00:00:00
Jason, do you want to tell us about your
00:00:02
new favorite podcast? Oh, it's so good.
00:00:06
My feed is now because, you know, since
00:00:10
cancel culture ended, Sachs, everybody
00:00:12
uses the R word and the f word right
00:00:14
now. My entire feed on Instagram is
00:00:17
either gay or down syndrome or bulldogs.
00:00:20
It's one of those three. And then I
00:00:22
stumbled upon the Miss Thing pod,
00:00:25
Miss Thing. And they do a bit called gay
00:00:28
name, straight name. Here's gay name or
00:00:30
straight name for David. This good news
00:00:33
and bad news freeird. Here we go.
00:00:35
>> Gay name or straight name?
00:00:38
>> David.
00:00:40
>> David to me is straight.
00:00:42
>> Okay. But he has my perfect body. It can
00:00:46
be confusing because I'm kind of like,
00:00:48
are you gay? And it's like, no, I just
00:00:50
want to be you, David.
00:00:51
>> Totally. Well, it's so like the
00:00:54
Michelangelo's David the male ideal.
00:00:56
It's like incredible body kind of small.
00:00:58
Sorry.
00:00:59
>> Yeah.
00:00:59
>> Oh,
00:01:00
>> it's a little rough.
00:01:02
>> What?
00:01:03
>> What are you watching there, Jal?
00:01:04
>> They basically nailed these two, but
00:01:06
okay, keep going.
00:01:07
>> I don't think Chimoth is on their short
00:01:09
list, but I know Jason will come up at
00:01:10
some point.
00:01:11
>> Gay name or straight.
00:01:13
>> Maybe this is it.
00:01:14
>> Chimoth
00:01:16
on the count of three. Yeah.
00:01:18
>> Three, two, one. gay. I'm seeing like
00:01:22
Italian sweater, like really kind of
00:01:25
like a loud vibrant sweater. He like
00:01:28
wears it to like poker night with his
00:01:30
like his boys and like not I'm not
00:01:33
talking like straight poker. I'm talking
00:01:35
like gay poker nights like at the bar.
00:01:38
>> Yeah. Always talking about wine.
00:01:40
>> Talks about wines.
00:01:42
>> Always sort of like
00:01:44
Yeah, exactly.
00:01:45
>> Yep. Also, it's so like the guy at the
00:01:48
gym taking off his shirt, taking
00:01:51
selfies.
00:01:52
>> Yeah. And everyone else is kind of like,
00:01:54
"Excuse me, Chimoth. I'd like to use the
00:01:56
mirror. I'd like to see myself.
00:01:58
>> See you at the next day." Poker night.
00:02:00
>> Totally.
00:02:01
>> You bring the wine.
00:02:03
>> We'll bring the sweater.
00:02:04
>> Yeah,
00:02:04
>> there it is. Wow. They did do.
00:02:06
>> That is fantastic. That is fantastic. A
00:02:09
shout out to my guys at the Miss Ding
00:02:11
podcast.
00:02:12
>> Wow, that was awesome. I think I'm gay.
00:02:16
I never knew.
00:02:19
>> Let your winners ride.
00:02:26
>> And it said we open sourced it to the
00:02:28
fans and they've just gone crazy with
00:02:29
it.
00:02:35
>> What did you do like cameo? Did you pay
00:02:37
them to do that?
00:02:37
>> Did it for me as a favor. So they did it
00:02:39
for
00:02:40
>> That's awesome. Well, thanks to those
00:02:41
guys to the miss. I've seen those guys
00:02:44
before in clips. I find them very funny.
00:02:46
>> It's so great. Shout out to my guys.
00:02:49
>> That was awesome.
00:02:50
>> All right, everybody. Seriously, welcome
00:02:52
back to the number one podcast in the
00:02:54
world. It's the All-In podcast with me
00:02:56
again, David Freebergia, David Saxs, and
00:03:00
of course, I'm Jason Calcanis. You can
00:03:01
call me Jay Cal if you're here for the
00:03:03
first time. Topic one, open AI. They
00:03:07
missed their targets for chat GPT
00:03:09
Freedberg both on users and revenue.
00:03:13
Let's talk about it. The Wall Street
00:03:14
Journal says in a uh a breaking
00:03:18
investigative report on Tuesday that
00:03:21
OpenAI expected to hit 1 billion wows
00:03:24
weekly active users before the end of
00:03:26
2025. They missed that and they still
00:03:28
haven't hit the milestone 4 months into
00:03:30
2026. Also, Chamath, they missed their
00:03:33
2025 revenue target for Chad GPT. Exact
00:03:37
number wasn't specified, but as we've
00:03:39
talked about here, they're at a 2030
00:03:41
billion run rate. There's a little bit
00:03:43
of accounting nuance that is yet to be
00:03:46
worked out in the industry. Two reasons
00:03:48
why this matters. Sachs, OpenAI has $600
00:03:52
billion in spending commitments for
00:03:54
compute. Just to put that in
00:03:56
perspective, that's about what they're
00:03:58
trading for on secondary markets. In
00:04:00
other words, the entire value of the
00:04:01
Open AI enterprise equals their spend
00:04:04
commitments in the coming year. CFO
00:04:05
Sarah Frier, who is coming to liquidity,
00:04:08
is reportedly worried, hey, that revenue
00:04:10
isn't growing fast enough to keep up
00:04:13
with expense and OpenAI wants to IPO
00:04:16
later this year. This has put Frier and
00:04:19
Oughtman in conflict or uh maybe there's
00:04:22
some natural tension there. Frier
00:04:24
doesn't think OpenAI is ready for public
00:04:26
reporting standards. According to the
00:04:28
Wall Street Journal, Altman obviously
00:04:30
wants to move faster, so they released a
00:04:33
joint statement. this is ridiculous yada
00:04:35
yada yada. Let's go to you Saxs. What do
00:04:38
you think's going on here? Are these
00:04:39
major headwinds or is this just managing
00:04:41
expectations as the leader of the pack
00:04:44
in the most important race of our
00:04:46
lifetimes, the race towards super
00:04:48
intelligence?
00:04:49
>> Well, I actually have a little bit of a
00:04:51
contrarian take on this. I know that
00:04:53
OpenAI had a really bad week. Like you
00:04:55
said, they had that Wall Street Journal
00:04:57
article which said that they missed
00:04:58
their numbers. They missed their 1
00:05:00
billion user growth target. They missed
00:05:03
their revenue numbers. That's called
00:05:04
into question whether they can afford
00:05:06
the data center commitments that they've
00:05:08
made. And then in addition to that,
00:05:10
they've also had the lawsuit with with
00:05:12
Elon happening this week. So in the
00:05:14
press, it ended up being I think a
00:05:15
pretty bad week for them. But I have a
00:05:16
contrarian take on this, which is I
00:05:18
think that over the past week or two, if
00:05:20
you look at kind of what's happening at
00:05:22
the product level, it's been a pretty
00:05:23
good couple of weeks for them. They
00:05:25
released chat GBT 5.5 and the reviews
00:05:29
from, you know, people I talked to in
00:05:30
Silicon Valley have been really strong.
00:05:32
You talk to developers, coders, they're
00:05:34
very happy with it. At the same time,
00:05:37
Opus 4.7, which is the latest anthropic
00:05:40
release, appears to be a bust. People
00:05:41
are complaining about it. They're in a
00:05:43
lot of cases are rolling back to 4.6.
00:05:46
They're saying that Opus 4.7 is
00:05:48
rationing compute. It's reducing
00:05:50
thinking time, not as good. there were
00:05:52
some bugs and clawed. So if you just
00:05:55
compare chat GPT 5.5 to Opus 4.7, it
00:06:00
does appear that OpenAI has had a better
00:06:02
couple of weeks on a product level. And
00:06:04
I think there's reason to believe that
00:06:07
the product improvements will continue.
00:06:09
GPT 5.5 is based on a new base model
00:06:12
called Spud, which is the first base
00:06:15
model upgrade they've done in I don't
00:06:16
know over a year. and having a new base
00:06:18
model will pave the way for future
00:06:20
improvements as well. So I think OpenAI
00:06:23
is feeling pretty optimistic about their
00:06:25
product right now and I think you're
00:06:26
starting to see on X some of the
00:06:29
developer mojo is shifting. I'm seeing a
00:06:31
lot of people saying that they are
00:06:33
shifting their their coding usage from
00:06:36
Opus to GPT 5.5.
00:06:39
So I think that SAM may end up being
00:06:43
right but for the wrong reason. And what
00:06:46
I mean by that is that when he made
00:06:49
these big compute commitments, it was
00:06:52
based on those estimates of hitting the
00:06:54
billion users on the consumer side and
00:06:56
hitting those revenue targets. The
00:06:58
consumer business ended up being weak.
00:07:00
So they missed those targets. But in the
00:07:02
meantime, coding has become the
00:07:04
allimportant sector of AI. And because
00:07:09
they made all these compute commitments
00:07:11
and they built out these data centers,
00:07:13
they have more compute than anthropic
00:07:15
right now. Anthropic is token
00:07:17
constrained. It's reducing their ability
00:07:19
to serve mythos, for example. It's
00:07:21
causing them to engage in compute gating
00:07:23
with Opus 4.7. And I understand why
00:07:26
Daario made that decision. I'm not
00:07:27
saying I mean it was a prudent business
00:07:29
decision. I'm not criticizing him for
00:07:30
it, but I think again I think Sam may
00:07:33
end up being right here for the wrong
00:07:35
reason, which is he missed on consumer,
00:07:37
but enterprise is going gang busters and
00:07:40
is giving him the ability now, I think,
00:07:42
to catch up on code. your poly market
00:07:44
>> which is the all important market right
00:07:46
now
00:07:46
>> of course and we talked about gro and
00:07:48
cursor teaming up last week Elon and the
00:07:50
team over there poly market showing now
00:07:52
a 32% chance that openai goes public by
00:07:55
the end of 2026 this is down from 60% in
00:07:58
December and Shimath you gave a bit of a
00:08:01
warning hey there's only so many dollars
00:08:04
to go around SpaceX IPO is obviously
00:08:06
getting out first and now if openai
00:08:10
doesn't go out this year and anthropic
00:08:13
does the sets up an interesting dynamic.
00:08:14
What are your thoughts here generally
00:08:16
speaking about the massive commitment
00:08:19
that OpenAI has made? Are they going to
00:08:21
run off the cliff or will it wind up
00:08:24
being brilliant? Uh even if it wasn't
00:08:26
strategically for the exact reasons,
00:08:29
>> I think they're going to be fine. I
00:08:30
think this is a multi-t trillion dollar
00:08:32
company. I think Anthropic is a multi-
00:08:34
trillion dollar company. I think the
00:08:36
thing that's happening right now is uh a
00:08:39
complete misunderstanding
00:08:41
of what's actually happening
00:08:44
inside of the world of AI. And there is
00:08:46
one very specific choke point
00:08:49
that is constraining everything which is
00:08:50
access to the power that's necessary to
00:08:53
drive these tokens. To the extent that
00:08:55
open AI missed, I think what that is is
00:08:58
an insight to not enough compute
00:09:00
capacity today. And that problem is only
00:09:03
getting worse. You've already seen that
00:09:05
with Anthropic as well where they just
00:09:08
found a way to economically induce
00:09:11
Amazon to give them enough capacity so
00:09:14
that you don't have to route through
00:09:16
bedrock to get to the anthropic models.
00:09:19
You're also seeing them do
00:09:20
differentiated deals now with economic
00:09:23
participation on top of what they
00:09:25
already had from folks like Google to
00:09:26
give them more capacity. What is my
00:09:28
point? Everything in this market is
00:09:31
power constrained. The reason that these
00:09:34
folks may miss a number or a forecast
00:09:36
have nothing to do with demand. It is
00:09:39
entirely 100% due to the supply of the
00:09:43
power necessary to generate the output
00:09:44
token. There is a really interesting
00:09:46
thing that was just announced today that
00:09:49
will make this problem even worse, which
00:09:52
is what you're starting to see now is
00:09:55
backlogs build up of not just the access
00:09:59
to the power, but then the componentry
00:10:01
that's actually necessary. Not just
00:10:03
resips and not just NAT gas turbines,
00:10:05
but now you're talking about
00:10:06
transformers and all the actual tactical
00:10:08
grid infrastructure.
00:10:10
Why is this important? If you look at
00:10:13
the actual amount of gigawatts that are
00:10:16
under construction, we have a huge
00:10:18
mismatch now,
00:10:20
people have announced all these
00:10:22
projects, Jason,
00:10:24
but less than half of it is actually
00:10:26
being built. Less than half. Most of it
00:10:29
is stuck in red tape.
00:10:31
Most of that is because there are these
00:10:33
supply chain delays. So there's no
00:10:36
credible strategy to turn any of this
00:10:38
stuff on. Who will this hurt? It will
00:10:41
hurt Anthropic and OpenAI the most. Who
00:10:44
will this benefit? It will benefit the
00:10:47
hyperscalers, specifically Oracle,
00:10:49
Amazon, Meta, Microsoft, and Google. And
00:10:52
now what you're going to see is a
00:10:55
negotiation and a trade back and forth.
00:10:57
How much equity do I have to give up?
00:11:00
How much control do I have to give up to
00:11:01
get access to the compute versus how
00:11:04
badly will I miss my growth forecasts if
00:11:07
I don't? And now what that means is, and
00:11:09
we spoke about this last week, that's a
00:11:10
huge lane for Grock to just run through
00:11:13
and SpaceX to run through cuz they have
00:11:16
a ton of excess capacity. And so I think
00:11:19
the cursor deal was the appetizer. But
00:11:22
if I were Elon now, I'd be running all
00:11:24
over this market because if the models
00:11:27
catch up in quality, I think he could
00:11:29
also do something really crazy with
00:11:31
anthropic or open AI right now. Maybe
00:11:33
not open AI because of the
00:11:35
>> we'll get into the lawsuit in a bit
00:11:36
>> the baggage.
00:11:37
>> Yeah.
00:11:38
>> But man, he and Dario should do a deal
00:11:40
tomorrow.
00:11:41
>> So you're framing, hey, the the limited
00:11:44
resource here is compute. The demand is
00:11:47
off the charts.
00:11:48
>> No, the limiting resource is power.
00:11:50
power which then powers compute which
00:11:52
then provides tokens which then services
00:11:55
the massive uh developer and co-work and
00:11:59
all these other projects that consumers
00:12:01
and enterprises can't get enough of. Got
00:12:03
it.
00:12:03
>> And Jason, the other factor that
00:12:05
complicates that for anthropic and open
00:12:07
AI is all the stuff that's sort of
00:12:09
sitting around thumb twiddling. 40% of
00:12:13
that is going to get cancelled because
00:12:14
they've done such a poor job of creating
00:12:17
a good positive halo around AI that 40%
00:12:20
of all the announced projects get
00:12:22
cancelled because 40% of all projects in
00:12:25
the last four years have been cancelled.
00:12:28
Yeah. And there's there there are some
00:12:30
bad feelings about data centers, AI,
00:12:33
jobs, etc. And that's causing some
00:12:35
headwind. People are you literally doing
00:12:39
violent things in society and blaming
00:12:41
data centers and AI for it. I don't want
00:12:43
to give it too much air time. Freeberg,
00:12:45
what's your take on the chessboard we're
00:12:47
looking at here? Either through compute,
00:12:50
energy or through going public on a
00:12:53
business level, you know, the strategic
00:12:56
nature of capital, compute and energy
00:12:59
now playing a role in this massive
00:13:01
amount of demand. still a ball in the
00:13:04
air kind of game. BCG had this theory, I
00:13:09
think I talked about this once before,
00:13:10
called the rule of three where they've
00:13:12
shown time and again that any stable,
00:13:14
mature, competitive market evolves to a
00:13:18
4:21
00:13:20
ratio of market share for basically 90%
00:13:22
of the market. So there's a market
00:13:23
leader that has four times the market
00:13:26
share of the second place, that's two
00:13:28
times the market share of the third
00:13:29
place. This is the case in pretty much
00:13:30
every mature kind of competitive market.
00:13:33
So you can kind of think about AI
00:13:34
probably evolving into a consumer market
00:13:36
and an enterprise market. Open AAI, even
00:13:39
if they're not at a billion, they're
00:13:40
still at 900 million weekly users, which
00:13:43
is well ahead of whatever Claude is at.
00:13:46
I think Claude is like probably subund
00:13:48
million sacks, you may know. And then
00:13:49
Gemini is probably closer to them at 700
00:13:52
to a billion somewhere in that range.
00:13:54
Probably pretty neck and neck with open
00:13:56
AI. So, you know, the consumer market
00:13:59
looks like it's trending towards a chat
00:14:02
GPT/Google
00:14:04
fight for first place and second place
00:14:06
and then probably anthropic in third
00:14:08
place and maybe Elon emerges and takes
00:14:10
off enabled by his compute capacity and
00:14:13
then the enterprise market is a little
00:14:14
bit of a different story and that's its
00:14:16
own market which is kind of anthropic or
00:14:18
probably Google in the lead actually if
00:14:20
you look at all the vertex use. Google
00:14:22
claims that 75% of GCP customers are
00:14:26
active users of Vertex. So there's
00:14:29
probably a pretty sizable
00:14:32
market share that Google's captured on
00:14:33
the enterprise side as well. This is
00:14:34
also probably why Google stock has
00:14:36
absolutely ripped over the last couple
00:14:38
of months is they're literally in first
00:14:39
place or fighting for first place in
00:14:42
enterprise and consumer.
00:14:44
But I still think that there's a lot of
00:14:46
opportunity to Chimoff's point about the
00:14:48
compute and energy capacity constraints
00:14:50
in improving how we actually scale and
00:14:53
deploy models in both the enterprise and
00:14:56
the consumer setting. And it is such
00:14:57
early days and I just want to highlight
00:14:59
this paper that came out from MIT from
00:15:02
these two scientists and these guys
00:15:05
published a paper on pruning techniques
00:15:07
and neural networks. This paper showed
00:15:09
that you could actually reduce the size
00:15:10
of these networks by 90%.
00:15:13
And get the same accuracy out by pruning
00:15:17
very large models down to smaller
00:15:19
models. And then you can make a
00:15:20
selection on which model to run for
00:15:22
inference. And by doing this, you can
00:15:24
actually reduce inference costs by 10x.
00:15:27
You can get 10x the output per energy
00:15:29
unit that goes into the data center with
00:15:31
no loss of accuracy. And so it's a
00:15:34
really interesting call it algorithmic
00:15:35
technique that can be applied to the
00:15:38
existing large models to actually make
00:15:40
them much lower energy use. So if you
00:15:42
think about it, you're firing up a very
00:15:44
large model to answer a very simple
00:15:46
question. You can actually prune away
00:15:48
that model. Now this is probably going
00:15:51
to be the case in AI applications as it
00:15:54
is in traditional Google search. There's
00:15:56
a long tale of searches, but there's a
00:15:58
few searches that account for a large
00:16:00
percentage of search volume. It's like
00:16:02
what is the weather? What are the movies
00:16:04
times? You know, what's the stock price?
00:16:06
Like there's a certain set of things
00:16:08
that make up the bulk of consumer
00:16:10
energy. And there's probably a certain
00:16:11
set of things that probably make up the
00:16:12
bulk of coding output as well. And so if
00:16:15
you can get that 80% of searches or chat
00:16:18
interfaces or coding requests reduced
00:16:22
down through pruning techniques to
00:16:23
smaller models and then you have a whole
00:16:25
set of smaller models that can be called
00:16:26
dynamically and you reduce inference
00:16:28
cost by 90%. you can make much more use,
00:16:31
call it 10 times the use on data center
00:16:34
and energy capacity than we can today.
00:16:36
So I would argue that we're still in the
00:16:37
very early days of getting efficiency in
00:16:40
terms of output and tokens and we're
00:16:42
just in the very kind of early stage of
00:16:43
that which also unlocks the opportunity
00:16:45
for guys like Elon to reinvent how this
00:16:47
is done and potentially compete pretty
00:16:49
aggressively.
00:16:50
>> There are two ways to win. You could
00:16:51
throw compute at it or you can do SLM's
00:16:54
small language models and V SLM's
00:16:57
verticaliz small language model. So if
00:17:00
you had a verticalized small language
00:17:01
model for the weather, let's say that
00:17:03
doesn't exist, but uh you can you can
00:17:05
use it as an example. They will have one
00:17:07
for travel as an example. When you hit
00:17:09
Google for flight information, it's
00:17:11
obviously going to route you to
00:17:13
something lighter and faster that uses
00:17:15
Google flights. And Google flights has
00:17:16
been incor incorporated into Gemini.
00:17:19
Gemini now is right behind 700 750
00:17:23
million users
00:17:25
>> and it's exactly what we discussed I
00:17:27
don't know 18 months ago on this podcast
00:17:28
Freeberg that what if they put it at the
00:17:30
top
00:17:31
>> and what would that do to their search
00:17:33
revenue search revenue is surging and
00:17:37
they're also surging so they figured out
00:17:39
a way to balance those two competing
00:17:41
forces having search results that are AI
00:17:43
enabled and still getting people to
00:17:46
click on links they've done it
00:17:47
brilliantly apparently and the stock is
00:17:49
rewarding.
00:17:50
>> I'll just add one statement to what you
00:17:51
said, which is like you're using what I
00:17:53
would call a humanistic on
00:17:58
humans don't intuitively know what this
00:18:01
model is. It's not just a verticalized
00:18:02
model, but there are going to be models
00:18:05
that will be discovered through
00:18:07
automated pruning techniques that will
00:18:09
then be working in concert. So, lots of
00:18:12
small models that link together. And we
00:18:13
don't define each model by some human
00:18:15
heristic like this is a search travel
00:18:17
model.
00:18:18
>> This is a maps model. We don't we don't
00:18:20
know why these models work the way they
00:18:22
do when they get broken down. But I do
00:18:24
think that that's really where the
00:18:25
evolution is happening. So effectively a
00:18:27
model becomes a macro model. It's got
00:18:29
lots of smaller models underneath it
00:18:31
that can be dynamically called and that
00:18:33
allows you to have 10x the inference for
00:18:35
the same unit of energy.
00:18:36
>> Sax,
00:18:37
>> let me just build on your point about
00:18:38
Google. Jcal, I would say that if
00:18:40
there's a single reason why OpenAI did
00:18:44
not hit its user targets and its revenue
00:18:48
targets, certainly around consumer,
00:18:50
you'd have to say it's because Google
00:18:52
managed to take meaningful share, you
00:18:55
know, they were basically nowhere a year
00:18:58
or so ago. Sergey came out of
00:19:00
retirement, helped focus the company,
00:19:03
and like you said, they did a brilliant
00:19:05
job improving Gemini and putting it at
00:19:07
the top of search, incorporating it.
00:19:09
Now, that being said, again, I don't
00:19:11
think the news is all bad for OpenAI
00:19:13
because I do think that the 5.5 release
00:19:15
was great. We're hearing really good
00:19:17
things about Codeex. I do think that
00:19:19
Codeex is taking share in coding tokens
00:19:23
right now. And I just think we're in a
00:19:25
really interesting place where these
00:19:27
companies are constantly oneuping each
00:19:29
other. I mean, two weeks ago it looked
00:19:31
like Anthropic was going to be
00:19:32
completely dominant, right? I mean,
00:19:34
Anthropic was growing at 10x. Open AAI
00:19:36
was growing at 3x and it looked like
00:19:38
>> and then the servers started going down.
00:19:39
Did you see that this week? The server
00:19:41
going down. People were in my office
00:19:43
were complaining we can't get on claud.
00:19:45
>> Listen, competition brings out the best
00:19:47
in everyone. Anthropic forced open AI to
00:19:50
compete. Google's forced open AI to
00:19:52
compete in consumer. I just hope the
00:19:54
market stays competitive for as long as
00:19:55
possible. I do think that's what's best
00:19:57
for consumers, our economy, and for our
00:20:00
country overall. Let me just say one
00:20:03
other area where I think OpenAI had a
00:20:06
good week is in this red-hot area of
00:20:10
cyber. Obviously, Anthropic made a huge
00:20:13
splash with Mythos. It hasn't been
00:20:14
commercially released. Their compute
00:20:16
constraint, but as a proof of concept or
00:20:18
training model, it hit a new level of
00:20:20
capabilities with cyber. But now OpenAI
00:20:22
has released a new model called GPT 5.5
00:20:25
cyber which has just been through a
00:20:26
bunch of tests and they've shown this
00:20:29
was testing done by the AI security
00:20:31
institute that GPT 5.5 is the second
00:20:34
model to complete one of their
00:20:36
multi-step cyber attack simulations end
00:20:38
to end. So it has the same level of
00:20:40
capability as Mythos and it does appear
00:20:43
to be commercially
00:20:46
ready. You know, they've got the compute
00:20:47
to serve it. So I do think that that's a
00:20:51
big accomplishment. I mean, look, we
00:20:53
knew that other cyber models were
00:20:55
coming. It wasn't just going to be
00:20:56
Mythos. In fact, within 6 months or so,
00:20:59
all the Frontier models are going to
00:21:00
have Mythos level cyber capability. But
00:21:03
it's impressive that OpenAI got this GPT
00:21:07
5.5 cyber out. so quickly and I think
00:21:11
5.5 might be the first cyber model that
00:21:14
cyber defenders actually get to use
00:21:17
because again I don't think they're as
00:21:18
compute constrained as anthropic is
00:21:20
>> and this is an incredible opportunity
00:21:22
you know for the crowd strikes and
00:21:24
PaloAlto networks of the world both of
00:21:26
which have been on the program they come
00:21:29
out and they start attacking this space
00:21:31
man you could really
00:21:33
see everything get tightened up and this
00:21:37
could be an incredible revenue stream
00:21:39
for everybody who's got whether it's
00:21:41
cursor claude or open eye or Gemini.
00:21:44
This is an amazing opportunity to
00:21:45
tighten up as much as it is to get
00:21:48
attacked.
00:21:48
>> Can I make a point about that? Cuz look,
00:21:50
there is so much fear right now almost
00:21:52
the level of panic about mythos. People
00:21:54
are treating it like a doomsday weapon
00:21:56
or something like that. It's not. is
00:21:58
simply that the frontier models have
00:22:00
reached the point where they're capable
00:22:02
of automating cyber activities just like
00:22:05
they're capable of automating coding.
00:22:08
But that means that a model could power
00:22:11
up a cyber attacker or cyber defender
00:22:13
the same way they can power up a coder
00:22:16
and allow them to discover a lot more
00:22:18
vulnerabilities. So there is obviously a
00:22:20
risk there. But I think it's important
00:22:22
to understand that Mythos or GPT 5.5, it
00:22:26
doesn't create the vulnerabilities. It
00:22:28
just discovers them. The bugs were
00:22:30
already in the code. They were sitting
00:22:31
there waiting for some hacker to
00:22:33
discover. If we can now use AI to find
00:22:37
these bugs in advance, these
00:22:39
vulnerabilities and patch them, then you
00:22:41
actually harden our infrastructure and
00:22:44
and you harden our security. I also
00:22:46
believe that this leap from let's call
00:22:49
it preAI cyber to post AAI cyber it's
00:22:52
going to be I think a big one-time
00:22:53
upgrade cycle because again you're going
00:22:55
to find all these dormant bugs and
00:22:57
vulnerabilities but I think that once we
00:23:00
get past that upgrade cycle you're going
00:23:02
to reach a new equilibrium between AI
00:23:04
powered cyber offense and AI powered
00:23:06
cyber defense it's going to become a lot
00:23:08
more normal it's not going to feel like
00:23:10
this huge disruption which is to say I
00:23:12
think you know people are treating this
00:23:14
as like some existential threat. I don't
00:23:17
think it is as long as everyone does
00:23:18
what they're supposed to do, which is
00:23:19
use the new capabilities to harden their
00:23:22
code bases and infrastructure and
00:23:24
security before the hackers get a hold
00:23:25
of these capabilities.
00:23:27
>> Yeah. And if Chimath, if you were to
00:23:29
look at this to build on Sax's point,
00:23:32
there are about 5 million or so security
00:23:34
experts in the world. We talked about
00:23:36
token cost. 40 hours of tokens just
00:23:39
pounding it, you know, a week. You could
00:23:42
create another five million for a
00:23:44
hundred dollars per chief security
00:23:47
officer per security expert. So it's the
00:23:50
volume of security expert agent saxs to
00:23:52
your point. Yeah. You could have 50 50
00:23:55
million of them 100 million of them.
00:23:57
They're not finding something unique.
00:23:59
They're just they never sleep. They're
00:24:01
relentless in their pursuit of these
00:24:03
problems. It's a really great point.
00:24:05
>> Just kind of just refine that. So yeah,
00:24:06
there's probably 5 million people in the
00:24:07
cyber industry, but there's probably
00:24:09
only a few thousand really elite
00:24:10
hackers.
00:24:11
>> Sure,
00:24:12
>> those hackers didn't have the time to go
00:24:14
after the entire surface area of every
00:24:17
possible attack vector out there. And so
00:24:19
if you train a model to do what they do,
00:24:22
obviously, like you said, it can operate
00:24:24
with a scale and speed that a human
00:24:26
hacker can't. So obviously, you know,
00:24:28
what you need to do is get these tools
00:24:30
in the hands of the white hats, let them
00:24:33
do the cyber attacks themselves to then
00:24:35
find the vulnerabilities and patch them
00:24:37
before the black hats get a hold of
00:24:39
these capabilities. But I think it's
00:24:41
just just one last point on this, I'll
00:24:42
stop. It's just it's really important to
00:24:44
understand that the Chinese models are
00:24:46
going to have these capabilities within
00:24:47
approximately 6 months.
00:24:49
>> Oh, they have them now in Deep Seek 4
00:24:50
for sure. They've got some level. Well,
00:24:52
no. Deepc4, I mean, Dec 4 is impressive
00:24:55
in a lot of ways, but its capability is
00:24:57
not at the frontier. It's maybe 80 80
00:25:00
85%. Let's call it the American
00:25:02
frontier.
00:25:02
>> Chimoff, you wanted to get in on this.
00:25:03
Let's get Chim in.
00:25:04
>> Two things. The reason that this is even
00:25:08
possible is because humans are
00:25:10
errorprone and when humans code, they
00:25:12
create holes. And so, humans exploiting
00:25:15
humans is where we've been for a long
00:25:18
time. Now we have computers exploiting
00:25:20
humans because the computers go and seek
00:25:21
out all these bugs that humans wrote.
00:25:25
In the next phase it'll be machines
00:25:27
versus machines.
00:25:29
And so I think the nature of cyber is
00:25:31
going to completely change. Probably in
00:25:32
the next five or six years there'll be
00:25:34
so much reason to rewrite all of the
00:25:38
software that runs the world.
00:25:41
In one part because you're going to be
00:25:42
asked to show more operating leverage
00:25:44
and revenue growth, but in another part
00:25:47
because everything else that was
00:25:48
handmade in the past is just
00:25:50
fundamentally insecure. Either way, all
00:25:52
roads will lead to all the operational
00:25:54
software that runs the world will get
00:25:56
rewritten. More and more of it will be
00:25:58
written by machines. More and more of it
00:26:00
will be impregnable as a result. But
00:26:02
then the cyber threat actually will only
00:26:04
increase because then you're going to
00:26:06
try to figure out how to use a machine
00:26:07
to inject something into another machine
00:26:09
so that some agentic loop inject some
00:26:11
malware or injects a bad token. And I
00:26:14
think that's a very complicated thing.
00:26:16
What I will tell you is
00:26:18
I'm not even sure if I'm allowed to say
00:26:19
this, but
00:26:22
a very good probably the best cyber
00:26:24
security company in the world run by one
00:26:26
of the very best CEOs in the world who
00:26:29
may or may not be speaking at liquidity
00:26:32
would tell you that they have penetrated
00:26:36
and can essentially
00:26:39
manipulate every model. Let me just let
00:26:42
me just say it roughly that way.
00:26:43
>> Okay, perfect. Yeah. And I uh at the
00:26:46
breakthrough prize uh which three of the
00:26:47
four of us were at I talked to George
00:26:49
Kurtz the other person you were kind of
00:26:51
describing was not that
00:26:52
>> sitting beside Nash. Yeah I'm talking
00:26:54
about
00:26:54
>> Nash and George are the two guys leading
00:26:56
this Palo Alto Networks Crowd Strike and
00:26:59
they understand the what George told me
00:27:01
was there is just a line out the door of
00:27:03
people who want this product or service.
00:27:06
And if you look at it, Freeberg, like
00:27:09
the murder rate, like we're sitting here
00:27:11
with the lowest murder rate in the
00:27:12
history of humanity. It has gone down
00:27:15
massively in our lifetimes, but
00:27:17
massively over the arc of history. I
00:27:18
think that's what's going to happen with
00:27:19
cyber. There is only so many attack
00:27:21
vectors, and the remaining attack
00:27:23
vectors are just going to be human
00:27:24
factors, right, Freeberg? That's always
00:27:26
been the case. And as we make the
00:27:28
software more resilient, then the the
00:27:31
weak link is, you know, the secretary
00:27:33
who puts her post-it note, you know,
00:27:36
with the password there or the
00:27:38
accountant who, you know, uses their
00:27:40
dog's name plus one, two, three for
00:27:42
their password, right? That's the the
00:27:44
historical one. Okay, let's any anything
00:27:46
you want to add, Free Bird, as we wrap
00:27:48
there? Oh, that's
00:27:49
>> Why is your bed so messy, by the way?
00:27:50
Why can't you just ask the room service
00:27:52
to come in?
00:27:52
>> Listen, I'll tell I can tell you what
00:27:54
happened. Listen, I'm here in Atlanta.
00:27:55
And also, why don't you have a suite
00:27:57
like where there's two rooms? Like, is
00:27:58
it just one room? This hotel only has
00:27:59
one.
00:28:00
>> It's just one room. Yes.
00:28:01
>> You know, either you're cheap or poor.
00:28:03
Which one is it? I'm cheap. Here's I'll
00:28:06
tell you what. Here's a situation. I'm
00:28:08
in Atlanta for the Knicks game tonight.
00:28:09
>> You're in one room.
00:28:10
>> Here's what I do. I just want to explain
00:28:12
to you value for value. Some people
00:28:13
spend their money on private jets and
00:28:15
they spend $30,000 flying to Atlanta. I
00:28:18
spend 30,000 on courtside seats. I don't
00:28:20
want the suite. I want to put it into
00:28:21
the seat side.
00:28:23
>> You can do both. I guess I could do
00:28:24
both, too. I don't I'm I'm in the
00:28:26
process of becoming
00:28:28
>> I don't understand
00:28:29
>> of embracing my richness. Okay.
00:28:31
>> If you've already convinced yourself
00:28:32
that you should spend $30,000 for
00:28:35
courtside tickets, which I think is
00:28:37
outrageous, but okay. You've already
00:28:38
convinced yourself like 10k each, but
00:28:40
yeah. Yeah.
00:28:40
>> A hotel room that has two rooms. Okay.
00:28:44
>> Probably cost 15% more than what you're
00:28:46
paying.
00:28:46
>> It's 2x, but yes, you're right.
00:28:48
>> I'll get the I'll get the hotel room. 20
00:28:50
or 20% more, but you room like 200 bucks
00:28:53
a night. So you pay 400 a night. You get
00:28:54
another hotel.
00:28:55
>> I mean it's it's Atlanta. The most I'm
00:28:57
in the best hotel the most expensive
00:28:58
hotel is 500 a night in Atlanta. It's no
00:29:00
big deal. But everything's sold out
00:29:02
because all the Knicks people are coming
00:29:03
here.
00:29:03
>> So you're selling me double that would
00:29:04
have been a thousand and you couldn't
00:29:05
spend,000.
00:29:06
>> Everything is sold out because the
00:29:08
Knicks are here.
00:29:08
>> So we have to look at your dirty beds.
00:29:10
>> It's gross.
00:29:11
>> The bed's not that dirty. Come on. Just
00:29:13
deal with it. Okay. Take it out and
00:29:14
post.
00:29:15
>> I have a private jet story about flying
00:29:17
to Atlanta.
00:29:20
>> You reminded me. Okay. So yeah, there
00:29:22
was some event there. So I I flew my
00:29:23
team there, you know, there's a few
00:29:26
people on my plane and it's kind of a
00:29:28
long flight. Was it like 4 hours or
00:29:29
something
00:29:30
>> from the back? Yeah.
00:29:31
>> Yeah. So I went in the back to to sleep.
00:29:34
Well, first, you know, we we started the
00:29:35
flight and I had a few bottles of Papy
00:29:37
Van Winkle on on the plane. And so we
00:29:40
started off with like a drink and then I
00:29:42
went in the back and and fell asleep and
00:29:43
I woke up basically when we landed.
00:29:45
>> So I come out and like all three bottles
00:29:47
are basically cashed of like Happy Van
00:29:50
Wink.
00:29:50
>> Oops. Those were like two grand a
00:29:52
bottle.
00:29:53
>> No, no, they're more. These were like
00:29:54
antique bottles. Like one of them was
00:29:56
>> I have one of those from your plane. I
00:29:57
have one of those from the old Falcon.
00:29:58
Yeah.
00:29:59
>> Yeah. They were like these vintage
00:30:01
>> $4,000. I remember. Yeah.
00:30:03
>> Anyway, you can't even find this [ __ ]
00:30:04
anymore. So these guys, they asked me
00:30:07
like when we land like, "Hey, Sax, how
00:30:09
much did it cost for you to fly us to
00:30:12
this event?" And I said, "Well, about
00:30:14
$8,000 in jet fuel and about $12,000 of
00:30:17
Happy Van Winkle."
00:30:20
Well, you gota you got to fuel the the
00:30:22
vibes as well as the plane. It's uh
00:30:25
>> Is Atlanta nice? I've never really
00:30:27
>> Do those people still work for you or
00:30:28
are they are they uh
00:30:31
>> they called in Atlanta? Um is Atlanta
00:30:33
nice? Listen, last year I went to the
00:30:34
Detroit games and that city was on the
00:30:36
rebound. Atlanta has an incredible
00:30:37
opportunity to rebound. I'll say it that
00:30:40
way. There's a great opportunity for
00:30:42
them to upgrade the city. I I went to
00:30:44
Waffle House at midnight last night.
00:30:46
There was no shootings. Okay, let's keep
00:30:47
moving. By the way, do you get royalty
00:30:49
points at the Best Western Atlanta or
00:30:51
No,
00:30:52
>> I get double points because I use my
00:30:54
Best Western uh Visa card. Yeah, it's
00:30:56
everywhere you want it to be. All right.
00:30:59
Use the promo code Jcal and get a
00:31:01
thousand extra points. In other Open AI
00:31:04
news, Musk versus Alman, the trial of
00:31:08
the century or maybe the decade has
00:31:11
started. Elon is of course accusing Open
00:31:13
AI of breach of charitable trust, unjust
00:31:15
enrichment. He's accusing Open AAI of
00:31:18
essentially flipping a nonprofit into a
00:31:21
for-profit. He's seeking 150 billion in
00:31:23
damages that they revert back to a
00:31:25
nonprofit that Alman and Brockman be
00:31:28
removed. And there were some fireworks
00:31:30
between Elon and the Open AI lawyers.
00:31:32
Elon kind of leveled up the discussion.
00:31:34
He said, quote, "If we make it okay to
00:31:37
loot a charity, the entire foundation of
00:31:39
charitable giving in America will be
00:31:42
destroyed. That's my concern."
00:31:44
Obviously, there's a ton of interesting
00:31:47
nuances here. Specifically, Greg
00:31:49
Brockman keeping a diary where he was
00:31:52
journal maxing his plans uh like a Bond
00:31:55
villain here. And uh the excerpts from
00:31:59
his diary include conclusion, we truly
00:32:02
want the BC Corp. The true answer is
00:32:04
that we want Elon out. If 3 months later
00:32:07
we're doing BCorp, then it was a lie.
00:32:09
Can't see us turning this into a
00:32:11
forprofit without a nasty fight. I'm
00:32:13
just thinking about the office and we're
00:32:15
in the office and this story will
00:32:16
correctly be that we weren't honest with
00:32:19
him. In the end, it's still about
00:32:20
wanting a for-profit just without him.
00:32:22
yada yada yada. Freeberg, your thoughts
00:32:24
on this case? Is Elon going to win? I
00:32:27
just don't know why Greg Brockman's got
00:32:29
a freaking diary where he's like
00:32:31
literally documenting. I mean, I love
00:32:33
the guy, but what the [ __ ] is he
00:32:34
thinking? Like, you're just sitting here
00:32:36
at home and like, let me write about the
00:32:38
the crime I'm committing or let me write
00:32:40
like and let me record it. And by the
00:32:42
way, let me never delete it. I don't
00:32:44
understand this.
00:32:45
>> It's not just journal maxing. It's
00:32:47
discovery maxing.
00:32:49
>> It's smoking gun maxing.
00:32:51
>> I don't get it. I don't get it, man.
00:32:54
>> I mean, do you guys remember from the
00:32:57
wire in that scene where the guy's like,
00:33:00
"Is you taking notes on a criminal
00:33:02
conspiracy?"
00:33:04
He's got everybody in the room. Can we
00:33:06
play that clip? It's like,
00:33:08
>> "What are you doing, Greg? [ __ ] is you
00:33:10
taking notes on a criminal conspiracy?
00:33:14
What the [ __ ] is you thinking, man?
00:33:16
>> If you're going to commit a crime, you
00:33:18
do not write down the date and time of
00:33:21
the crime in your journal.
00:33:23
>> Well, look, we don't know it's a crime.
00:33:24
Let's not.
00:33:25
>> Okay, sure.
00:33:27
>> A crime, but yes, you keeping
00:33:30
shenanigans. Jamat, do you keep a diary?
00:33:32
>> What do you think, Juel? Do you keep a
00:33:34
diary?
00:33:35
>> I I believe ruminate. No, I'll tell you
00:33:38
right now, rumination is the path to
00:33:41
unhappiness. Nobody gives a [ __ ] about
00:33:43
your feelings. Writing your feelings
00:33:45
down is only going to make you
00:33:46
miserable. Talking to your spouse about
00:33:48
your feelings.
00:33:50
>> Just go to a beautiful dinner, sit
00:33:52
courtside at the next, and do what I've
00:33:54
been doing for 30 years.
00:33:56
>> [ __ ] maxing.
00:33:57
>> [ __ ] maxing. And the register goes up.
00:34:00
All you have to do is work. Start new
00:34:02
projects. Nine out of 10 foul. Place
00:34:04
nine out of 10 bets. One wins and you're
00:34:06
golden. Go sit courtside at the Knicks
00:34:08
game.
00:34:09
>> Keep going. Life's too short.
00:34:10
>> And just keep moving forward. Don't
00:34:12
write anything down. Period. Full stop.
00:34:15
>> It's good advice. Yeah, I just The
00:34:17
biggest surprise to me was this guy's
00:34:18
got a diary. I just I don't know anyone
00:34:20
that has a diary. I've never heard of
00:34:21
this. So anyway, that was shocking.
00:34:24
Besides that, I have no view on what's
00:34:26
going to happen with the case or what
00:34:27
the judge will do.
00:34:28
>> I have no comment on the case either. I
00:34:30
think it's weird that Poly Market hasn't
00:34:33
budged even as all of this discovery has
00:34:35
been published. It's effectively at 42
00:34:37
or 43% that Elon wins. So, one of the
00:34:41
friends in our group chat said what may
00:34:44
just happen is that Elon technically
00:34:45
wins and he's just credited back the $40
00:34:48
million. And so, maybe that's what this
00:34:51
poll is front running.
00:34:54
But on a totally separate note, I think
00:34:56
Jason, I know you say it as a joke, but
00:34:59
this idea of just keep moving forward,
00:35:02
don't ruminate, I think is very good
00:35:03
general life advice for everybody to
00:35:06
follow.
00:35:06
>> The modern-day therapy industrial
00:35:09
complex and the medication industrial
00:35:11
complex, I believe, is
00:35:14
>> around rumination. Well, it does pivot
00:35:15
around rumination.
00:35:16
>> Yes.
00:35:17
>> That is the gateway drug to all these
00:35:18
things.
00:35:19
>> Yep. Talk about your problems.
00:35:21
You know, when these people go to
00:35:22
therapy, you ever hear these people?
00:35:23
Howard Stern's like, "I've been in
00:35:24
therapy with the same person 2 or three
00:35:25
days a week for 40 years." I'm like,
00:35:27
"Okay, what's the incentive for the
00:35:28
therapist to stop charging you $1,200 an
00:35:31
hour? There is none." Then they lose a
00:35:33
revenue stream. They lose a customer.
00:35:34
It's all a giant [ __ ] fraud. Facts.
00:35:37
Uh, in terms of this case,
00:35:38
>> I wouldn't go that far. I do think that
00:35:40
there's a lot of value in kind of
00:35:42
untying some of these Gordian knots that
00:35:44
people have because of how they grew up.
00:35:47
But there's a difference between that
00:35:49
and being specific and just randomly
00:35:51
ruminating cuz I don't think there's a
00:35:53
lot of productive.
00:35:53
>> You've got an acute issue like in trauma
00:35:57
in your life. Yeah, sure. Unpack it,
00:35:59
figure it out. I'm just talking about
00:36:00
this neverending self-improvement,
00:36:03
you know, ruminating thing. Uh but
00:36:05
getting back on topic here, Saxs,
00:36:09
what's the And we're we're talking about
00:36:11
a jury, I believe, in Oakland.
00:36:13
>> No, but it's a bench trial. This is
00:36:14
important. It's a bench trial where the
00:36:16
jury is advisory in capacity,
00:36:19
>> but ultimately that judge, she will make
00:36:21
the final call
00:36:22
>> and she'll do the damages. And so, is
00:36:25
this a case sacks of like we've got a
00:36:28
Bay Area jury judge and we've got Elon
00:36:33
who's considered, you know, a bit
00:36:35
right-wing and people don't all agree in
00:36:37
that area in terms of his politics. And
00:36:39
then you have this Sam Alman New Yorker
00:36:42
story and people finding out that so
00:36:45
many different people feel they got
00:36:47
screwed by him. You put these two things
00:36:48
together, it's impossible to handicap
00:36:50
where this turns out. Sachs, your
00:36:52
thoughts?
00:36:53
>> Well, yeah, I don't think this is about
00:36:55
politics. I mean, I guess you could
00:36:58
argue that what Elon is seeking, which
00:37:01
is to protect the charity, is if
00:37:03
anything a left-coded sort of principle,
00:37:06
although I don't really think it's left
00:37:07
versus right. Look, I don't want to take
00:37:09
sides on this trial. I'm just watching
00:37:11
like everyone else. The last time I
00:37:12
weighed in on some Elon litigation, I
00:37:16
got deposed for six hours. Remember
00:37:18
that? Cuz they just assume that somehow
00:37:20
I know something,
00:37:21
>> right?
00:37:21
>> I've never talked to Elon about the
00:37:22
case. I don't know anything about it.
00:37:24
>> Yeah.
00:37:25
>> I'm going to see what happens like
00:37:26
everyone else. Now, one thing I I will
00:37:29
say having just read some of the
00:37:31
coverage is that apparently the company
00:37:35
at some point did offer Elon shares in
00:37:38
the company, but he thought that there
00:37:41
was something kind of icky about it. Do
00:37:43
you remember this that
00:37:44
>> Yes. Because at at one point I said on
00:37:46
our show when this dispute started
00:37:49
happening but before it became a court
00:37:50
case I said look if Open AAI at a
00:37:53
certain point decided they had the wrong
00:37:54
structure they should just gone and done
00:37:56
a make right with Elon and he should
00:37:59
have been a shareholder on the cap
00:38:00
table. What I didn't know is that
00:38:02
apparently they did try to do something
00:38:04
like that but Elon turned it down
00:38:06
because he did want the entity to remain
00:38:10
a charitable entity. In other
00:38:13
>> had a principled view of it according to
00:38:15
the reports and was like no we're trying
00:38:17
to save humanity and then you're giving
00:38:20
this keys to the kingdom to Microsoft
00:38:21
that's all come out
00:38:24
>> and I I also have not talked to Elon
00:38:26
about any of this but my guess is like
00:38:30
most of these things there'll be some
00:38:32
sort of settlement or something here but
00:38:34
maybe he takes it to the mat who knows
00:38:36
Judge Rogers who's doing this
00:38:38
61-year-old Obama appointee politics has
00:38:42
played a role. Saxs, they have had to
00:38:44
tell the jury like however you feel
00:38:45
about these individuals politically,
00:38:47
whatever, please put that aside. But of
00:38:49
note is that she oversaw the Epic Games
00:38:51
versus Apple trial over App Store
00:38:54
exclusively ruled in favor of Apple with
00:38:56
some caveats um that they don't have a
00:38:59
monopoly, etc., etc. So, this is going
00:39:02
to be a really interesting one. And I
00:39:04
think the worst case scenario is open AI
00:39:07
for for OpenAI is they have to unravel
00:39:09
this somehow and that would delay the
00:39:12
IPO. That would cause chaos in
00:39:14
shareholders and I guess the best case
00:39:16
is some sort of settlement. And if Elon
00:39:19
put the first 40 or $50 million in, he's
00:39:21
he's due 10 20 30% of the company after
00:39:24
dilution.
00:39:26
All right, let's keep moving through the
00:39:27
docket. Lots more to discuss and uh good
00:39:30
luck to everybody in their lawsuit and
00:39:31
those of you betting on market all-in
00:39:34
summit selling up fast. Our fifth
00:39:37
edition Los Angeles September 13th to
00:39:39
15th. Go to allin.com/events
00:39:41
and uh speakers are going to be top
00:39:43
tier. Apparently Freeberg is having this
00:39:47
as his major creative outlet. I heard
00:39:49
some back channel chimoff today that
00:39:51
he's going to be doing Broadway musical
00:39:55
uh illusionist. tap. I got a tap dancing
00:39:58
situation.
00:39:59
>> He's literally going fullon entertainer.
00:40:02
This is going to be vaudeville sachs
00:40:05
wrapped up. He's just going to take it
00:40:07
to a whole new level. Musical numbers
00:40:10
>> like Nathan Lane.
00:40:12
>> I think if you're coding that it's going
00:40:15
to be his big gay summit. Yes, it could
00:40:17
be a big gay summit.
00:40:19
>> Might be our last year in LA, guys.
00:40:22
>> Why?
00:40:23
>> Not might be.
00:40:24
>> Might be. Oh, everybody wants to go to
00:40:25
Vegas apparently.
00:40:28
Those bones, baby. Can you imagine
00:40:31
leaving the summit for lunch and going
00:40:33
and playing crabs? Jimoth, we get a
00:40:35
fresh. Yes, I can. Yes, I can imagine.
00:40:39
>> I got some bricks. Oh, I got some bricks
00:40:41
right here. Let's go. Yum, yum.
00:40:44
>> That way Sachs can come.
00:40:46
>> Yeah, Sax is like, I'm never setting
00:40:48
foot in California, but I will.
00:40:52
You know, we're doing a couple live
00:40:53
events. Are you coming to them?
00:40:54
>> Liquidity or something different?
00:40:56
>> Liquidity. And then there's the the
00:40:57
all-in summit happens in September.
00:40:59
>> Yeah, I'm going to do those, too.
00:41:00
>> All right. Big tech smashed their
00:41:02
earnings on Thursday. Google, Microsoft,
00:41:05
Amazon, and Meta all reported. I don't
00:41:07
know why they do this on the same night,
00:41:09
folks, but they do. And performance was
00:41:11
spectacular. It was great. However, the
00:41:14
capex announcements were really the
00:41:19
story here. Let me just cue this up and
00:41:23
show the chart.
00:41:25
$725 billion in capex guidance in 2026
00:41:30
from but four companies. Amazon,
00:41:32
Microsoft, Google, and Meta. Amazon
00:41:34
leading the pack with 200 billion, 190
00:41:37
billion each for Microsoft and Google,
00:41:38
145 billion for Meta. You add Grock, you
00:41:42
add OpenAI and some other players to
00:41:45
these plans. And we haven't heard from
00:41:46
the new Apple CEO yet, but he's going to
00:41:48
be taking over. And he's going to have
00:41:50
some plans here. I'm sure we are going
00:41:53
to see the large a trillion dollars a
00:41:55
trillion dollars in buildout over the
00:41:57
next year. I don't know if this is even
00:41:58
possible, but this is all being driven
00:42:02
by AI and cloud computing. Google Cloud,
00:42:06
which includes the Google Suite, that
00:42:08
grew 63% year-onear.
00:42:11
Let that number sink in. 63% on 20
00:42:14
billion in revenue. That's in a quarter.
00:42:16
Microsoft cloud, that includes Azure,
00:42:18
Windows Server, SQL Server, they bundled
00:42:20
some things together there to get the
00:42:22
number to go up. Uh that grew 30% on
00:42:25
34.7 billion in revenue. Amazon Web
00:42:28
Services, the original cloud, that grew
00:42:31
28% on 37.6 billion in revenue. That's a
00:42:34
bit of a pure play. Just counts Amazon's
00:42:37
web services. Obviously, these are all
00:42:39
moving to NeoClouds. These are all
00:42:41
serving AI jobs and tokens now. They
00:42:44
have a massive customer base and the
00:42:46
customers from the smallest startups all
00:42:48
the way to the biggest frontier models
00:42:49
cannot get enough compute and it is
00:42:52
going to the bottom line. But this is
00:42:54
shrinking Chimath cash flow massively.
00:42:57
These were free cash flow machines, the
00:43:00
largest money printing machines in the
00:43:02
history of humanity. But they are giving
00:43:05
up on free cash flow, stock buybacks and
00:43:07
dividends and the focus on those three
00:43:11
to invest in infrastructure. Amazon's
00:43:13
free cash flow down 97%
00:43:16
Google, Microsoft and Meta down 12, 12
00:43:18
and 8% respectively. your thoughts on
00:43:21
this free cash flow, the end of the free
00:43:23
cash flow deluge and the massive massive
00:43:27
investment we're seeing in capex.
00:43:30
I think we're seeing a very important
00:43:33
structural shift in the capital markets.
00:43:36
I think the last 20 or 30 years, well 20
00:43:39
years, it's been that the mag 7 just
00:43:42
kind of ran away with it. that these big
00:43:45
companies got bigger and bigger and it
00:43:47
absorbed all of these investment dollars
00:43:49
and the biggest reason was that it had
00:43:52
these very assetike business models,
00:43:55
right? You just built some more software
00:43:56
and it just has all this leverage and it
00:43:58
all just kind of worked except maybe for
00:44:00
Amazon cuz they needed physical
00:44:01
infrastructure for warehouses and
00:44:03
delivery and whatnot. But by and large
00:44:04
it was a very asset light investment
00:44:07
cycle. Now all of a sudden the pendulum
00:44:09
is swinging violently in the other
00:44:11
direction. And there's something that I
00:44:13
think people misunderstand which is as
00:44:15
it moves back to these asset heavy
00:44:18
infrastructure investments.
00:44:22
The hyperscalers are signing checks that
00:44:25
I mean I suspect their body can cash but
00:44:27
there's a world in which they can't.
00:44:29
I'll give you an example. You know when
00:44:31
Microsoft convinced the owners of three
00:44:33
Mile Island to turn their
00:44:37
>> nuclear site back on?
00:44:38
>> Yeah.
00:44:39
>> Do you know what their Ford purchase
00:44:41
agreement was? It was for more than 2x
00:44:43
the prevailing spot rate for energy.
00:44:45
More than 2x. The problem is that's not
00:44:47
for an enormous percentage of their
00:44:50
overall energy needs.
00:44:53
So if you play that out and you think
00:44:56
these five or six companies all of a
00:44:58
sudden are not just spending Jason 700
00:45:01
billion a year of capex which they are
00:45:05
but then from an operating cash flow
00:45:07
they're going to be spending 2x the
00:45:09
prevailing spot rate because they just
00:45:11
want guaranteed demand into the future.
00:45:15
Where's all this cash going to go? It's
00:45:18
not going to go
00:45:20
to the shareholder and it's not going to
00:45:22
stay on the balance sheet. These
00:45:24
companies will now get levered. They're
00:45:26
going to get highly sophisticated around
00:45:29
the financial engineering. They'll have
00:45:30
more debt. They'll have all kinds of
00:45:33
different vehicles and term loans and
00:45:35
revolvers and all of this stuff.
00:45:37
And so, they're going to look like this
00:45:39
big bulky industrial business in five
00:45:41
years. And I'm not sure that there's a
00:45:44
good valuation case to be made at that
00:45:46
point.
00:45:47
And so I think it may be simpler and
00:45:50
this is what I tweeted to just follow
00:45:52
the dollars like a trillion dollars a
00:45:55
year going out of the hyperscalers.
00:45:57
Where is it going? Just follow those
00:45:59
dollars and buy those companies because
00:46:00
those companies are already underpriced.
00:46:03
This is uh obviously reminiscent of
00:46:05
something we all experienced. Uh Nick,
00:46:07
can you pull up the Cisco chart I just
00:46:08
sent you and put it at max? uh we had a
00:46:11
massive buildout of the infrastructure
00:46:14
of the internet in the late 1990s and
00:46:17
into 2000. And what that caused was a
00:46:20
lot of aggressive companies to do
00:46:21
massive amounts of spending, a lot of
00:46:23
retail investors to embrace these stocks
00:46:25
like we're seeing with people trying to
00:46:28
get into these private companies and
00:46:30
saxs. Look at the 2000 peak of Cisco.
00:46:33
This is the most extraordinary chart
00:46:35
ever. It took them 25 years to get back
00:46:37
to that peak and uh they had a lost two
00:46:41
decades and we had a massive amount of
00:46:44
fiber that wound up getting bought. We
00:46:45
talked about that a couple years ago on
00:46:47
the program. But there's something for
00:46:48
you to build off of here when you look
00:46:50
at this massive infrastructure. You
00:46:52
think it's going to be Cisco systems all
00:46:54
over again, World Warcom, etc.?
00:46:55
>> No, I really don't. The issue we had in
00:46:58
2000 was dark fiber. You had all this
00:47:00
infrastructure being built out and it
00:47:02
wasn't being used. There's no dark GPUs
00:47:05
today as you know Brad Gersonner likes
00:47:07
to say. So what's driving the capex now
00:47:11
is the voracious demand for compute for
00:47:15
tokens and the demand is now pulling
00:47:19
forward this additional um investment in
00:47:22
infrastructure. So I think what's
00:47:25
happened here is that the bull thesis
00:47:26
for AI just got validated in a single
00:47:30
afternoon. I mean again you got
00:47:31
Microsoft Azure, Google Cloud, Amazon
00:47:34
AWS, Meta, they're all basically
00:47:37
exceeding expectations, exceeding
00:47:39
guidance in terms of where their cloud
00:47:42
revenue would be and therefore how much
00:47:44
they're going to reinvest in capex this
00:47:47
year. I think we were supposed to have
00:47:48
660 billion of hyperscaler capex up from
00:47:52
350 last year. I think there's now the
00:47:54
new estimate is it's going to be over
00:47:55
700. So this is you know again it's more
00:47:58
than 2% of GDP. This is a huge tailwind
00:48:00
to GDP. There's another article saying
00:48:02
that I think in the last quarter
00:48:05
AI was 75% of GDP growth. And by the
00:48:08
way, this is just the capex part. This
00:48:10
is the physical infrastructure. This is
00:48:12
not the economic impact of the tokens
00:48:14
that are generated inside the token
00:48:16
factory. This is the building of the
00:48:18
factories. How do those tokens get used?
00:48:20
like we're seeing they're being used not
00:48:23
just to do research or to answer
00:48:25
questions but to create code. And so
00:48:29
we're seeing this explosion of
00:48:31
productivity in software development.
00:48:33
And we're seeing an explosion of bespoke
00:48:36
software being created and that's going
00:48:38
to accelerate every part of the economy.
00:48:40
Every business that now wants to get
00:48:42
code will be able to get code for the
00:48:44
first time. Before they couldn't even
00:48:45
hire the engineers, they needed to
00:48:47
generate it. Now they will be able to.
00:48:49
So that is a huge unlock of productivity
00:48:52
across the economy. Then you're getting
00:48:54
into these new use cases like the the
00:48:56
co-working use cases and agents, right?
00:48:59
So the the workflow automations that are
00:49:01
happening, it's still early. I don't
00:49:04
believe that this is going to replace
00:49:05
humans. We had that um in the past week,
00:49:06
we had that crazy case of an agent
00:49:09
deleting a production database in 9
00:49:11
seconds because because of a bug. Look,
00:49:13
what that said to me is that
00:49:16
>> it's not that agents aren't valuable.
00:49:18
They are valuable, but they have to be
00:49:19
supervised. You know, this idea that
00:49:21
you're just going to be able to like
00:49:22
automate all the jobs away. It is a
00:49:24
massive amount of handwaving over the
00:49:27
real technical problems and issues. The
00:49:29
agents have to be supervised. Someone
00:49:31
has to be accountable. It's not going to
00:49:33
be the CEO. The CEO doesn't want to be
00:49:35
accountable for thousands of agents. You
00:49:38
need people
00:49:38
>> despite what Jack had block said.
00:49:40
>> Yeah. Thousand direct reports is a great
00:49:43
like goal, but it's not realistic. Yeah,
00:49:46
>> you need IT people who are savvy who can
00:49:49
supervise this and make sure it's
00:49:50
working. They have to be accountable to
00:49:51
the CEO. Someone has to drive the
00:49:53
productivity. It's like Bology always
00:49:56
said, AI is not end to end is middle to
00:49:58
middle. You have to have someone to do
00:49:59
the prompting and you have to have
00:50:00
someone to do the validating and I would
00:50:02
add the supervision and accountability.
00:50:04
So anyway, the larger point though is
00:50:06
I'm speaking to the fact that I don't
00:50:08
think there's going to be this huge job
00:50:10
loss associated with this productivity
00:50:12
boom that we're going to get. And in
00:50:13
fact, I think what's actually happening
00:50:15
now is that AI is becoming synonymous
00:50:18
with the American economy. I mean, the
00:50:20
fact that it's generating 75% of GDP,
00:50:23
you have this capex explosion, this
00:50:26
energy explosion that feeds it, and
00:50:28
again, just the beginning of the
00:50:31
applications that are being unleashed by
00:50:33
these new token factories. I think it's
00:50:35
all a very, very positive thing. and all
00:50:38
these doomers who are trying to throw a
00:50:40
wet blanket on it or constantly scaring
00:50:43
the daylights out of people. I mean,
00:50:45
what do they want the American economy
00:50:47
to do? Just to stop I mean, they just
00:50:49
don't want any progress. I mean, like
00:50:51
again, you know, when you talk about
00:50:52
stopping AI or halting AI progress? What
00:50:55
you're really doing is stopping the
00:50:57
American economy now. You're basically
00:50:59
saying you don't want economic growth.
00:51:01
AI is now synonymous with the growth of
00:51:04
the American economy. And if there's no
00:51:06
economic growth, there's not gonna be
00:51:07
money to pay for all the social
00:51:08
programs. There's not gonna be money to
00:51:09
pay down the national debt. There's not
00:51:11
gonna be money to basically build up our
00:51:13
national defense. All these things we
00:51:14
want to spend money on. We have to have
00:51:16
a vibrant economy. And that is now
00:51:19
synonymous with AI. So I know that AI
00:51:21
may not be popular. I see those polls.
00:51:23
But having a strong economy is popular.
00:51:26
And I believe that those things are now
00:51:28
synonymous. It's almost like there was
00:51:30
some architect or ZAR who set up the
00:51:33
chessboard in the first year of this to
00:51:35
make sure that it was ultra competitive.
00:51:38
>> President Trump set the table on this.
00:51:39
>> Absolutely. With some good advice, I
00:51:41
think. Maybe.
00:51:42
>> Freeberg, your thoughts?
00:51:43
>> It's always good to have good advisors.
00:51:45
>> Always good to have good advisors.
00:51:46
Absolutely. Absolutely.
00:51:47
>> No, but look, I've said it before. The
00:51:49
president just wants America to win.
00:51:50
>> Literally, there are people who if we
00:51:52
were looking at this, you know, I don't
00:51:54
know, a hundred years ago, it'd be like
00:51:56
people were like, "Yeah, you know what?
00:51:57
we shouldn't build the highway system or
00:51:58
we half built the highway system. Let's
00:52:00
stop let's stop building the highways.
00:52:02
>> No, the highway system was funded by the
00:52:04
federal government. There was no
00:52:05
competition. It was the most expensive
00:52:07
on a on a inflationadjusted basis. I
00:52:10
think it was the most expensive project
00:52:11
in US history.
00:52:13
>> Yeah. And the railroads before that like
00:52:15
you can't stop these things. They have
00:52:17
to keep going. It's interesting point
00:52:20
you know there is so much demand for the
00:52:22
resource of tokens of intelligence
00:52:24
freeberg and it's quite different than
00:52:26
the fiber situation as Sax correctly
00:52:28
points out where we built all this but
00:52:31
we didn't actually have an application
00:52:32
here the application is pretty um pretty
00:52:36
wellnown and you've got a large number
00:52:38
of people in businesses who are trying
00:52:40
to vibe code their way to success trying
00:52:43
to push this stuff and we had an
00:52:45
interesting story referenced earlier in
00:52:47
the show
00:52:48
where
00:52:50
uh Claude ate somebody's homework. This
00:52:52
is the nightmare of all nightmares.
00:52:55
Somebody was vibe coding. Uh it was the
00:52:57
founder of Pocket OS. Apparently, they
00:52:59
make software for rental car companies.
00:53:00
He was using Opus 4.6 through Cursor's
00:53:03
AI platform, their coding platform, and
00:53:07
uh you know, which is like the most
00:53:09
expensive tier. Uh and he said he
00:53:12
configured it with enough safety rules,
00:53:13
but the agent was working on a routine
00:53:16
task. They saw some sort of credentiing
00:53:18
mismatch and they decided to fix the
00:53:20
mismatch by deleting a railway volume
00:53:22
without user confirmation and uh they
00:53:25
pushed the code from a repo to a live
00:53:27
app and they deleted everything
00:53:28
including the backups. Literally a scene
00:53:32
from Silicon Valley's HBO clip of Son of
00:53:36
Anton. Hilarious.
00:53:37
>> You gave your AI permission to overwrite
00:53:39
code in the internal file system. Were
00:53:42
you going to tell me about this? No, I
00:53:44
thought that was the company policy
00:53:46
these days.
00:53:47
>> Okay, well, your AI just failed
00:53:50
epically.
00:53:51
>> That's unclear.
00:53:53
>> It's possible the Son of Anton decided
00:53:55
that the most efficient way to get rid
00:53:56
of all the bugs was to get rid of all
00:53:58
the software, which is technically and
00:54:01
statistically correct. But artificial
00:54:03
neural nets are sort of a black box. So,
00:54:05
we'll never know for sure.
00:54:06
>> How did they get that so right, Zach?
00:54:08
Five or six years ago, art and neural
00:54:10
networks are a black box. So, I guess
00:54:11
we'll never know. But technically, it
00:54:14
was correct. Freeberg, when you blow up
00:54:16
a hollow system with your vibe coding,
00:54:19
which you were absolutely showing off in
00:54:21
front of Jensen a couple of weeks ago
00:54:23
about how much code you're pushing, who
00:54:24
are you going to blame? You going to
00:54:26
take responsibility yourself? Are you
00:54:27
going to blame Cla Claude or Kurser? Who
00:54:30
are you going to blame when you blow up
00:54:32
the entire stack over at Ohio?
00:54:36
>> Who you blame?
00:54:37
>> Yeah, I'll blame Dario.
00:54:38
>> You blame Dario. Okay, that's what I
00:54:40
thought. That's a correct answer.
00:54:41
Correct answer. Blame Daario. He's the
00:54:42
one who says it's a doomsday machine.
00:54:44
Uh, come on the prodio.
00:54:47
17th invitio.
00:54:50
>> I mean, I've invited the guy like 17
00:54:52
times. He is totally going to me. He
00:54:54
wants nothing to do with this podcast.
00:54:57
>> Actually, let me speak to that. So, I
00:54:59
think I think that um there's maybe a
00:55:02
misperception that this error occurred
00:55:05
because of, you know, quote unquote AI
00:55:07
scheming,
00:55:09
>> like kind of in that video that the AI
00:55:12
decided that the best way to get rid of
00:55:13
bugs is to basically eliminate the
00:55:14
codebase. This is kind of like the, you
00:55:16
know, AI is going to turn the world into
00:55:18
paper clips type thing where somehow
00:55:19
it'll like miss scheme. That's not
00:55:21
really what happened here. This is a
00:55:22
case of just a of old-fashioned bugs
00:55:26
occurring at an edge case. You know,
00:55:28
you've got the fact that this API was
00:55:30
not designed for permissioned usage.
00:55:34
You've got the fact that a credential
00:55:36
was left kind of lying around. Probably
00:55:38
it should not be. There's kind of like a
00:55:40
perfect storm that caused the AI to do
00:55:42
something or the agent to do something
00:55:43
that didn't quite understand it was what
00:55:44
it was doing. I think that if there is a
00:55:48
systemic problem here rather than just
00:55:50
kind of a like a random edge case is
00:55:53
that AI still doesn't know what it
00:55:58
doesn't know. You know, like a human
00:56:00
would stop before deleting a production
00:56:03
database and just say, "Oh, I'm about to
00:56:04
do something like really serious, really
00:56:06
destructive. Am I sure I want to do
00:56:08
this?" You know, and a human would have
00:56:10
stopped and said, "Oh, wait a second.
00:56:11
like I need to be more confident in what
00:56:13
I'm doing before I take that action. And
00:56:16
AI still has this issue where again it
00:56:18
can be kind of overcon. This is where
00:56:19
like the hallucinations come from is it
00:56:22
doesn't know when it should have a low
00:56:24
confidence in its output, right? But
00:56:26
this is why it has to be supervised. You
00:56:29
know, the longer the time horizon for a
00:56:31
task, the more likely it is to go off
00:56:34
the rails.
00:56:35
>> And a drift. Exactly. And this is why I
00:56:38
think people are starting to realize
00:56:40
that this idea of eliminating all
00:56:41
software developers was the peak of
00:56:44
inflated expectations. Yes.
00:56:45
>> Right. There was actually a really good
00:56:48
tweet on this by Aaron Levy who's got
00:56:50
the right take on this. Aaron retweeted
00:56:53
Matthew Glacius who sort of sardonically
00:56:56
tweeted that 5 months in I think I've
00:56:59
decided I don't want to vibe code. I
00:57:01
want professionally managed software
00:57:02
companies to use AI coding assistants to
00:57:05
make more better, cheaper software
00:57:07
products that they sell to me for money.
00:57:08
>> Just lower your prices. Don't make me
00:57:10
vibe code is the translation.
00:57:13
>> Yeah. I mean, I think like rare win for
00:57:15
for Madaglacius there. Anyway, Aaron
00:57:17
Levy then says
00:57:19
Agent Coding is a huge win for software
00:57:21
developers that want to get more done
00:57:23
and it's fantastic for anyone curious to
00:57:26
learn how to start coding. What it's
00:57:28
less great for is casually building
00:57:30
complex software that you have to
00:57:32
maintain on an ongoing basis and take
00:57:33
all the risk for upgrades, maintenance,
00:57:36
keeping up to date with latest security
00:57:37
issues, you know, the bugs, cyber, those
00:57:40
are taxes on most knowledge workers who
00:57:42
aren't familiar with the system.
00:57:45
>> It's not a tax. It's a huge risk.
00:57:47
>> Yes, it's a risk has to be managed if
00:57:49
you don't. People will get fired because
00:57:52
there will be some public companies
00:57:53
where some goofball tries to vibe code
00:57:55
their way out of something and they're
00:57:56
going to torch the enterprise value.
00:57:58
It's going to be glorious to watch
00:58:00
because we're all going to laugh and
00:58:01
realize that was stupid and should never
00:58:02
have happened in the first place.
00:58:04
>> Yeah. I mean, it's there is a chance
00:58:07
that this improves to the point passes
00:58:10
trial of disillusionment and becomes
00:58:11
super productive and you'll be able to
00:58:13
get an agent to do reasonable things
00:58:15
without deleting your data set. But we
00:58:17
have a way to go. Here is your, you
00:58:20
know, this is the tech adoption chart.
00:58:22
Basically, you got a technology gets
00:58:23
triggered. You have the trial, you have
00:58:24
this peak of inflated expectations. You
00:58:26
go into the trial of disillusionment and
00:58:27
then the slope of enlightenment invest
00:58:29
and eventually it becomes deer and it's
00:58:32
an opportunity. Hey, uh, Freedberg,
00:58:36
you have become reddit tide curious. You
00:58:41
have and also
00:58:43
>> tell me tell me about rea cuz I want it.
00:58:46
I want to get on it. M
00:58:47
>> I want to use it and I need you to tell
00:58:50
Nat that it's okay for me to take it.
00:58:52
>> I I have a friend who has some advice as
00:58:54
well.
00:58:55
>> Freeberg, the coverage is coming out of
00:58:57
this phase three clinical trial data
00:58:59
release that Lily put out last month. So
00:59:01
everyone's going crazy over the data
00:59:05
which continues to show pretty amazing
00:59:08
results. So unlike trazepatide which is
00:59:11
kind of Lily's main
00:59:15
product today, it's a which is a dual
00:59:16
agonist. It's got two peptides in it
00:59:18
that that bind to different receptors,
00:59:20
the GLP1, the GIP receptor. This other
00:59:22
one now also binds to glucagon, which is
00:59:24
a third receptor. And that glucagon
00:59:27
receptor binding peptide causes the
00:59:30
cells to increase their metabolism,
00:59:31
which actually accelerates fat energy
00:59:34
consumption
00:59:36
over what would typically be muscle
00:59:38
energy consumption. It's more likely to
00:59:40
burn up fat early on, which causes more
00:59:44
quick fat loss, but also reduces muscle
00:59:47
loss. And some of the other data that's
00:59:50
now coming out shows non-HDL cholesterol
00:59:53
down 27%, triglycerides down 41%.
00:59:57
>> Liver fat down 80% to
01:00:00
>> 80% reduction of liver fat. A1C drops
01:00:04
from 7.9% to 6% in 40 weeks, which is
01:00:08
amazing, by the way. If you're diabetic
01:00:09
and your A1C drops that much in a couple
01:00:12
of months, it's literally a life-saving
01:00:14
product. The average user in this phase
01:00:17
3 trial saw their weight decline from
01:00:20
214 pounds. They lost 37 pounds. That's
01:00:23
compared to six pounds on placebo
01:00:26
in 40 weeks. And you know, modest side
01:00:29
effects. 20% people felt more nauseous
01:00:32
than the people that were on the
01:00:33
placebo. There's a lot of other separate
01:00:36
studies that are being done now that are
01:00:37
showing significant reductions in
01:00:39
inflammatory signaling molecules. So
01:00:42
systemic signaling of like hey cells are
01:00:46
in distress triggers this kind of
01:00:49
inflammatory process that can have a lot
01:00:51
of other damage to your body can
01:00:53
accelerate aging. And so one of the
01:00:55
other conversations is that retride
01:00:57
might actually be kind of a deaging drug
01:01:00
as well.
01:01:02
Hercules, Hercules, Hercules,
01:01:05
you know, and a lot of the studies, by
01:01:06
the way, are done on the the the very
01:01:08
high dose, 12 milligram dose, but you
01:01:10
could probably get this thing dosed down
01:01:11
to 2 milligrams and still see a lot of
01:01:13
the anti-inflammatory
01:01:14
maintenance and other benefits. I'm no
01:01:16
doctor, but people are going nuts over
01:01:19
this being more widely useful than just
01:01:22
for clinical obesity or type when the
01:01:24
FDA when's the projected date for
01:01:27
>> 2027. Mid 27.
01:01:28
>> That's what they're saying. Could happen
01:01:30
sooner. I mean, the data is in the, you
01:01:32
know, the FDA will take their time to
01:01:36
evaluate it, but I think given the way
01:01:37
this is all looking,
01:01:39
>> could happen sooner, could happen
01:01:40
sometime later this year. Swim Chimath
01:01:43
Swim said it's incredible and that uh
01:01:47
it's living up to the hype in their
01:01:49
experience.
01:01:50
>> Who?
01:01:51
>> Swim.
01:01:52
>> What is that? What is that?
01:01:53
>> Someone who isn't me. Swim.
01:01:55
>> Oh,
01:01:55
>> this is a Reddit term. Someone who isn't
01:01:57
me said who has a guy swim has a guy and
01:02:02
has cycled on reddatride and does
01:02:06
push-ups and says muscle gain has been
01:02:09
spectacular
01:02:10
no muscle loss and a lowering of fat.
01:02:14
>> If you go on X and you just search up
01:02:16
rea
01:02:17
>> Mhm.
01:02:19
>> it's like incredible. You see these like
01:02:21
65 year old guys that go from a dadbod
01:02:24
to looking like an incredibly ripped
01:02:27
athlete in weeks. And and I
01:02:31
I mean I'm shocked. And then for me I
01:02:34
don't need that help per se, but my
01:02:36
liver health is important to me. My
01:02:37
cardiac health because I'm South Asian
01:02:39
and it just looks like a wonder drug. I
01:02:41
can't wait. When you starve your body,
01:02:43
when you turn off the the appetite,
01:02:45
which is the GLP-1 agonist function,
01:02:47
normally your body goes into this kind
01:02:49
of mode of starvation and you have this
01:02:52
process by which your body tries to
01:02:53
generate energy from your existing
01:02:55
cells. And because muscle is much denser
01:02:59
than fat, you can have a favoring of
01:03:01
muscle tissue being kind of broken up
01:03:03
over fat tissue. But what this new
01:03:05
agonist, this glucagon agonist that they
01:03:07
put into this neutrutide
01:03:10
is it favors fat burning over muscle
01:03:12
burning. And so that actually can drive
01:03:15
short-term use at low dose for people to
01:03:18
cut weight and maintain muscle and get
01:03:20
ripped. And so that's why a lot of
01:03:21
people in the kind of fitness community
01:03:23
are talking about, hey, I want to get
01:03:24
access to this and get on it for a
01:03:26
while. So you'll see a lot more hype
01:03:27
probably in that community as well as
01:03:30
the all the health effects. It just
01:03:32
feels like we're about to have an
01:03:34
absolute avalanche of peptides to choose
01:03:36
from.
01:03:36
>> On November of 2025, Lily cut a deal
01:03:39
with the Trump administration. I saw
01:03:40
this to drop the price on Drespatide
01:03:43
pretty significantly. I think it's like
01:03:44
50 bucks on Medicare.
01:03:45
>> 50 bucks from Medicare. Yeah.
01:03:47
>> Yeah. Which is a pretty cheap price
01:03:48
point, but it starts to make sense as
01:03:50
you think about the portfolio of Lily
01:03:51
products. You get Tzepide for 50 bucks,
01:03:54
but if you want to upgrade, get the
01:03:56
Retatride. That's the high premium
01:03:57
product and that's where they're going
01:03:58
to start the Mercedes to the Honda. I'm
01:04:01
sure if I'm Lily and I'm sitting there
01:04:02
and I'm looking at this data coming out,
01:04:03
I'm like, "My god, people will pay for
01:04:05
this and that starts to become sort of
01:04:07
like the upgrade to the BMW or the Model
01:04:09
S plat if you will."
01:04:10
>> Yeah. The Trappetide is like the one
01:04:12
bedroomedroom messy bed hotel room
01:04:15
>> and the other one's the sweetide
01:04:18
is like the twobedroom suite.
01:04:20
>> Well, you can also, by the way, you guys
01:04:22
know I'm a spokesman for Row. They also
01:04:26
have the Waggoi pill uh row.co/twist
01:04:28
cotwist uh to get your
01:04:31
>> Wait, are you a paid talking about Are
01:04:34
you a paid What are you talking about?
01:04:36
We're putting Charles Barkley and
01:04:38
>> we're not having sales team
01:04:42
and then you come over on Allin and you
01:04:44
start promoting.
01:04:44
>> No, no, no, no. Trust me, we'll get one
01:04:46
of those as well. We'll get a row.
01:04:48
Sponsorship here.
01:04:49
>> What was the ro pill that you had me
01:04:50
get? What was it called?
01:04:51
>> Oh, sparks. Sparks. Sparks. Did you take
01:04:53
it?
01:04:54
>> I have taken it and now
01:04:56
>> Amore please. No, not
01:04:58
>> maybe just a half of a lousy.
01:05:00
>> I want to hear the story. I want to hear
01:05:01
the story. Go.
01:05:02
>> It's so out of control.
01:05:04
>> I told what I told you.
01:05:06
>> So then what happens is Nat and I are
01:05:08
like, you can't you can't just randomly
01:05:09
use it. It's scheduled. We discuss it.
01:05:11
We put it on the calendar.
01:05:12
>> You need a plan.
01:05:13
>> You need a plan.
01:05:14
>> You need a plan. Can't go in
01:05:16
>> because otherwise otherwise it's too
01:05:17
much. You just can't randomly take it.
01:05:19
>> What do you mean?
01:05:20
>> It's going to be a sesh.
01:05:22
>> It's a whole thing, man. It's like I
01:05:24
don't have the energy for that.
01:05:25
>> It's an extended session. You you have
01:05:26
to be well rested. This don't do this at
01:05:29
1:00 a.m. This is like a 10 p.m. This is
01:05:31
like a No, this is more like a This is
01:05:33
more like on vacation, you know, like
01:05:35
>> 10:00 a.m. to 12:00 p.m., you know, to
01:05:37
noon, you know, you got to really
01:05:40
>> you got to really schedule it.
01:05:42
>> Schedule it because kids around
01:05:44
>> otherwise it's got to be empty.
01:05:46
>> Otherwise, it's unfair to her and it's
01:05:48
just a li it's a lie. Put it out.
01:05:50
>> It's a It's a big commitment literally.
01:05:52
>> You You look embarrassed, Jim. Do you
01:05:54
feel embarrassed talking about it? It's
01:05:56
just a lot, man. It's like It's a lot to
01:05:57
handle. It's a lot.
01:05:59
>> It's If you want to get the extra 20% in
01:06:02
your performance,
01:06:03
>> it's a lot, bro.
01:06:04
>> It's a lot. It's basically over time.
01:06:07
>> What happened? What happened was I was
01:06:08
like, "Oh, what is this thing?" Jason's
01:06:10
like, "Dude, you must get it. You must
01:06:11
get it." So, we got it. We tried it and
01:06:14
>> we were like, "What the was that?" And
01:06:17
so, then I've been trying to bleed the
01:06:18
pills out. So, I gave some to Stant
01:06:20
Tang. I'm like, "Stanley, you try it."
01:06:22
Literally, he's dealing them like cards
01:06:25
>> but when we're having poker dinner, I'm
01:06:27
like, does anybody want to try these
01:06:28
things? But these what is it? Rose
01:06:30
sparks. Is that rose sparks? Shout out
01:06:31
to my friends out. All right, let's keep
01:06:33
moving here. Freedberg, guys. Friedberg
01:06:36
had his own personal Super Bowl. You see
01:06:38
me getting ready for Nick's playoff
01:06:40
season. I get my courtside. Freeberg had
01:06:43
the equivalence acts. He went to the
01:06:45
Supreme Court in order to hear them talk
01:06:49
about chemicals. This was a big deal for
01:06:52
him. The Supreme Court coming together.
01:06:54
Did you wear it? Yes. The Monsanto trial
01:06:57
happened in the Supreme Court and he
01:06:58
went he got courtside. He went to the
01:07:01
Supreme Court and listened in the
01:07:02
building. Have you guys ever seen a live
01:07:04
Supreme Court hearing?
01:07:05
>> No. I'd love to.
01:07:06
>> I'd love to. Tax, have you been?
01:07:08
>> No, I haven't actually.
01:07:09
>> I mean, honest honestly, I think it was
01:07:11
one of the most amazing experiences I've
01:07:13
ever had. There was a massive protest
01:07:14
out front. We went through the
01:07:16
marshall's office to get in. And that
01:07:18
building, you walk in, it's like sacred.
01:07:20
It's all marble. It's you're not allowed
01:07:23
to talk. You have to be super quiet when
01:07:24
you're in the building. Like they keep
01:07:25
going like you're in some quiet library.
01:07:28
It's like people treat it with this
01:07:30
level of kind of sanctity that that and
01:07:32
respect. And they're like, there is no
01:07:34
politics here. There is no [ __ ]
01:07:36
There is no freedom of speech. This is
01:07:38
the court. When you come into this
01:07:40
court, the justices tell you how you
01:07:43
will speak, how you will behave, what
01:07:44
you will do, and you will not speak
01:07:46
unless spoken to. You put all your stuff
01:07:48
in a locker, you go up the stairs, you
01:07:50
go into the the courtroom. And the
01:07:52
courtroom, it's just so amazing being in
01:07:54
there. They have this amazing marble
01:07:55
freeze above the justices that has some
01:07:57
of the great people of human history,
01:07:59
Moses and these kind of amazing
01:08:01
historical figures. And then below them
01:08:03
are the nine justices and the court
01:08:05
case. If you guys haven't watched the
01:08:07
case, you can listen to them, I think,
01:08:08
online. on you. Is it worth listening
01:08:10
to?
01:08:10
>> Hold on. Wait, wait, wait. I have a
01:08:11
question. So,
01:08:12
>> does Robert sit in the middle cuz he's a
01:08:14
chief? Yes.
01:08:15
>> And then do all of the right justices
01:08:17
sit on the right?
01:08:18
>> No, they're mixed. So, they're I think I
01:08:20
don't know I don't know the exact
01:08:21
seating, but they're mixed in terms on
01:08:24
appointments of the court. Is
01:08:25
>> that right? I think that's right. And
01:08:27
then so yeah, that's right. And then it
01:08:29
kind of goes out from the middle with
01:08:30
Roberts in the middle. Roberts
01:08:32
occasionally will name the justices and
01:08:35
say, "Hey, do you have a question? Do
01:08:36
you have a question?" if no one's
01:08:37
talking, but otherwise the justices will
01:08:39
jump in with their questions when they
01:08:40
want and they'll ask. Now, honest to
01:08:42
God, watching this is like watching
01:08:44
LeBron James play basketball. These
01:08:46
lawyers are so mind-blowingly impressive
01:08:50
on both sides that you would just like
01:08:52
sit there and I was like in awe. It was
01:08:54
so I I felt like my energy was
01:08:57
completely sapped from me at the end of
01:08:58
this process because you were just so
01:09:00
engaged and so caught into the way that
01:09:02
these guys are thinking and talking.
01:09:04
>> Did you take a rose sparks? Did you take
01:09:05
a rose sparks when you were there?
01:09:06
>> No. And if you're familiar if you're
01:09:08
familiar enough with the case or the
01:09:10
case history or the law that's being
01:09:12
debated because again when when you get
01:09:13
to the Supreme Court, you never debate
01:09:14
the case. What you're debating is the
01:09:17
legal interpretation of the the
01:09:19
decisions that were made on the case.
01:09:21
And so is this constitutional? How do
01:09:23
you interpret this particular act, this
01:09:25
law, this federal law? What's the right
01:09:27
way to think about it? So you don't
01:09:28
actually talk about the case. You talk
01:09:29
about the interpretation of American
01:09:31
law,
01:09:33
of our laws, of our constitution, of the
01:09:36
global.
01:09:36
>> You're saying the facts have already
01:09:38
been determined.
01:09:38
>> That's right.
01:09:39
>> Right. At a lower court, there's
01:09:40
questions of fact and questions of law.
01:09:42
>> The facts have already been determined
01:09:44
by the lower court. It's just Supreme
01:09:45
Court is ruling on questions of law.
01:09:47
>> That's right. And so they have a full
01:09:49
briefing with the full history of the
01:09:50
case. And remember, they only hear two
01:09:52
cases a day. So they're one hour each
01:09:54
for each hearing.
01:09:55
>> So you go in and they only do it Monday,
01:09:56
Tuesday, Wednesday on the last two
01:09:58
weeks. and they only hear cases from
01:10:00
October to April. There's only a handful
01:10:02
of cases that are selected.
01:10:03
>> Wow. So you're really on a shock clock
01:10:04
then to make
01:10:05
>> you're on a shot clock and you only have
01:10:06
and it's 30 minutes aside and then the
01:10:08
justices will ask question.
01:10:09
>> So this was Monsanto and Roundup, right?
01:10:11
So what was the law that was being
01:10:13
debated
01:10:14
>> for yours? The regulatory body, the EPA
01:10:18
sets the label for pesticides. Does this
01:10:20
cause cancer or not? What are the
01:10:22
warnings? This can be damaging for birth
01:10:23
defects, pregnancy, all the things that
01:10:25
we're all used to seeing on labels. when
01:10:27
you buy a product, a chemical product
01:10:29
and the EPA and their regulatory
01:10:31
authority determined that Roundup does
01:10:34
not cause cancer. When you sell a
01:10:36
pesticide, you first have to register it
01:10:38
with the EPA, get it approved, and then
01:10:39
the EPA gives you a label. And the label
01:10:41
is written by the EPA. It says exactly
01:10:43
what you're supposed to say. And in this
01:10:45
case, it said all this stuff doesn't say
01:10:47
cancer because they determined it does
01:10:49
not cause cancer. And I'm not going to
01:10:51
debate whether or not it causes cancer,
01:10:52
but that's the case that was made is
01:10:54
that the EPA is the regulatory body
01:10:56
under a federal act called FIFRA,
01:10:59
fungicide, insecttoide or denicide act.
01:11:01
And that's where the EPA is given their
01:11:03
regulatory authority to put the label on
01:11:06
these products. And all of the cases
01:11:08
that have been lost have been state
01:11:11
failure to warn cases. To date, Bayer,
01:11:14
which now owns Monsanto, has paid out
01:11:16
$10 billion in these lawsuits, and they
01:11:19
have reserved 10 billion on their
01:11:21
balance sheet. They have 90,000 cases
01:11:22
still outstanding in the courts. 90,000.
01:11:25
>> Wow.
01:11:25
>> And so, this one case got kind of
01:11:27
appealed up to the Supreme Court. Last
01:11:30
year, the White House solicitor general,
01:11:32
and if the solicitor general steps up
01:11:33
and asks the Supreme Court to take a
01:11:35
case, it's more likely the case gets
01:11:37
taken. So, the White House said, "Please
01:11:38
take this case. We need to have federal
01:11:41
preeemption, meaning the federal
01:11:42
government has the right to set the
01:11:44
label because all of the cases that have
01:11:46
been lost and that are being adjudicated
01:11:49
are in state courts where the state has
01:11:51
a law like in California called a
01:11:53
failure to warn law, which means if a
01:11:55
manufacturer knows that a product
01:11:57
carries a risk, you have to warn the
01:11:59
consumer. And so the the lawyers have
01:12:02
been arguing that Monsanto or Bayer knew
01:12:04
that this product caused cancer and
01:12:06
didn't warn the consumer. And they've
01:12:08
been winning cases. they've been losing
01:12:09
cases, but they've won enough cases that
01:12:12
this has now become a multi-dea billion
01:12:13
dollar problem. And so the argument is
01:12:16
that the EPA says it doesn't cause
01:12:18
cancer and they have federal
01:12:19
preeemption. So the EPA has the right to
01:12:21
determine. So that's the one argument.
01:12:24
But then when the other attorney came
01:12:27
up, this guy was like literally like
01:12:28
watching LeBron James. And so going in,
01:12:30
we're like, "Oh, 63 Bayer's going to
01:12:31
win." And then the other guy comes up
01:12:33
and he was like, "Well, hey, you guys
01:12:36
overturned the Chevron doctrine last
01:12:37
year. You guys remember that case?
01:12:38
>> Yeah. Where basically when the Chevron
01:12:40
doctrine got overturned, it basically
01:12:42
said that no longer
01:12:44
>> does the federal agency get to decide it
01:12:46
has to be a direct reading of the law.
01:12:50
>> Duh. So now, so he's saying like the
01:12:52
states should have a right to read the
01:12:55
law themselves. They shouldn't have to
01:12:57
just defer to the EPA. And that's what
01:12:59
this will come down to. So at the end of
01:13:00
it, we were like, "Oh my god, this could
01:13:02
be a 50/50 coin flip, 54 either way."
01:13:05
And going into it, we were kind of like
01:13:06
trying to say, "Hey, maybe this could be
01:13:08
63." So honestly, the whole experience
01:13:10
was incredible. The case is interesting.
01:13:12
>> These are very complicated matters. How
01:13:13
are these people able to make a
01:13:16
wholesome argument in like one side gets
01:13:19
30 minutes, the other side gets 30
01:13:20
minutes, there's a little Q&A, and then
01:13:21
you're done in an hour.
01:13:22
>> There's this whole art and science and
01:13:24
Sax, you're probably familiar with this
01:13:26
on how do you distill down a Supreme
01:13:27
Court case in the briefing dock? Like
01:13:29
what is it you're petitioning around the
01:13:31
court? and you try and distill it down
01:13:32
to the exact legal interpretation you
01:13:35
want the judges to rule on, not all the
01:13:37
other [ __ ]
01:13:38
>> And this is oral arguments. Yes.
01:13:39
>> Oral arguments, just a discussion. And
01:13:41
then the judges jump in and all they're
01:13:42
doing is asking the lawyer questions,
01:13:44
one lawyer at a time, the one side and
01:13:46
then the other side. And by the way, the
01:13:47
solicitor general came up in the middle
01:13:49
and kind of made a few comments and they
01:13:50
asked her some questions from the White
01:13:52
House and she sat down and then the two
01:13:54
sides kind of went back and forth and
01:13:55
and they just it's like 30 minutes Q&A
01:13:57
each on that one specific legal question
01:14:00
and Katanji Brown Jackson said, "But
01:14:03
what if after the EPA issued the label,
01:14:05
they found out information that it does
01:14:07
cause cancer? Shouldn't they update the
01:14:09
label?" And he's saying, "Well, no,
01:14:10
they're not allowed to. They can only
01:14:11
issue the label the EPA says." And he
01:14:13
says also and it's it's a criminal case
01:14:16
if they find out that it does cause
01:14:17
cancer and they don't report it to the
01:14:19
EPA and then she's saying well what if
01:14:21
the EPA doesn't act and shouldn't the
01:14:23
states have a right to protect their
01:14:24
people? So those are the legal
01:14:26
arguments, the discussions that are
01:14:27
going on in all of this. And there's
01:14:29
interesting implications which is
01:14:30
fundamentally if the states get to
01:14:32
interpret federal law and ignore federal
01:14:35
regulatory bodies, it opens up a whole
01:14:37
new can of worms in terms of like all
01:14:39
the states can start to ignore federal
01:14:41
regula regulatory bodies like the EPA or
01:14:44
the FDA or the USDA or and on and on and
01:14:46
on. So the whole case has a whole bunch
01:14:48
of really interesting implications wound
01:14:50
up in it. when you hear these guys and
01:14:51
they're just talking chimat about that
01:14:53
exact like interpretation of the law and
01:14:55
that's what this comes down to. It's not
01:14:56
the actual case that matters
01:14:58
>> and after the sachs they oral arguments
01:15:01
and then they have like a private
01:15:02
conference where they'll write their
01:15:04
papers and give their final judgment.
01:15:05
Yeah.
01:15:07
>> Saxs.
01:15:07
>> Yeah. Yeah, I think what happens is that
01:15:10
so I guess there's some discussion that
01:15:12
happens behind closed doors and they
01:15:14
figure out where the majority is and
01:15:16
then the chief gets to assign who writes
01:15:18
the opinion for the majority
01:15:19
>> in that meeting nobody is allowed in and
01:15:22
in fact you have a double door system
01:15:24
where like if anything needs to come in
01:15:25
and out you have to like kind of like
01:15:27
knock on the door you're led into this
01:15:28
anti chamber then
01:15:29
>> oh is it an airlock is an airlock
01:15:31
>> it's effectively we I had I don't know
01:15:34
if you were there Jason but we had Ted
01:15:35
Cruz come to play in the poker game
01:15:37
>> uh And Ted Cruz clerked for William
01:15:40
Ranquist and if you want to have an
01:15:42
incredible dinner, ask him about the
01:15:45
Supreme Court and Bill Ranquist. He's a
01:15:48
real student of the Supreme Court and it
01:15:50
just makes the Supreme Court free to
01:15:51
your point sound like the most
01:15:53
incredible body that's ever been created
01:15:57
anywhere. By the way, more than the
01:15:59
White House, more than the Capital
01:16:01
Building, more than any of these other
01:16:03
big agencies, this place has it's almost
01:16:06
like being in England, it has these kind
01:16:08
of ways that people operate. The the the
01:16:10
security is so different. They kind of
01:16:12
stand there in the court and they all
01:16:14
exchange places every 20 minutes. It's
01:16:16
very coordinated. They're dressed very
01:16:18
differently than any other
01:16:19
>> courtroom listening.
01:16:20
>> Maybe like 150, I would say. How do you
01:16:23
take it? Are they on
01:16:24
>> I actually think everyone is a guest of
01:16:26
a clerk or someone that works at the
01:16:27
court. I don't think that it's like very
01:16:29
publicly available to get in there.
01:16:30
>> You can't line up. There's no lineup.
01:16:32
>> There's I don't know if there's a
01:16:33
lineup. Um this was a connection through
01:16:36
we got in through the chief justice. Um
01:16:38
he gave us the pass, but I think it was
01:16:40
like very um
01:16:42
>> I think at the Elon versus Open AI case
01:16:45
there's you can line up and then the
01:16:47
judge gave like 30 tickets to the press
01:16:50
court. No, no. Yeah, but I think there's
01:16:52
a lineup for the Supreme Court as well.
01:16:54
There's some public access that they're
01:16:55
>> It did not look like anyone from the
01:16:57
public was in this court. Everyone,
01:16:59
everyone is dressed respectfully. I
01:17:00
mean, this court has an incredible
01:17:02
amount of like,
01:17:04
>> you know, cool experience.
01:17:07
>> I would I would just say uh enjoy it
01:17:09
while you can. I mean, I think the
01:17:11
Supreme Court is one of the last highly
01:17:12
functional institutions in the United
01:17:14
States. And%
01:17:16
>> you know at some point we're going to
01:17:17
have like 13 or 21 or some crazy number
01:17:20
of justices up there
01:17:23
and get jersey
01:17:25
after justices there and so enjoy it
01:17:28
while it's still
01:17:29
>> in the current in the current form it's
01:17:31
in.
01:17:32
>> Can you imagine showing up with jerseys
01:17:34
with the justices names on them and like
01:17:36
having sections and like somebody
01:17:38
selling cracker jacks
01:17:40
>> theocracy version of the Supreme Court
01:17:43
>> version. Exactly.
01:17:45
>> The popularity of the court really
01:17:47
depends on whether it's issuing
01:17:51
decisions that people agree with. That's
01:17:52
what it comes down to. If like if you
01:17:54
ask people whether they like the Supreme
01:17:56
Court or not, it really just depends on
01:17:58
whether they agree with the decisions
01:17:59
are recency as opposed to
01:18:02
>> the process of the decisions and how
01:18:04
well argued it is and all these things
01:18:05
that you're pointing to. And actually
01:18:08
the the court I mean I just checked the
01:18:09
numbers. The court is relatively popular
01:18:12
right now. I think that it got as low as
01:18:16
35% in the 2024 Gallup survey, but I
01:18:19
think it's back up to, you know, like 44
01:18:23
to 50% favorability, which for something
01:18:26
that's involved in politics is
01:18:28
relatively high, right? Like you look at
01:18:30
Congress or
01:18:31
>> any particular politician, they're going
01:18:33
to be lower than that typically. I just
01:18:35
felt so assured of like the institution
01:18:40
when I visited and saw these guys
01:18:41
interact and behave and how they behave
01:18:43
the process. It was like
01:18:45
>> man this what an amazing country. Yeah.
01:18:48
>> Well, the reason I say what I say is
01:18:49
there was an interview with James
01:18:50
Carville recently. Did you guys see
01:18:52
this? He saidaw he said look when we get
01:18:54
power we're packing the court.
01:18:56
>> So we're not even going to we're not
01:18:57
going to worry about it.
01:18:58
>> And we're going to get to 13, right? He
01:19:00
said we're going to make
01:19:01
>> I think Yeah. They're going to go from 9
01:19:02
to 13 and then they're going to create
01:19:04
some new states and all the rest of it.
01:19:06
So that'll be that.
01:19:07
>> Uh
01:19:08
>> enjoy enjoy while it last. Enjoy.
01:19:10
>> Enjoy while it lasts.
01:19:11
>> Uh by the way,
01:19:12
>> end on a high note.
01:19:13
>> Wait. Yeah. It's the end of the empire.
01:19:15
That'll be that.
01:19:16
>> By the way, there is uh a Supreme Court
01:19:19
on I was correct. There is an online
01:19:21
ticketing lottery. So we can all sign up
01:19:24
and you can get a fourack of tickets. I
01:19:26
think they should make this I we should
01:19:28
talk to Howard Lutnik. Maybe he can make
01:19:30
this an auction. We get a revenue stream
01:19:31
from the US. We could sell like 10 of
01:19:33
the tickets as courtside seats for 20
01:19:35
grand.
01:19:35
>> Jason, you're exactly what they're
01:19:37
trying to protect against.
01:19:38
>> Exactly. Like, how can we how do we
01:19:40
monetize the Supreme Court?
01:19:43
>> All right, everybody. That's it. That's
01:19:44
the world's greatest podcast for you for
01:19:46
Chimoth Poly Hatia, David Freeberg, and
01:19:48
David Saxs. I am the world's greatest
01:19:51
moderator. We'll see you chief justice.
01:19:54
>> I'm like the chief justice of the allin
01:19:56
podcast.
01:19:59
We'll let your winners ride.
01:20:01
>> Rainman David
01:20:06
and it said we open sourced it to the
01:20:08
fans and they've just gone crazy with
01:20:09
it.
01:20:10
>> Love you queen of
01:20:13
winners.
01:20:19
>> Besties are gone.
01:20:21
>> That is my dog taking a driveway.
01:20:26
>> Oh man. My habitasher will meet.
01:20:29
>> We should all just get a room and just
01:20:30
have one big huge orgy cuz they're all
01:20:32
just useless. It's like this like sexual
01:20:34
tension that you just need to release
01:20:35
somehow.
01:20:40
>> Your feet.
01:20:42
We need to get Mercury's already.
01:20:52
I'm going all in.
