throw a fight or
00:24:32
shave points and the sharps are involved
00:24:35
in that. But it's increasingly harder
00:24:38
and harder to do because the sports
00:24:39
leagues
00:24:41
analytically are studying these things
00:24:44
so closely to make sure that that never
00:24:46
happens. But what you get are people
00:24:48
with a smarter sense of what's going to
00:24:50
happen and people with less of a smart
00:24:52
sense of what's going to happen. The
00:24:53
thing with prediction markets is it's
00:24:55
not just that. There will be those
00:24:58
things. But then there are going to be
00:25:00
these fundamental markets that are
00:25:02
purely about inside information.
00:25:06
And the question is what can a
00:25:09
regulatory body or a society do about
00:25:11
that? And I think the answer is not
00:25:13
much. And the reason is is that if you
00:25:16
try to regulate this, it looks like a
00:25:18
securities market. And I think the
00:25:20
problem there is that these things are
00:25:22
too fluid and too dynamic and too
00:25:25
ephemeral for them to be legislated like
00:25:28
a security. And so why are these things
00:25:32
happening? It's because there's too many
00:25:34
of these prediction markets that can be
00:25:36
manipulated this way. Somebody knows
00:25:38
something that somebody else doesn't
00:25:39
know. And there's no way to arbitrate
00:25:40
that. This used to exist in the
00:25:43
securities market, too. And this is
00:25:44
where now I'm going to get a lot of
00:25:45
people really upset with me.
00:25:48
In 2000, we introduced the law called
00:25:51
REGGG FD. And what was the point of REGG
00:25:54
FD? It was basically that if you're a
00:25:57
CFO, you cannot talk to an individual
00:26:01
stock manager and tell him something
00:26:03
that you then don't tell everybody else.
00:26:05
Essentially, inside information
00:26:08
that used to be not illegal. I won't say
00:26:10
that it was legal. I would just say that
00:26:12
used to be not illegal. You call your
00:26:14
buddy, he says, "Hey, how you doing?" He
00:26:16
goes, "Man, Quarter was a blockbuster.
00:26:18
You would go and buy the stock." And
00:26:21
starting in the 2000s, it became
00:26:23
illegal. And there used to be these
00:26:24
networks of information arbitrage that
00:26:26
that took advantage of this. Now, this
00:26:28
is an example of Warren Buffett's
00:26:29
returns pre and postreg. Now, what do
00:26:32
you see?
00:26:34
His returns were double the market
00:26:37
returns
00:26:39
when this kind of information sharing
00:26:41
was legal.
00:26:43
And the minute that it became illegal
00:26:45
and you had to basically act on the same
00:26:48
edge as everybody else,
00:26:50
his returns went to the market return.
00:26:53
He generated zero alpha. In fact, he
00:26:55
probably on the margins lost a little
00:26:57
bit. So this is the single best investor
00:26:59
in the world. This is what happens when
00:27:01
you have information symmetry.
00:27:05
So it's just meant to explain that
00:27:07
markets thrive when there's asymmetry.
00:27:11
billions and billions of dollars will be
00:27:12
made in asymmetry. The prediction
00:27:14
markets today unless they are regulated
00:27:17
out of existence or shut down will look
00:27:20
like the stock market free and there's
00:27:23
nothing we can do except choose not to
00:27:26
bet it because otherwise what you're
00:27:28
going to have are a ton of sharps taking
00:27:31
advantage of a ton of squares and I
00:27:33
think that's the end state.
00:27:34
>> Jim, why is it good or bad for society
00:27:36
that these exist? You have a take on
00:27:37
that? There are a certain percentage of
00:27:39
these prediction markets that are about
00:27:42
the well functioning of society and
00:27:46
the use of inside information gets to
00:27:48
the truth faster
00:27:51
and I think that has value especially if
00:27:54
it uncovers corruption or misdeeds
00:27:57
and so if people make money along the
00:27:59
way and that's the incentive that it
00:28:01
takes for folks to work around what
00:28:04
would otherwise be whistleblower laws or
00:28:06
something else to get to the truth and
00:28:08
get it out there faster. That probably
00:28:11
benefits society.
00:28:13
Now, there's a bunch of other things
00:28:14
where some people will just set up a
00:28:16
market that they know about and that
00:28:18
they can control that other people
00:28:19
aren't unaware. That's not good. But
00:28:23
unfortunately, there's no way to discern
00:28:24
when a prediction market gets created
00:28:26
whether it's A or B. And so, you have to
00:28:29
decide whether it's more important that
00:28:31
you can understand these current events
00:28:33
faster with more accuracy or not. And I
00:28:36
think that's where this decision has to
00:28:37
come to and that's what politicians need
00:28:39
to decide and society needs to decide.
00:28:41
>> All right. We're really excited that
00:28:43
we're doing another event. Yes. A new
00:28:45
event from your friends at Allin. The
00:28:47
besties are hosting a new conference uh
00:28:50
a retreat, a summit in uh wine country
00:28:55
May 31st through June 3rd. It's called
00:28:56
liquidity. This is for capital
00:28:58
allocators and LPs and GPS. Chamatop,
00:29:01
maybe you could talk a little bit about
00:29:03
the vision we have here for the event.
00:29:06
>> There are a handful of conferences that
00:29:09
happen every year where money is made.
00:29:11
I'll give you a couple of examples. All
00:29:13
the top
00:29:16
market traders have been invited to this
00:29:18
thing called Iris every year where you
00:29:22
go in front of a large audience, present
00:29:25
your best long or short idea and you can
00:29:28
be a debt trader, a credit trader, you
00:29:31
can be an equities trader. I've done it
00:29:34
several times. Aman has done it. David
00:29:36
Einhorn has done it. Cliff Robbins has
00:29:38
done it. These are incredible places and
00:29:40
you pay like $10,000 a ticket and if you
00:29:42
take those portfolios, they tend to do
00:29:44
really well. Separately, there are
00:29:45
conferences that investment banks
00:29:47
organize that are off the record, not
00:29:49
publicly accessible where they ask their
00:29:52
biggest traders to come to a room, and
00:29:56
they'll give them each a few minutes to
00:29:58
present their best long and short ideas
00:29:59
of public stocks. Then there are these
00:30:01
equivalent conferences that investment
00:30:04
banks do for private companies where the
00:30:06
best fast growing private companies show
00:30:08
up and the CEOs get on stage and they
00:30:10
give presentations.
00:30:12
All of these things have been closed. I
00:30:15
would like to blow that wide open. So
00:30:17
what will we do? We will convene
00:30:22
the best investors in public markets,
00:30:26
the best hedge fund managers, the best
00:30:28
private market investors, the best
00:30:30
growth investors, the best credit
00:30:31
investors, and the largest cohort of LPs
00:30:35
representing trillions of dollars of
00:30:37
capital, and the CEOs of the fastest
00:30:40
growing and most important companies in
00:30:42
technology.
00:30:44
And what we will do over the course of a
00:30:46
few days is we'll have some
00:30:47
presentations. We'll have best ideas.
00:30:49
We'll build relationships.
00:30:52
There may be some investments that may
00:30:53
happen as a result of that. We're going
00:30:55
to shut down all of Yianville. We're
00:30:58
going to shut down the French Laundry.
00:30:59
We're going to shut down all of it. And
00:31:01
it'll be ours for a two-day playground
00:31:03
where we will build relationships,
00:31:05
allocate capital, and maybe make some
00:31:08
money as a result. So, you need to
00:31:11
apply. We will make some allocations to
00:31:14
some folks that may not otherwise get
00:31:16
in. We'll make some allocations to
00:31:18
emerging managers who may need to raise
00:31:21
capital and scale up but can show us
00:31:23
good returns.
00:31:26
And over time, we'll find a way to
00:31:28
increase a lot of this and make it more
00:31:29
and more publicly accessible. We But we
00:31:32
are going to essentially take all of
00:31:33
these things that I've been a part of
00:31:34
that have been in closed rooms and we're
00:31:37
going to put them together and open it
00:31:38
up.
00:31:39
>> Yeah. Well said. Well said. It's going
00:31:40
to be a wonderful event. Freeberg, uh,
00:31:43
anything you're excited about in terms
00:31:45
of the event?
00:31:47
>> No, I love Yfville. We're going to
00:31:49
Yanfell, so I'm looking forward to that.
00:31:50
It's going to be great.
00:31:51
>> I mean, beautiful location and I think
00:31:53
there's going to be ample time for
00:31:54
meetings, networking.
00:31:56
>> Jal, if you're an investor, you can go
00:31:58
to the website to
00:32:00
>> allin.com/events
00:32:02
and you can submit your application. We
00:32:03
can't have everybody there and this is
00:32:06
not like a general admission type event.
00:32:08
It is specifically for this group of
00:32:10
people, capital allocators. So apply at
00:32:11
the website allin.com/events.
00:32:13
It's going to be wonderful. And Shimoth
00:32:14
is putting his focus on it. I can tell
00:32:16
you because I brought him my first five
00:32:19
ideas and he was like, "No, no, no. Yes,
00:32:21
but better." Yes. Yes. So he is engaged
00:32:24
and he's going to make it super tight
00:32:26
and tight.
00:32:28
>> I'm being judgy.
00:32:29
>> Good. I like it. I like it. You know,
00:32:30
all great events, all great art is has
00:32:34
some perspective behind it and we're
00:32:36
excited to have your sharp perspective
00:32:38
behind this one. Liquidity May 31st to
00:32:40
June 3rd. allin.com/events. Okay, let's
00:32:43
move on to our next topic. The new CBO
00:32:45
report is out. Freeberg, you said we are
00:32:47
in a debt death spiral. The
00:32:50
Congressional Budget Offer released its
00:32:52
long-term budget forecast on Wednesday,
00:32:54
February 11th. Here are the numbers.
00:32:56
2026 deficit is 1.9 trillion. And that's
00:33:00
nearly 6% of GDP, much higher than the
00:33:03
3% GDP target we heard from Scott Besson
00:33:06
on this podcast. Social Security, I
00:33:08
talked about that before. Freeberg Trust
00:33:10
runs out in 2032,
00:33:12
uh, one year earlier than previously
00:33:14
expected. That's obviously going to
00:33:15
trigger all kinds of discussions around
00:33:17
austerity measures that folks will not
00:33:19
like. The debt will now grow from 31
00:33:22
trillion today to 56 trillion in 2036.
00:33:26
So, it is not stopping, folks. We are
00:33:29
looking at an average of 2.5 trillion
00:33:31
per year from 2026 to 2036.
00:33:35
Also, currently we're at 120% debt to
00:33:38
GDP. House Committee on Budget expects
00:33:41
it to be 135%. So slightly up in 2036.
00:33:45
For comparison, Japan is 237, Singapore
00:33:48
176, Venezuela 164, the Greeks 154, UK
00:33:53
94. 20 years ago, our debt to GDP was
00:33:55
but 60%. Here's a direct quote from the
00:33:59
report. The fiscal trajectory is not
00:34:01
sustainable.
00:34:03
Okay, Dr. Doom,
00:34:07
what do you think? Freeberg, this is
00:34:09
your story, your chance to shine.
00:34:10
>> Well, there's no outlook to 3% deficit
00:34:13
to GDP.
00:34:14
>> There he is.
00:34:16
And if you look at the assumptions, one
00:34:19
of the key assumptions is that the
00:34:21
short-term interest rate, which is
00:34:23
largely how a lot of the debt is getting
00:34:25
refinanced, is modeled to be around
00:34:29
3.1%.
00:34:31
But if rates climb closer to 5%, as I
00:34:34
mentioned in the past, just using the
00:34:35
current debt levels, it adds another
00:34:37
$650 billion a year of interest expense,
00:34:40
which takes interest expense almost up
00:34:42
to$2 trillion a year just paying the
00:34:45
interest on the past debt. And because
00:34:47
we're running a deficit, that new
00:34:49
interest expense increases the debt
00:34:52
every year. So the debt goes up and up
00:34:54
and up just by adding interest on past
00:34:56
debt. And so this becomes the death
00:34:59
spiral that we've kind of highlighted
00:35:02
many times. So there's nothing in this
00:35:03
report that I think changes the outlook.
00:35:06
It's pretty scary. Um, I'll say that the
00:35:09
trigger point that I'm getting more and
00:35:10
more concerned about, if the Democrats
00:35:13
win the midterms and you end up with a
00:35:16
Democrat in the White House in 2028,
00:35:19
I think that there's a bigger problem at
00:35:21
foot, which is all of the state and
00:35:25
local obligations. We've talked about
00:35:27
Social Security looks like it's going to
00:35:28
run out of money in a few years here.
00:35:30
And so, they're going to need to print a
00:35:32
lot more money to fund Social Security
00:35:34
obligations. uh it's very unlikely
00:35:37
they're going to make a massive cut to
00:35:38
social security because no one will get
00:35:39
elected if they did that and no one will
00:35:42
get elected if they promise to do that.
00:35:44
Um and there's a similar problem at the
00:35:46
state and local level which is that
00:35:48
there's pension obligations. We've
00:35:49
talked about this extensively.
00:35:51
California has nearly a trillion dollars
00:35:52
of unfunded pension obligations to its
00:35:55
public retirees or public employees that
00:35:58
are going to retire. If you end up with
00:36:00
a Democratc controlled House and a
00:36:03
Democrat president in 2028, you'll very
00:36:06
likely see a federalization of that
00:36:08
obligation, meaning that the federal
00:36:10
government will step in to bail out or
00:36:13
support those state and local
00:36:14
governments because otherwise there's
00:36:16
going to be a real kind of economic
00:36:18
crisis a foot. So when you add that
00:36:20
liability coming to hit this CBO report,
00:36:24
which doesn't include any of that in the
00:36:26
next 5 to 10 years, I think that could
00:36:28
be not just the straw that breaks the
00:36:30
camel's back, but the concrete that
00:36:32
breaks the camel's back. And that's the
00:36:34
thing I'm most worried about. There is a
00:36:36
deep connection between what's going on
00:36:38
with the socialist movements at a city
00:36:40
level and now increasingly at a state
00:36:42
level and what we should expect to
00:36:45
happen with the US dollar and how it
00:36:48
relates to federal spending and federal
00:36:50
deficits and federal debt and these are
00:36:52
going to be dragging each other into a
00:36:55
bad place in the next couple of years
00:36:56
one way or the other. So, you know,
00:36:58
that's kind of what I'm more worried
00:37:00
about at this point. It seems if it's
00:37:02
very hard to cut spending or get
00:37:04
Congress to approve budget cuts that we
00:37:06
need to save ourselves from this debt
00:37:08
death spiral, imagine how much worse
00:37:10
it's going to be in the next couple of
00:37:13
years if we have to bail out or
00:37:14
federalize state and local uh debt and
00:37:17
state and local pension obligations.
00:37:19
It's going to be really nasty. So,
00:37:20
that's the thing I worry about the most.
00:37:21
Yeah.
00:37:22
>> In my Dr. Doom hat.
00:37:23
>> Yeah. And I think that that's one of the
00:37:24
things that no one talks about at the
00:37:26
federal level and everyone ignores it
00:37:28
because they assume it's a state and
00:37:29
local problem
00:37:30
>> as we've talked about and I'll bring it
00:37:31
up again and I'll ask my colleague who
00:37:33
works in the administration to think
00:37:35
about this idea that you know if we can
00:37:37
find a way to declare bankruptcy to
00:37:40
restructure the the fiscal obligations
00:37:42
or the the pension obligations that sit
00:37:44
at the state and local level we may have
00:37:47
a way out but short of that uh that's
00:37:49
going to pile onto this this federal
00:37:51
problem. Sax, your thoughts on the CBO
00:37:53
report and this uh death spiral, debt
00:37:58
death spiral.
00:37:59
>> Well, we all agree about the problem of
00:38:01
federal spending and the deficit and the
00:38:03
debt and we're all concerned about that.
00:38:05
With respect to the CBO study, however,
00:38:09
I'll just note that one of the key
00:38:10
assumptions here is that CBO projects
00:38:14
that real GDP will only grow by 2.2%
00:38:18
this year in 2026. That's a very low
00:38:22
assumption given that we grew by over 4%
00:38:25
in Q3 last year and the preliminary
00:38:27
number for Q4 was over 5%. And I think
00:38:30
all of our predictions for GDP growth
00:38:32
this year when we did our predictions
00:38:34
episode was 5% plus. So 2.2% is a pretty
00:38:38
low number. And then they predict that
00:38:40
it's going to slow to 1.8%
00:38:44
after 2026. So again, these are very
00:38:48
meager anemic growth assumptions. And if
00:38:50
you believe that all of this capex
00:38:53
that's being invested in AI
00:38:55
infrastructure is going to have a payoff
00:38:57
then the growth rates could be a lot
00:39:00
higher. And that ultimately I think is
00:39:01
the way to get out of the debt spiral is
00:39:04
we need strong growth. Without that
00:39:05
we're not going to get out of this
00:39:07
problem. So look I think that if you
00:39:09
believe in growth then the situation is
00:39:11
not quite as dire. You know what would I
00:39:14
do? Well, I mean, if I could wave a
00:39:15
magic wand, the two key charts you want
00:39:18
to look at are federal net outlays as a
00:39:20
percent of GDP. This is from Fred,
00:39:22
right? And then you want to look at
00:39:24
federal receipts, which is tax receipts
00:39:27
as a percent of GDP. And you just don't
00:39:29
want those lines to be more than call it
00:39:31
3% apart. I think that's what Secretary
00:39:33
Besson said is try to reduce federal
00:39:36
deficits to 3% of GDP. Historically,
00:39:41
tax receipts have bounced around 17%.
00:39:45
And the federal net outlays have bounced
00:39:47
around 20%. So, if you get back to that,
00:39:50
we'd be in pretty good shape. And we
00:39:52
were before CO our federal net outlays,
00:39:56
which means spending as a percent of GDP
00:39:58
was around 20%. But then with CO, it
00:40:01
bounced all the way up to 30% in 2020
00:40:04
because of both a function of all the
00:40:05
stimulus, but then also the fact that
00:40:07
the economy shrank because of CO and
00:40:09
we've never quite gotten back to that
00:40:11
magic 20% number. Right now, it's
00:40:14
trending around 23%. So, we're doing a
00:40:16
lot better than we did under CO, but
00:40:18
it's still just a few percent higher. I
00:40:21
mean, if it was up to me, I would just
00:40:22
freeze federal spending until the
00:40:24
economy grew to the point where federal
00:40:28
spending as a percent of GDP is 20%. And
00:40:30
then you could let federal spending
00:40:32
continue to grow as the economy grows.
00:40:36
And we're not even talking about cuts
00:40:37
here. We're not even talking about
00:40:39
shrinking the size of the government.
00:40:41
We're just talking about limiting the
00:40:43
rate of growth until the overall size of
00:40:45
the economy can catch up with it. But
00:40:47
look, as we know, it's very hard to get
00:40:49
Washington to go along with that because
00:40:50
there is just a lot of spending pressure
00:40:53
in Washington. One thing I will say
00:40:55
though, I mean, just to give some credit
00:40:57
to the administration here, is that the
00:41:00
level of federal employment is at the
00:41:03
lowest level since 1966. So during uh
00:41:07
President Trump's second term here,
00:41:09
we've gone from roughly 3 million
00:41:11
federal employees to a little bit under
00:41:13
2.7 million. So, you know, over 300,000
00:41:16
federal employees have been cut. I think
00:41:17
that is a good start. I mean, you
00:41:19
>> 10%
00:41:20
>> is a good start for you,
00:41:20
>> by the way. I think that's I think
00:41:22
that's really important to just pause on
00:41:24
just so people understand this isn't
00:41:25
like some hurtful thing about firing
00:41:28
people. They lose their jobs. But when
00:41:30
people move from the government
00:41:31
workforce into the private workforce,
00:41:33
they become productive. They're making
00:41:35
things that grow the economy. And
00:41:37
theoretically, they should also make
00:41:38
more money. So this is positive from an
00:41:41
economic point of view to move the
00:41:43
workforce from public to private. It
00:41:45
also to my point historically I think
00:41:47
it's very important to avoid the
00:41:49
socialist spiral that if you have too
00:41:51
many people employed by the government
00:41:53
it becomes impossible to not employ
00:41:55
people by the government and that
00:41:56
becomes ultimately deacto socialism.
00:41:58
Shimath your thoughts here obviously
00:42:01
great thing that we're shrinking the
00:42:04
size of the government those people
00:42:05
becoming more productive going into the
00:42:07
private sector. That's a big win. We all
00:42:08
agree 10% great job in the first year.
00:42:11
Hey, maybe 5% the next two or three
00:42:12
years would be even better, but the debt
00:42:14
continues to be a problem. Uh, are you
00:42:17
worried? Do you think there's a solution
00:42:18
here? What would you do if you were
00:42:20
running the show?
00:42:23
I think you have to take a broader
00:42:25
historical context to this.
00:42:28
Does debt to GDP matter?
00:42:32
It depends on many things, but mostly I
00:42:35
would say it doesn't matter.
00:42:38
And it's very easy for people to get
00:42:40
agitated about that. Now, there are
00:42:43
things that matter when you print too
00:42:46
much money, which is the value of the
00:42:48
dollar, the value of exports, the cost
00:42:50
of imports, and how to actually protect
00:42:53
your earnings and your wealth. That's a
00:42:55
different question. This is a historical
00:42:58
look back from about 300 years of debt
00:43:01
to GDP of the largest functioning
00:43:03
economies in the world. Now, what do you
00:43:05
see? What you see is the trend where
00:43:08
you, you know, if you smooth it out for
00:43:10
wars, which by the way has this weird
00:43:12
effect of first escalating the debt to
00:43:15
GDP, but then severely impacting it in a
00:43:18
positive way. The Napoleonic War, the
00:43:20
FrancoRussian War, World War II, these
00:43:23
things all had positive effects on
00:43:26
bringing debt to GDP once the war was
00:43:28
over. But the general trend since 1700
00:43:31
to now is up and to the right. And the
00:43:35
key observation is that it moves in
00:43:37
unison that these things are relative
00:43:41
problems. So if the entire world moves
00:43:45
in unison like this there is an argument
00:43:48
to be made which is that you could end
00:43:51
up at 300
00:43:53
250 200% of debt to GDP. But if
00:43:56
everybody is there nothing really
00:43:58
changes that much. The real question is
00:44:01
if one country is able to decouple
00:44:03
itself and its economic output is so
00:44:07
meaningfully different than everybody
00:44:09
else's. So my first take on this whole
00:44:11
debt to GDP thing is I think you have to
00:44:13
look at it together as a group.
00:44:15
Separately,
00:44:17
is it important to contain the debt?
00:44:19
Absolutely. But for these other reasons,
00:44:22
for earnings, for inflation, for all of
00:44:25
those very practical reasons that impact
00:44:27
your daily lived life. And what do we
00:44:29
know there? We know that President Trump
00:44:32
was elected on a massive mandate
00:44:36
to secure the border on one hand, but to
00:44:38
look at waste, fraud, and abuse on the
00:44:39
other. And on that side, what did he do?
00:44:42
He drafted the most important and
00:44:44
prolific private businessman in the
00:44:46
history of the world to be his tip of
00:44:48
the spear.
00:44:50
And what happened? They identified
00:44:52
hundreds of billions of dollars. But
00:44:53
when it came down to it and Congress had
00:44:56
to act to solidify these cuts, they
00:44:59
haven't done much of anything. Which is
00:45:01
a way of saying that if the most
00:45:02
conservative Congress in the history of
00:45:04
the United States has not done much to
00:45:08
solidify these cuts that were identified
00:45:10
by the White House and Doge, then as
00:45:13
Freeberg said, it'll only get worse if
00:45:15
there's ever a Democratic House and
00:45:17
Democratic control. So what do we have
00:45:20
to do? I think we have to just
00:45:21
acknowledge that if debt to GDP
00:45:24
continually moves in unison, the music
00:45:26
isn't up for a very long time. That's
00:45:29
just an observation. I'm not saying it's
00:45:30
right or wrong. It's just the
00:45:31
observation. But you got to find ways of
00:45:33
hedging and owning real durable assets
00:45:36
because the underlying currency that is
00:45:39
used in these economies even on a
00:45:42
relative basis will fluctuate wildly and
00:45:44
just fall off of a cliff which will mean
00:45:46
that it will erode the value that you
00:45:47
have created for yourself and your
00:45:49
family. That I think is the most
00:45:50
important takeaway from all of this,
00:45:52
which is we probably see things like
00:45:57
gold
00:45:59
do much much better over time because
00:46:02
people will be afraid about the
00:46:04
durability of their dollar denominated
00:46:06
resources. But it will also be true for
00:46:08
all these other denominated resources.
00:46:11
But I think debt to GDP quite honestly
00:46:13
if I had to be a betting man will trend
00:46:15
into the 2 3 4 5 600 on a relative basis
00:46:18
for all countries because I just think
00:46:20
the governments of these countries are
00:46:22
addicted to spending and there is no
00:46:25
reason to stop safe of some other
00:46:28
planetary species invading planet earth.
00:46:31
>> Yeah. A black swan event. Yes.
00:46:34
>> Yeah. There's also a question of what
00:46:38
Fed action will do to the capacity for
00:46:42
excess deficit spending. So if Kevin
00:46:45
Worsh
00:46:47
really does want to tighten the Fed's
00:46:49
balance sheet and the Fed is effectively
00:46:52
the first in line buyer of treasuries,
00:46:54
meaning they are printing money to fund
00:46:57
the government spending and they slow
00:46:59
down or actively slow down and stop
00:47:01
doing that, then there is a real um kind
00:47:04
of question on what action will Congress
00:47:06
and the administration need to take
00:47:08
because what will happen as you know if
00:47:10
the Fed stops buying treasuries,
00:47:13
Treasury yields will go up. And if
00:47:15
Treasury yields go up, that means the
00:47:16
interest on the existing debt will start
00:47:18
to go up. And if that lasts for a period
00:47:20
of time and you start going from 3 to 4
00:47:23
to 4 12 to 5% on the short end of the
00:47:26
the yield curve, then it starts to
00:47:28
become way too expensive to fund this
00:47:31
level of deficit spending because the
00:47:34
interest expense will just start to
00:47:35
climb and eat it all up. So I think like
00:47:37
the Kevin Worsh question is if he really
00:47:40
is going to reduce the balance sheet,
00:47:41
what's that going to do to rates? What's
00:47:42
that going to ultimately force Congress,
00:47:44
force the administration to do with
00:47:46
spending?
00:47:46
>> Jason, what do you think?
00:47:48
>> Uh, you know, we are in a
00:47:51
consumer-driven
00:47:52
economy and the employment rate in this
00:47:55
country is absolutely fantastic. So,
00:47:58
just three quick charts here. You know,
00:48:00
this is
00:48:02
the number of job openings. We still
00:48:04
have, even after we burned off uh in
00:48:07
2022 from 12 million to 7 million jobs,
00:48:10
we still have a ton of jobs available.
00:48:12
Then if you look at our unemployment
00:48:14
rate, it's still at historical lows for
00:48:16
our lifetime. If you were born in 1970,
00:48:18
this is as good as it gets. 4.456
00:48:21
is what it's been. It's ticking up
00:48:23
modestly, but still lowest of our
00:48:25
lifetimes. And then finally, the
00:48:26
employment participation rate, number of
00:48:28
people in our society who are working
00:48:30
and able to work. It peaked at 68% or so
00:48:33
during the Clinton years and this is
00:48:36
still low 62%.
00:48:38
We still have people who could be
00:48:40
participating. So all of these problems
00:48:42
will be solved if more people were to
00:48:44
participate and take those jobs. Why
00:48:46
don't they take those jobs? Sometimes
00:48:47
it's a geographic mismatch. Sometimes it
00:48:51
is a skills mismatch and but very often
00:48:54
it is the jobs are not paying enough. So
00:48:58
if you want to give uh Trump his flowers
00:49:00
by closing the border, you've reduced
00:49:03
the number of people taking the jobs uh
00:49:05
off the books and then you have the
00:49:09
businesses are going to have to raise
00:49:10
their minimum wage. They're going to
00:49:12
have to raise their offering wage, which
00:49:14
then might get this 7% or so that are
00:49:17
sitting on the sidelines to take their
00:49:18
jobs. Crazy prediction. I wouldn't be
00:49:21
surprised if we see Trump, who is
00:49:23
obviously a populist, and I tweeted
00:49:24
about this the other day, got almost a
00:49:26
half million views or 400,000 views.
00:49:29
What if um Trump decides he's going to
00:49:31
raise the minimum wage? Not saying I
00:49:33
endorse this or not, but it's incredibly
00:49:35
low at seven bucks an hour. Obviously,
00:49:37
in different cities and states, it's 15
00:49:38
to 20. But what if Trump said we're
00:49:40
going to add a dollar to it or $2 to it
00:49:42
over each year of the next three? this
00:49:45
would be incredibly popular and it would
00:49:48
get some of those people off the
00:49:49
sidelines and maybe take these jobs. So,
00:49:52
just a crazy prediction there, but I
00:49:54
think it's a possibility and I think
00:49:57
they're going to lose the midterms as it
00:49:58
stands right now. It looks like I think
00:50:00
that's the consensus opinion and they
00:50:03
haven't been able to do something with
00:50:05
this affordability. Well, I think most
00:50:09
Americans would say if you raise the
00:50:10
minimum wage uh that that would increase
00:50:13
affordability. You can make the counter
00:50:14
argument it's going to just be
00:50:15
inflationary, but I think most Americans
00:50:17
are going to believe in that. So, I
00:50:19
wouldn't be surprised if you saw Trump
00:50:20
take action there because he does take
00:50:22
populist actions like this from time to
00:50:24
time.
00:50:24
>> You actually the economic literature on
00:50:27
what raising the minimum wage does.
00:50:30
>> Yes. It can increase inflation and it
00:50:32
can lower the it can raise inflation and
00:50:34
it can lower the profitability of
00:50:36
businesses. Yeah. And move stuff
00:50:37
offshore. No, what it does is it makes
00:50:40
it illegal to hire someone whose labor
00:50:44
is worth less than the minimum wage and
00:50:47
so it is shown to create higher
00:50:49
unemployment in those segments of the
00:50:51
economy. It's like one of those core
00:50:55
findings of economists. So yeah, it's
00:50:57
true that some people will be a
00:50:59
beneficiary of getting a higher minimum
00:51:02
wage, but then there'll be other people
00:51:04
who just lose their jobs and it creates
00:51:06
an incentive for those employers to
00:51:08
shift more labor towards automation. So
00:51:11
if you're already worried about those
00:51:12
people losing their jobs to automation,
00:51:14
that's a downside. So anyway, if the
00:51:17
minimum wage were a panacea and it just
00:51:20
increased everyone's living standards
00:51:22
without having downsides, why wouldn't
00:51:24
you make the minimum wage $100 an hour?
00:51:26
You know, why would you know, everyone
00:51:27
would just keep raising it infinitely?
00:51:29
Obviously, it doesn't work because if
00:51:31
you raise the minimum wage too much,
00:51:33
which is to say more than the value of
00:51:35
someone's labor, then they just get
00:51:36
unemployed. looking at what happened in
00:51:39
the different cities or in Australia or
00:51:42
other countries, they have a much higher
00:51:44
minimum wage and they have much more
00:51:45
happiness. Businesses and prices go up
00:51:48
about 20% 10 to 20%. So in Australia, if
00:51:51
you go to a restaurant or if you go to a
00:51:53
Scandinavian country, things might cost
00:51:55
10% 20% more, but you have a happier
00:51:58
population. Uh and yes, it could lead to
00:52:01
more, you know, automation. We got rid
00:52:03
of cashiers because it became too
00:52:05
expensive in New York to pay 15 to 20
00:52:06
bucks for a cashier. Sure. But we have
00:52:09
really low minimum. We have very low
00:52:12
unemployment now. And the businesses can
00:52:14
clearly afford to pay an extra buck an
00:52:15
hour or two bucks an hour. So there's
00:52:17
the theoretical academic argument which
00:52:20
you are correct on and I understand it
00:52:21
fully well. And then there's the reality
00:52:23
on the field which is Seattle, San
00:52:26
Francisco, New York, Los Angeles,
00:52:28
Australia, other places have a much
00:52:30
higher minimum wage. they have higher
00:52:31
happiness in the population. I don't
00:52:33
actually think it will have any impact
00:52:34
because I think it's artificially low,
00:52:37
but that's just one man's opinion. I
00:52:38
think it would change the game here in
00:52:40
America and I think it would actually do
00:52:42
something to your uh concern Freeberg
00:52:44
about socialism. I think that if people
00:52:46
felt that there was a, you know, kind of
00:52:48
a backs stop against this low low cost
00:52:52
of labor, it might actually make people
00:52:54
pretty stoked, you know, that they could
00:52:56
get a higher paying hourly job and it
00:52:58
might take some of that edge off in the
00:52:59
same way universal healthcare might do
00:53:00
that. But again, just one man's opinion.
00:53:03
>> I got to say on all this economic data,
00:53:05
I I think we're kind of missing the lead
00:53:07
here, which is we are at the beginning
00:53:10
of an economic boom. Again, we saw it in
00:53:13
the GDP growth rates in Q3 and Q4 last
00:53:15
year. Over 4% Q3, over 5% Q4. We just
00:53:19
had a January job report where the
00:53:21
economy added 172,000 new private sector
00:53:24
jobs. This blew away the expectation
00:53:27
which was around 70,000. At the same
00:53:29
time, the government shed 42,000 jobs.
00:53:33
The net of this was to bring the
00:53:35
unemployment rate down to 4.3%. So, I
00:53:38
remember a few months ago, JCL, you were
00:53:40
ringing your hands about the fact that
00:53:41
the unemployment rate had ticked up.
00:53:43
Well, now it's back down and you're
00:53:45
seeing a lot of jobs being created in
00:53:48
construction, especially non-residential
00:53:50
construction. Has to do with the data
00:53:52
centers, the AI boom that's going on.
00:53:54
33,000 new construction jobs in January.
00:53:59
We've seen in President Trump's second
00:54:00
term, you've had 615,000
00:54:03
new private sector jobs been created.
00:54:05
While again like we talked about over
00:54:07
300,000 government jobs have been cut
00:54:10
which increases the productivity of the
00:54:12
economy and it does what Secretary
00:54:14
Besson says which is rep privatize the
00:54:16
economy. So I just think that the
00:54:19
overall economic news is really good.
00:54:21
Again we have this AI boom going on.
00:54:23
There's a new chart showing that the
00:54:27
capex for this year that's expected just
00:54:31
from the four leading hyperscalers is
00:54:34
$600 billion just from four companies.
00:54:37
That's a roughly 2% tailwind to GDP
00:54:40
growth right there. That is just the
00:54:42
capex. That doesn't include all the ROI
00:54:45
that you might get from that
00:54:47
infrastructure on the software side, on
00:54:49
the application side, the productivity
00:54:51
side. So, we have a boom going on and I
00:54:53
feel like everyone's kind of
00:54:54
blackpilling about this. Uh, you know,
00:54:57
they're focusing on this
00:54:58
>> I agree
00:54:58
>> CBO report that has unrealistically low
00:55:02
growth rates.
00:55:03
>> We're going to print 6%.
00:55:05
>> Or they're doom scrolling about Epstein
00:55:06
or what have you.
00:55:08
>> And I just think when we look back on
00:55:09
this period, it could end up being a
00:55:11
little bit like the late 90s. Remember
00:55:12
when we had, you know, we look back on
00:55:14
the late 90s, we're like, "Wow, we had
00:55:15
like phenomenal economic growth.
00:55:17
>> Golden age.
00:55:18
>> Golden age. Economic labor
00:55:19
participation.
00:55:21
the internet,
00:55:23
>> right? But if you remember what politics
00:55:25
were like at that time period, all
00:55:26
anyone talked about was whether Bill
00:55:28
Clinton got a from Lewinsky.
00:55:30
>> So my point is just again, I'm not sure
00:55:33
we're focused on the right things. I
00:55:35
suspect we'll look back on this time
00:55:37
period as the beginning of a new golden
00:55:38
age.
00:55:39
>> I agree.
00:55:40
>> I think you're correct. And just in
00:55:42
terms of the hand ringing comment,
00:55:43
anytime a statistic is 10 15%, I
00:55:46
highlight it. I wouldn't use hand
00:55:48
ringing. I would just say we generally
00:55:50
look at that when we went from 4.1%
00:55:52
which is where Trump inherited it went
00:55:54
up to 4.5 it's about a 10% increase in
00:55:57
one year if that trend were to continue
00:55:59
that would be notable but to your point
00:56:01
it's gone down and that is because the
00:56:04
border I believe the southern border is
00:56:06
closed and as you're pointing out we've
00:56:08
got a lot of good news in the economy so
00:56:10
people are hiring still and uh so we are
00:56:13
in really good economic shape I would
00:56:15
say it's hard to deny that yeah
00:56:18
>> all The job creation has been enjoyed by
00:56:20
native born Americans as well. All the
00:56:22
job loss has been on non-nativeborn
00:56:25
Americans which is pretty remarkable. So
00:56:27
that I think is also going to acrude to
00:56:30
the benefit of of more Americans. By the
00:56:32
way, just on the unemployment thing,
00:56:34
there was a slight tick up in October
00:56:37
because of the October one buyouts.
00:56:39
Remember Doge created the the buyout
00:56:41
program
00:56:42
>> and September or October? When did they
00:56:44
hit?
00:56:44
>> It was October 1st was the deadline for
00:56:46
that. And so we had a tick up in
00:56:48
unemployment related to that. But
00:56:50
remember, all of those were voluntary
00:56:51
buyouts. They all chose the Doge option.
00:56:54
That's what created the tick up in
00:56:56
unemployment, but again, it was all, I
00:56:58
think, a good and voluntary tick up. And
00:57:00
now the unemployment rate has ticked
00:57:02
down. So again, the job creation right
00:57:04
now is strong.
00:57:05
>> And to just put a finer point on it, the
00:57:08
top two areas where illegal aliens are
00:57:10
working in the United States, most
00:57:11
people don't know this, construction,
00:57:13
number one, and the second one is
00:57:14
leisure and hospitality. So you got two
00:57:17
and a half million people working in
00:57:18
those two categories. Which is why I
00:57:19
said if you want to see more Americans
00:57:22
take jobs and you want to see wages go
00:57:24
up. If you went to those businesses and
00:57:26
you find those businesses for hiring
00:57:28
illegal aliens, which is the easiest
00:57:29
thing in the world to do, you just show
00:57:31
up at a construction site, you take
00:57:34
pictures of everybody who is working
00:57:36
illegally, which is what they used to do
00:57:37
in the ICE agency. They would then do um
00:57:40
you know surveillance of construction
00:57:42
sites and then they would go and find
00:57:43
the construction person and then they
00:57:45
had to hire Americans or that
00:57:46
construction company would get in
00:57:48
serious trouble. There's been
00:57:49
multi-million dollar fines done over the
00:57:51
last 20 years specifically on
00:57:53
construction sites and that would drive
00:57:55
more people to raise the wages of
00:57:57
construction workers which would even
00:57:59
lower unemployment more and increase
00:58:01
labor participation. That's where the
00:58:03
big win is. Go to construction sites.
00:58:05
>> You want you want ICE to randomly raid
00:58:09
employers, construction sites.
00:58:10
>> I would use the term raid. Raid is no. I
00:58:12
survey
00:58:13
>> and just check everyone's papers. You
00:58:14
want I showing up everywhere. You want
00:58:18
number one, they're doing this already,
00:58:20
gentlemen. This is well within their
00:58:21
purview. Look up the legal. They have
00:58:23
been doing this for 30 years. This is
00:58:25
actually the technique they used before
00:58:28
raiding cities in a chaotic way. They
00:58:30
went, they surveiled, which is their
00:58:32
right to do. They have the right to do
00:58:35
that. I didn't say raid. I said surveil.
00:58:37
That is a peaceful, quiet thing to do.
00:58:39
And then they find business owners. The
00:58:41
business owners are the people who are
00:58:42
causing this problem. The if there was
00:58:45
not a job available in construction for
00:58:47
20, 30, 40 bucks an hour off the books
00:58:49
and not paying taxes, those immigrants
00:58:52
who are crossing illegally would not be
00:58:55
here. If they couldn't get a $30 an hour
00:58:57
off the book job, working at a hotel or
00:58:59
as a dishwasher, they would not come.
00:59:01
And the blame need to stop hiring them.
00:59:04
What do you mean?
00:59:06
>> Surveil surveile. So what is how do they
00:59:08
how do they with a camera figure out if
00:59:10
someone's illegal? What's the camera
00:59:12
figuring out?
00:59:12
>> Okay. So you guys, it's very simple. And
00:59:14
I
00:59:14
>> Oh, you have to scan first and so then
00:59:17
you'll know who you guys are.
00:59:18
>> No, I want to hear I want to hear I want
00:59:20
to hear you want to hear the answer and
00:59:21
how misinformed the three of you are and
00:59:24
biased. I will tell you you're all
00:59:25
misinformed and biased.
00:59:27
>> Surveil someone to figure out they're
00:59:29
illegal.
00:59:29
>> It's super simple. You go to the
00:59:31
construction site. Everybody checks in
00:59:33
their morning. They have a truck. This
00:59:35
has been done for decades, gentlemen.
00:59:37
They take pictures of everybody. Then
00:59:39
they go in at the end of the day after
00:59:41
surveilling for weeks, Chama. And it's
00:59:44
they have done this already. This is all
00:59:46
facts. They've had multiple cases where
00:59:47
they go to the construction site, they
00:59:49
take pictures, they take a video, then
00:59:50
they go to the business owner and say,
00:59:51
"Show us these people's payubs." And the
00:59:54
business owner goes, "I don't have
00:59:56
payubs for these people." And they say,
00:59:58
"Okay, here's a video of them working
01:00:00
for 8 hours a day. Where's their payub?
01:00:02
Show us their taxes. The businesses are
01:00:05
paying people off the books. That is tax
01:00:07
evasion. And then they got multi-million
01:00:11
dollar fines. Here's a very important
01:00:13
case. This is from back in 2017. The
01:00:16
Justice Department and ICE went after a
01:00:19
group which was hiring illegal aliens.
01:00:23
This is the largest payment ever in an
01:00:25
immigration case. 95 million recovered,
01:00:28
80 million criminal forfeite, 15 million
01:00:30
in civil payments. That represented,
01:00:33
according to our Justice Department in
01:00:35
2017, the largest ever levied
01:00:37
immigration case, we can solve almost
01:00:40
all of the immigration issues with the
01:00:42
exception of maybe criminal gangs. just
01:00:46
by doing basic surveillance, basic, you
01:00:49
know, detective work, asking these
01:00:52
businesses to show the payubs of the
01:00:56
people working for them. And ICE has
01:00:58
been doing this. They've already been
01:01:00
doing this.
01:01:00
>> Your suggestion is to do this for every
01:01:02
company in America.
01:01:04
>> Okay. So again, you're being hyperbolic
01:01:06
and you're not negot
01:01:10
>> I said I said this at the top. You pick
01:01:12
the number one employer of illegal
01:01:14
aliens, 2.5 million people working
01:01:16
construction. You start with the largest
01:01:18
construction sites and then you work
01:01:19
backwards. Then you start with the
01:01:20
largest restaurant and hotel chains.
01:01:22
>> If Stephen Miller were doing this, you'd
01:01:24
say he's not compassionate enough. You
01:01:25
call him a fascist.
01:01:26
>> I have Nope. Incorrect. Once again,
01:01:28
incorrect. I have stated publicly here
01:01:29
on the pod and I have stated publicly on
01:01:31
Twitter that this is actually what
01:01:32
Steven Miller should do because this
01:01:34
would go after the people who are
01:01:37
causing the immigration problem. The
01:01:39
people immigration problem are the
01:01:40
people. Let me finish. Let me finish
01:01:42
that. The people causing this problem
01:01:44
are the business owners. They are
01:01:46
providing the incentive to come here.
01:01:48
Steven Miller should stop doing the
01:01:50
crazy raids and he should go and just
01:01:53
>> You don't think it's the government
01:01:54
benefits that are incentivizing people
01:01:55
to come.
01:01:57
>> I think that's like far down the list.
01:01:58
Two, three, four.
01:01:59
>> Far down the list.
01:02:01
>> Yes.
01:02:02
>> Is the free healthcare and the free food
01:02:03
and the free.
01:02:04
>> Actually have statistics. I can give you
01:02:05
a statistic on. According to this LA
01:02:06
Times survey, 75% of immigrants come
01:02:10
here for better job opportunities.
01:02:13
People coming to America illegally are
01:02:16
coming here for economic reasons. They
01:02:18
are not coming here to commit crimes.
01:02:20
They are not coming here to get
01:02:21
benefits. That is way down the list.
01:02:23
That is a small percentage.
01:02:26
>> How is this going to deport all the gang
01:02:28
bangers, the rapists, the murderers, the
01:02:30
ones who aren't working on a farm?
01:02:32
They're not doing
01:02:33
>> That's a totally separate issue. They
01:02:34
should go do that. That's a separate
01:02:35
issue. They should go do those and go
01:02:37
after
01:02:37
>> that is what ICE was doing. They're
01:02:39
trying to round up the known criminals
01:02:42
for whom they get warrants and then they
01:02:44
capture them and deport them. That's a
01:02:46
separate problem.
01:02:47
>> Yeah, those are two separate problems.
01:02:48
I'm not talking about the problem of
01:02:51
gang bangers. You can do gang bangers.
01:02:52
I'm talking about if you actually want
01:02:54
to move big numbers. The gang bangers
01:02:56
are small number. The people working in
01:02:58
construction working at hotels are a big
01:03:00
number. Yeah, they're both equally
01:03:02
important. Saxs, we're in agreement.
01:03:04
Okay.
01:03:05
>> Yeah. The the thing we're not doing at
01:03:07
scale is going after the businesses that
01:03:11
are creating the incentive for the
01:03:12
majority of people who come here.
01:03:14
Ferrari has a new car coming out. It's
01:03:17
going to be their first all electric
01:03:20
vehicle. Very polarizing.
01:03:22
Here's an illustration of the vehicle
01:03:24
from Car and Driver. This is not the
01:03:27
accurate one because it's going to be
01:03:29
revealed in May, but this is what they
01:03:30
think it's going to look like. 1,000
01:03:32
plus horsepower, four electric motors, 0
01:03:35
to 60 in under 2.5 seconds. Uh 330 mi
01:03:38
range. It's the heaviest Ferrari ever,
01:03:41
5100 lb compared to the iconic F40,
01:03:44
which was but 3,000 lb. It's going to
01:03:47
launch in May of 2026. But we got to see
01:03:50
the interior, and this is what
01:03:52
everybody's buzzing about. It's gone
01:03:53
viral on the interwebs. former Apple
01:03:55
design chief Johnny IV uh on his team
01:03:58
with his partner Mark Nuome who also
01:04:00
designed the iconic Ford 021C concept
01:04:03
car were involved in this.
01:04:05
>> Wait, what is that?
01:04:07
>> Uh it's this is like if you're a car
01:04:09
nerd, this was like this incredibly
01:04:11
innovative moment in design that never
01:04:13
happened that Ford did it. It looks very
01:04:15
similar to an Apple product.
01:04:17
>> Here's the key for the new
01:04:20
>> looks like an animated character in cars
01:04:22
or something like that, you know. You
01:04:24
have this beautiful square glass key
01:04:26
like an iPhone. You put it in and the
01:04:28
yellow Ferrari yellow drains out and
01:04:31
goes into the shifter. That was one
01:04:33
nuance that people thought was very
01:04:34
beautiful. The screen looks very Mac
01:04:37
inspired except unlike Tesla which is uh
01:04:40
no buttons and removing buttons. They're
01:04:42
adding buttons here and making the
01:04:44
buttons very tactile. All the sports car
01:04:46
enthusiasts love tactile memory based
01:04:48
buttons that you can just have fun with
01:04:51
and flip and feel like you're a fighter
01:04:52
pilot. Finally,
01:04:54
the uh turning the car on is like
01:04:57
starting up a jet. You have a launch
01:04:59
button you twist and press and it makes
01:05:01
the whole car turn Ferrari orange or
01:05:03
red. And uh yeah, that's the inside
01:05:07
sachs. You buying one or not? You like
01:05:11
it?
01:05:12
>> I saw I saw everyone just, you know, all
01:05:14
over this design and I thought it was a
01:05:17
little bit unfair in the sense that I
01:05:19
actually overall like the interior. I
01:05:22
thought it found a compromise between,
01:05:24
you know, let's call it the all glass
01:05:27
cockpit of a Tesla versus a totally
01:05:30
analog old Ferrari interior. Like you
01:05:34
said, it it had a combination of
01:05:36
screens, but then also buttons. And they
01:05:38
made a point of showing that the buttons
01:05:41
were not only nicely tactile, but they
01:05:43
also made pleasing sounds and that kind
01:05:44
of stuff. It seemed very heavy duty. So,
01:05:47
I thought the interior actually was
01:05:49
pretty good. Again, nice balance between
01:05:52
kind of the interior of a race car, the
01:05:54
simplicity of that iPad screen, but also
01:05:57
having enough sort of buttons that you
01:05:59
develop muscle memory around where all
01:06:00
the controls are. You don't have to go
01:06:02
hunting for them through a menu.
01:06:04
>> I thought the missier wasn't on the
01:06:06
inside. I thought it was on the outside.
01:06:07
I hate the look of the outside of this
01:06:10
car.
01:06:10
>> It looks to me,
01:06:11
>> by the way, just to be clear, that look
01:06:13
is what people are projecting. It's not
01:06:15
the final version.
01:06:16
>> I think this is terrible. This to me
01:06:17
looks like a Corvette maybe or even like
01:06:20
a Franazam. I mean
01:06:21
>> it looks like a Model 3. Yeah.
01:06:24
>> I don't like the What's that like the
01:06:25
black part of the front or even the
01:06:28
>> grill?
01:06:28
>> The grill. It looks terrible and the
01:06:30
things going on on the sides
01:06:31
>> and then the back almost looks like a
01:06:33
hatchback or something. It's just, you
01:06:35
know, uh a Ferrari should look swoopier.
01:06:38
It should look curvier and there should
01:06:40
be fewer different pieces to it.
01:06:42
>> 100%.
01:06:42
>> So, I don't know. Doesn't look right to
01:06:44
me as a Ferrari, but I thought the
01:06:46
inside actually was was fine.
01:06:48
>> I like it.
01:06:49
>> Yeah.
01:06:50
>> Chimop, you buyer. When's the last time
01:06:51
you actually drove yourself, Sax? Have
01:06:54
you actually used a steering wheel in
01:06:56
the last decade? When's the last time
01:06:58
you actually used the steering wheel?
01:06:59
>> Full full self-driving has made me a
01:07:01
driver again because I just set the full
01:07:03
self-driving.
01:07:04
>> Wow. And it's such a game changer. Yeah.
01:07:06
>> Well, with with FSD.
01:07:08
>> Yeah. Okay. So, you're now driving
01:07:10
around Texas. I like it
01:07:11
>> with FSD.
01:07:13
>> Okay.
01:07:14
>> Because the Uber takes forever. So now
01:07:16
I'm just like, you know,
01:07:16
>> you you like uh you've always liked to
01:07:19
drive Jamoth. I think you you drive
01:07:21
yourself these days or you uh
01:07:22
>> I drive myself in a Model Y with FSD or
01:07:25
I take a Whimo. One of the two.
01:07:27
>> Yeah, Whimo's in the valley now. Yeah.
01:07:29
On the piss.
01:07:29
>> I've had a Ferrari.
01:07:32
What I would tell you is that there's
01:07:34
just something that's very unique.
01:07:35
There's a Ferrari experience that's
01:07:37
different from every other car.
01:07:39
And I think that the new CEO,
01:07:41
Benedtovenia,
01:07:43
is a very talented executive and I think
01:07:46
that he's probably going to land
01:07:47
something beautiful.
01:07:50
The thing is that we are racing against
01:07:52
time. And I've said this before, but FSD
01:07:57
and autonomy is going to shift the
01:07:59
number of people that even know what it
01:08:01
means to drive. It will feel like
01:08:04
when we look at somebody who really
01:08:08
embraces thoroughbed racing, it's just
01:08:11
going to happen in smaller and smaller
01:08:13
places and less and less often. And
01:08:16
that's not because these cars aren't
01:08:18
beautiful, but it's because the risk
01:08:21
will not make any sense for most people
01:08:23
under most conditions. And I think
01:08:26
that's the big thing that's going to
01:08:27
change. Like the car culture in America
01:08:30
was a profound part of the American
01:08:32
culture. You know, driving from A to B
01:08:35
on vacation, the sense of freedom, the
01:08:38
building of the interstate highway
01:08:39
system. These were huge parts of what
01:08:41
made America great and the rails on
01:08:44
which all this productivity sat on top
01:08:46
of. And now I think it's all going to
01:08:49
change. So I don't know. I mean, I think
01:08:50
the car will probably be beautiful. Like
01:08:52
Ferraris are beautiful. There's a
01:08:54
Ferrari dealership in Redwood City. And
01:08:57
whenever I drive by it, I slow down and
01:08:59
I look out.
01:08:59
>> Yum, yum.
01:09:01
>> They make beautiful cars.
01:09:04
>> Piece of art. Yeah.
01:09:05
>> And I think in places like China and
01:09:07
India, they're always going to have a
01:09:09
market. But I think in places like the
01:09:10
United States, it's going to become so
01:09:12
expensive to pay for the insurance if
01:09:13
you are driving yourself
01:09:16
>> that the idea that you would buy any car
01:09:20
is going to feel tougher and tougher
01:09:23
just because I think the math is going
01:09:24
to be tough. But the experience inside
01:09:27
of the Ferrari is second to none. So it
01:09:29
probably is that there's going to be a
01:09:31
bunch of high-end cars like Ferrari
01:09:33
where you pay for the experience, you're
01:09:35
in a position to pay for the car, you'll
01:09:36
pay for the insurance, the luxury,
01:09:39
>> all of it, and then the rest of us will
01:09:42
be using FSD or Whimo
01:09:44
>> 100%.
01:09:46
I we have two Model Y's and we have to
01:09:48
get another car and it's like, well,
01:09:50
what else can we buy? We have no choice.
01:09:52
It's either buy one of the last X's.
01:09:54
>> I'm so mad the X well deprecating the X
01:09:56
really bought I have a real problem now
01:09:58
which is I have five kids so the X is
01:10:01
the only car that can manage seven
01:10:03
people so I need I need a new
01:10:05
>> there is a three row by the way Model Y
01:10:07
but it's a bit tight.
01:10:08
>> It's a bit tight. It's not
01:10:09
>> it's a bit tight. I This is I I wish
01:10:12
Elon would have made the minivan or the
01:10:14
three row SUV and who knows maybe he
01:10:17
does someday. When I was in Abu Dhabi, I
01:10:19
saw my dream car. It is this Lexus
01:10:23
minivan and the doors open and it's like
01:10:27
first class airline seats.
01:10:30
>> Yes.
01:10:31
>> And the front is completely blacked out.
01:10:33
So you have total privacy.
01:10:36
This car. Yes. This car.
01:10:37
>> Yes.
01:10:38
>> I
01:10:39
>> This is not It's not for sale in the US.
01:10:40
>> It's not available in America. Why is
01:10:42
this car or minivan, whatever, not
01:10:44
available in the United States?
01:10:46
>> I think it's the Lexus LM. And then
01:10:47
there's the Alfred in Japan. My favorite
01:10:50
cars.
01:10:51
>> This is the interior.
01:10:53
>> Show the view captain's chairs. It's got
01:10:55
a full screen and divider between the
01:10:58
two with the drivers in the front.
01:11:00
>> Um and then you have this luxurious.
01:11:02
Those aren't the captain's chair. Show
01:11:04
the captain. There's the captain.
01:11:05
>> See those beautiful captain's chairs?
01:11:06
They're gorgeous. Basically like an
01:11:08
executive van.
01:11:09
>> These are like Etihad
01:11:12
first class airline seats. It's
01:11:14
unbelievable. Look at these. Look at
01:11:16
these two seats.
01:11:18
>> Unbelievable. Gorgeous. And you have a
01:11:20
full monitor in front of you, David. You
01:11:22
press a button and the monitor rises and
01:11:23
falls so you can talk to your driver or
01:11:25
have CNBC on.
01:11:26
>> I like getting in and out of SUVs or
01:11:28
minivans. The height is good for me.
01:11:31
>> So easy to get. And the Alfred is the
01:11:33
other one. None of these are available
01:11:35
in the US. These are the number one cars
01:11:36
in China, Singapore, the Middle East for
01:11:39
chauffeer driven cars. They're
01:11:41
incredible.
01:11:42
>> It's called what? An Alfred. What's
01:11:44
that?
01:11:44
>> Alfred is the Toyota version. And then
01:11:46
Lexus is obviously the higher brand of
01:11:48
Toyota and they make I think it's called
01:11:50
the LS is the name for these. Cannot get
01:11:54
them in.
01:11:54
>> All right, boys. I love you very much.
01:11:56
That's another amazing episode of the
01:11:58
Allin podcast. 261 weeks and counting.
01:12:02
Strong.
01:12:03
>> Episode 261.
01:12:05
>> I think it's 261. Yeah.
01:12:06
>> All right. Back at you. Let's go.
01:12:07
>> Love you boys. Love you.
01:12:12
>> Let your winners ride.
01:12:14
Rainman David
01:12:18
>> and it said we open sourced it to the
01:12:20
fans and they've just gone crazy with
01:12:22
it.
01:12:24
>> Queen of
01:12:32
besties are
01:12:34
my dog taking a notice in your driveway.
01:12:39
Oh man, my habitasher will meet up.
01:12:42
>> We should all just get a room and just
01:12:43
have one big huge orgy cuz they're all
01:12:45
just useless. It's like this like sexual
01:12:47
tension that we just need to release
01:12:48
somehow.
01:12:52
>> Your feet.
01:12:55
We need to get Mercury's back.
01:13:05
I'm going all in.