0:00

Has anybody else seen due to an IMAX? No,

0:02

don't say anything. Oh, don't say anything about

0:05

Dune. I want to say

0:07

about Dune. No, don't say it is fantastic.

0:09

And I want to see it again. That's

0:12

all I'm saying is that it's worth seeing twice.

0:15

I don't even want to know that you think

0:17

it's good. It's over. It makes me stop. Both

0:20

of you see, he didn't say it. He didn't sax didn't

0:22

see it. No, I saw Dune. Dune

0:25

two. Yeah, I saw it. It's over.

0:27

Okay. Stop, stop. Okay. Stop. Everybody. You both

0:29

have. It's not over. There's

0:32

like five great scenes. Sax, the

0:34

scene. You

0:37

guys are so nice. All

0:55

right, everybody. Welcome to the All

0:57

In podcast with me again, the

0:59

chairman and dictator, Chamath Pali, Hapatiya,

1:01

David Sacks, the Rain Man. Yeah.

1:04

And Sultan of Science,

1:06

David Freiburg, the Treboy J.

1:08

Cowell here. And we have

1:12

so much to go through. The

1:14

DOJ dropped a Sherman X suit on

1:17

Apple today, right as we were about

1:19

to tape the program. It seems like

1:21

Thursday is the big news drop day

1:24

now. So this DOJ suit outlines five

1:26

alleged abuses and they claim obviously that

1:28

those abuses reduce competition while

1:30

limiting consumer choice and raising the prices that consumers

1:33

pay for their iPhone. We talked about this actually

1:35

just a couple of weeks ago, we talked about

1:37

peak Apple. The five

1:39

categories are very quickly super apps,

1:41

cloud gaming apps, messaging

1:44

apps, smart watches and digital wallets. If

1:46

you don't know about super apps, that's

1:48

the one that's maybe you haven't heard

1:50

of. If you're listening to this in

1:53

Asia, they mentioned how Alipay,

1:55

WeChat and Paytm are super

1:57

apps. What does that mean? You

1:59

get like five or six different functions in

2:01

one app, social media, images, purchasing, getting

2:04

rides, you know, all that stuff. And

2:07

when you have that, you can move from one

2:09

platform to another super easily. And

2:12

when it comes to messaging apps, we've all experienced

2:14

the green bubble friend. So

2:17

the main argument of the lawsuit, pretty interesting.

2:19

They explain, and it's a really well

2:21

written document, I read it this morning,

2:24

that when Apple faces new competition, instead

2:26

of lowering prices for consumers

2:30

or offering a better deal to developers,

2:32

they would quote, meet competitive threats by

2:34

imposing a series of shape-shifting rules and

2:37

restrictions in its app store guidelines and

2:39

developer agreements. I've faced this every

2:41

time I've invested in an app

2:43

startup. They complain about the

2:46

goalposts moving by Apple and

2:48

Apple shares down about 3% on the

2:50

news today. I'm sure everybody's got some

2:52

interesting thoughts on it. Chamath, you were

2:54

talking about peak Apple just last week

2:56

or the week before, I think. What

2:59

are your thoughts on the lawsuit that

3:01

dropped today? To be

3:03

honest, I haven't read it, and I don't really

3:05

know like

3:10

the odds of these things, just because it seems

3:12

like some of this stuff is so clearly political.

3:15

And I think that these things have seasons to

3:17

it in the sense that there are

3:20

moments where these things are more likely to

3:23

go in the favor of the company or more likely to go in

3:25

favor of the government. The one thing I'll say is

3:27

that these guys have been losing

3:29

a fair

3:31

number of these things. So they're

3:34

kind of 50-50 in their fight with Epic.

3:36

They've basically lost against their fight with

3:39

Spotify. They

3:43

definitely are in a moment where

3:46

people are looking at the

3:48

profitability of these devices as

3:50

the source of their long-term cash flow. And

3:53

I think they realize that there's not a lot of growth

3:55

because they haven't entered a bunch of new markets. So

3:58

the real question is knowing all of them. that did

4:00

they do something beyond what they've already

4:02

always done to try

4:05

to lock people in and can they prove it?

4:07

I think that that's really where the government's case

4:09

will come down to because then it's

4:11

clear that Apple probably

4:14

understood that their market was kind of

4:16

solidifying and they introduced

4:18

these blocks

4:20

essentially into processes that force people to stay.

4:22

That's probably bad news for them but I

4:24

don't really know because I haven't read the

4:26

lawsuit. Yeah, Zach said you get a chance to

4:28

read it or do you have any thoughts generally speaking? I

4:31

haven't read the filings but I read some of the press coverage

4:33

about it. I would say

4:35

based on the press covers that I

4:38

read that there's not really a smoking

4:40

gun here. At least the press

4:42

hasn't reported one. And what I mean by that

4:44

is, let me give you a couple of examples that were

4:46

in the press. One is that

4:49

apparently the filing made a point

4:51

about when Tim Cook was confronted some time

4:54

ago by a user saying when I try

4:56

to send a

4:58

video by text to someone with an Android phone,

5:00

it just doesn't work very well. It never works.

5:02

It was a guy trying to send a video

5:04

to his mom and Tim

5:06

Cook responded, well, your mom should buy an iPhone. It's

5:08

kind of like a throwaway comment by Tim Cook, probably not

5:11

a great idea for him to say that but that's an

5:13

example of what the lawsuit brings up. Another

5:15

example is that the Apple

5:17

Watch doesn't work with Android

5:19

phones. It

5:22

only works with iPhones obviously.

5:25

Again, is that really a smoking gun

5:27

issue? I don't think so. This

5:30

is Apple's whole strategy is that all of

5:32

its products work seamlessly together and Apple

5:36

Hardware doesn't really work with other operating

5:38

systems. It never has. So

5:40

at least based on the

5:42

press coverage, I've yet to see a single

5:44

smoking gun example coming out of this lawsuit.

5:47

Now I think what those examples highlight

5:49

is what this lawsuit is really about

5:51

is your view of

5:54

interoperability. What the

5:56

government is saying is that Apple needs to

5:58

make its apps. its

6:00

devices more interoperable with

6:03

other ecosystems. So the watch needs to

6:05

be interoperable with Android, Apple

6:09

messaging needs to be interoperable with Android messaging

6:11

and so on. What the government is saying

6:14

is that Apple is refusing to play ball

6:16

with other applications, other operating

6:18

systems and that creates a

6:21

monopolistic network effect. What

6:23

I think Apple would say is no,

6:25

the fact is that Apple from the

6:27

outset has always tried to do everything

6:29

soup to nuts. We've always done the

6:32

hardware and the software together and

6:34

that's what creates the magical experience. And

6:36

so Apple's entire product strategy is

6:39

based on creating a vertically

6:41

integrated stack that goes

6:43

from hardware to operating system

6:46

to key applications. And

6:48

I think Apple has a point saying if you try to

6:51

make us unwind all of that, the process is not going

6:53

to work as well. It's not going to

6:55

be the same product experience that everyone's used to. So

6:57

again, I think that how you view

6:59

interoperability at the core of this lawsuit, I

7:01

think that both the government

7:03

and Apple make good points about

7:05

that. And I'm a little

7:08

bit skeptical right now that the government has a

7:10

smoking gun, at least one hasn't

7:12

been reported in the press. And

7:14

so I'm a little bit doubtful right now that the government's

7:16

going to be able to win this case. Freeberg,

7:18

any thoughts? Yeah, I mean, I

7:21

think it's a function of consumer choice.

7:24

If consumers want to

7:26

have this closed system where they can

7:28

iMessage with other iMessage app holders and

7:31

not be able to have a

7:33

seamless integrated messaging experience with Android

7:35

users, they'll be happy. If they're

7:37

annoyed, they'll switch over to an Android. So

7:40

this has always been Apple's orientation.

7:44

I think I mentioned this to you guys. I've been

7:46

on a Mac since 1984 when the Mac original came

7:48

out. I still have my first Mac original, by the

7:51

way, I keep it in my office

7:53

on my shelf. And

7:55

Apple's always had a

7:57

really hard focus on the software.

8:00

that they make available to their users to create an

8:02

incredible product experience.

8:04

So I don't know if this is necessarily

8:07

about market abuse as much as it is

8:09

a consumer choice. If the consumers didn't like

8:12

the product, they didn't like the fact that things were

8:14

not available, that things were very expensive, they

8:16

would go elsewhere. And you do see that in segments of

8:18

the market. And as we talked about a few weeks ago,

8:21

you do see that XUS. Most

8:23

of the smartphone market is actually Android

8:25

driven. Yeah. And this is a

8:27

premium product that people are willing to pay a premium

8:30

price for, even if that means limited access. I think

8:32

developers are frustrated that they can't access this premium market.

8:35

But I'm not sure that Apple necessarily

8:37

should be coerced to

8:40

service developers when at the end

8:43

of the day, the consumers are paying for the product. And

8:45

to your point, like, is it really the case

8:47

that a green bubble versus a blue bubble is

8:50

really that limiting? You know, it's

8:52

not the end of the world. It's annoying. It's

8:54

annoying, but not super limiting. It's not the end of the world. Let's

8:56

say Apple has completely abused their

8:58

power every chance they get in order

9:00

to capture this 30% from the

9:02

App Store. They blocked the ability

9:04

for you to use Audible or, you

9:07

know, other marketplaces to buy books

9:09

and media. They

9:11

blocked the ability to use other

9:13

media players like VLC, blocked the

9:15

ability to use browsers. And

9:18

if you leave Apple unchecked, they just

9:20

keep abusing it. So I think

9:22

this is kind of one of those lawsuits that's like, if

9:25

we don't stop them, they'll just keep

9:27

making it worse. And when people file

9:29

complaints, then Apple will back down, which

9:31

they did on media players and

9:34

they'll stand down. If you think about

9:36

this from the PC era to now, then you

9:38

can see how pernicious their behavior is. When

9:41

you buy a laptop, the idea that you would

9:43

have to go through an App Store and pay a $30

9:45

tax to put a piece

9:47

of software on your computer would seem insane.

9:50

30% tax on your, which would seem insane. Like

9:54

if you buy a laptop, you should be able to put whatever software

9:56

you want on it. But they did this magic trick

9:58

where they said, oh no, if you want to... put software on

10:00

your phone, it has to go through the App Store.

10:02

It has to be 30%. And

10:05

that's really the anti-competitive thing here. I

10:07

think this will be settled. And if you look

10:09

at the different issues here, I think this is

10:12

going to be actually a huge win for Apple.

10:14

Because if iMessage were to exist on

10:16

Android, they would get all of those

10:18

users to download iMessage, and they would have

10:20

all those users. If they made the watch compatible, they

10:23

would open up many more people to buy the watch,

10:25

and you would get more watches. I think that Apple

10:27

is thinking way too short term the

10:29

lock in here, and they should allow

10:32

more of these apps, allow more watches,

10:34

and they should allow the gaming thing. I think

10:36

that they're being petty. I don't

10:38

think anything's going to happen here, because it's taken them

10:40

five years to file. There's

10:44

at least a 50% chance that the administration is

10:46

going to turn over, which means that this lawsuit

10:50

changes or goes away entirely. And

10:52

then even if it does kind of proceed, it's

10:54

going to take 10 years of very detailed arguments

10:56

for something to happen. And

10:59

frankly, probably in 10 years from now, we've already moved to a

11:01

different compute platform, and this is not going to matter. I

11:04

mean, there is the coin toss of what the Trump

11:06

administration would do here. For

11:08

sure, that's valid. But I think these things

11:10

all get settled. I think the settlement will

11:12

be iMessage releases. They allow the, so the

11:14

thing about the Apple watch, it's not just

11:16

SACS that they're allowing it on Android. It's

11:19

would you allow Android watches to connect

11:21

with the SDK directly into

11:23

the iOS operating

11:25

system, which they block you from doing. I

11:28

think they should just have a little switch in your Friedberg

11:31

settings that allows you to flip it

11:33

and say, I take

11:35

responsibility for allowing third party stores and I

11:38

want to be able to load any software,

11:40

it's your computer. I don't think the

11:42

government should regulate that. I don't think the

11:44

government should be able to mandate that. Yeah,

11:46

I mean, I like the fact that the

11:48

government is. Looking to the government to do things

11:50

that they want as a consumer, make your choice

11:52

with your dollars. Make your vote with the

11:55

product and service you want to buy instead

11:57

of running to the government and asking the government to come

11:59

and do stuff. that way about price

12:01

fixing? What do you mean price fixing? In what

12:03

context? If like a group of people price fix the

12:05

price cost of a phone or like let's say the

12:07

30% from the app store, which just happens to be

12:10

the same between two app stores. Getting

12:13

price fixing is something valid for the government to

12:15

come in on? It's not the same. Do you

12:17

think price fixing is something the government should come

12:19

in on, is the question? If there's limited access

12:21

to products in a market and all the participants

12:24

get together to set the

12:26

price in the market, I do think that's anti-competitive. And

12:28

that's, there's a good role for the government to play

12:30

there. But Google and Apple don't price

12:33

fix because Android is an open access system.

12:35

You can put any application you want, any

12:37

way you want. There are several

12:40

app stores to put Android apps

12:42

on your Android device. Google operates an app

12:44

store called Google Play, but there are other

12:46

app stores you can go to as well.

12:48

And they all have different prices. And Google's

12:50

app store costs, by the way, are not

12:52

apples and they're not 30%. And

12:54

we talked about this last time. I'll pull it up for

12:56

you again. It's like 15 to 12 going down the sliding

12:58

scale. So there is competition in that

13:01

market. Generally, I think the point about if

13:04

there's limited access to a market for products for

13:06

consumers and everyone in that market gets together

13:08

and sets a price, that is anti-competitive.

13:10

And there's a good role for government to play there.

13:13

But I don't think that having the government come in and say,

13:15

I want this feature to have this button and this flip in

13:18

my app is what I want my government doing.

13:20

How am I government protecting me from crime? And,

13:24

you know, defending the country. And I don't want

13:26

all the other stuff that the government does to

13:28

drive costs up, which is generally what the government

13:30

does. That's what you think the role of the

13:32

government is here in terms of anti-competitiveness,

13:34

especially in the case, I'm curious your thoughts

13:36

on a duopoly. Like if you look at

13:39

Google search monopoly, pretty easy to

13:41

understand like, hey, they got 90%. But here you got

13:44

two players who have, you know, roughly 50% of

13:46

the market each or 60 40, depending on where

13:48

you are. So what is the proper

13:51

role of government here? And is this an

13:53

overreach? No, I

13:55

mean, I actually think just to give the government lost

13:58

you some credit, I actually think it's good. to

14:00

hold Apple's feet to the fire and make

14:02

sure they're not engaging in anti-competitive tactics. One

14:05

of the things I've said previously on the show

14:08

is that I think the government's been making the

14:10

mistake of going after bigness for its

14:12

own sake. And I don't think that bigness in

14:14

and of itself is bad. It

14:16

might be a reflection that the company's done an incredibly

14:18

good job and that's why it has a lot

14:21

of customers and a lot of market share. What

14:23

I've said is I think the government should prevent

14:25

anti-competitive tactics. So I think that

14:27

this lawsuit is good in the sense that it's

14:30

targeted at the right types of

14:32

things. It's not just going after Apple because its

14:34

market cap is so big. It's going after specific

14:37

things that it's doing to kind of lock

14:39

in its network effect. Now the

14:41

problem that I see is just that the examples

14:43

that we've been given so far from the lawsuit

14:45

just don't seem that compelling. Again

14:48

I'm kind of waiting to find out where's the smoking gun here.

14:51

But to Freiburg's point, I mean look, I don't

14:53

think you can just say that if

14:56

you don't like what Apple is doing, go to a

14:59

different platform because the point is that there's only two

15:01

choices and they

15:03

can both engage in anti-competitive tactics and

15:06

they can both create lock-in. But

15:08

Saks, there are not just two choices. Android

15:10

is an open operating system. Google has

15:12

a fork of Android that they

15:15

put on certain smartphones. There are other smartphone

15:17

makers that use other forks of Android. So

15:20

Android has enabled and the reason

15:22

Google bought Android was to

15:24

provide a counterbalance to exactly the dynamic

15:27

of one handset manufacturer having an

15:29

operating system that could control access

15:31

to the internet and to apps.

15:34

So Google open sourced Android, many handset

15:36

manufacturers use their own versions of Android

15:38

to put their own experience on the

15:40

phone. And then Android users can

15:43

put apps from anywhere they want and

15:45

other companies offer app stores. So there are many

15:47

app stores to go to and there are many

15:49

different handset manufacturers. And a lot of people, I

15:52

think incorrectly, look at

15:54

the operating system landscape in mobile and say,

15:56

oh, there's Android and there's iOS, but Android

15:59

is an open... platform and there are many

16:01

different forks of it run by many different companies.

16:04

So I think that it is actually a very

16:06

competitive market. Like we talked about last week, I

16:08

think close to 80% of global handsets are run

16:10

on Android forks. So, you

16:12

know, it's not necessarily the case that there

16:14

are only two choices. And Google does not

16:16

control Android. There are many developers that contribute

16:18

to open source there. Google has their own

16:20

fork that they do a lot of work on. But

16:22

anyone can take their own fork and do it. I mean, you guys

16:25

can, you know, when you boot

16:27

up a Samsung or Sony TV,

16:29

that's running Android. And that's running their

16:31

own version of Android. Are you arguing that Apple doesn't

16:33

have market power? In the US

16:35

they do. XUS, they're, you know, they're... Yeah, I

16:37

mean, I think, sure, we're talking about the US

16:39

market. I mean, I think the US market, they're about

16:41

50... Are they about 50-50? We pulled this chart

16:43

up last week. But I think they're about 50-50. It's slightly more

16:45

in the US now. It might be 55-45 right now. Yeah, something

16:47

like that. The issue is the

16:49

cat and mouse that Apple is really good at playing,

16:52

they will push their advantage. And so we see

16:54

the five things that the Justice Department wants to go

16:56

after here, but the App Store is obviously one of

16:58

them. And before that, I mentioned

17:00

some of the other categories where they do this abusive

17:02

power. They basically keep it up

17:04

as much as they can until they get checked. And

17:07

I think interoperability is

17:09

actually in the long term going to be

17:11

in their best interest. They use interoperability when

17:13

they need it. So Apple Music and iTunes,

17:16

they allow you to use Windows and Android.

17:18

It's only when they see an advantage. And that's

17:20

really the hypocrisy of Apple. And I

17:22

think it's really important that the

17:25

industry stand for more interoperability and

17:27

the ability for you to load any software

17:29

that you want on your device. This is

17:31

a compute platform. The ability for them to

17:33

block you from installing your own apps on

17:35

it, I think is a really terrible

17:38

question. By the way Jason, what you're

17:40

saying is a very fair statement from a consumer point of view

17:42

and from an industry point of view. And

17:44

that's where that voice should come from

17:46

to force in the marketplace changes from

17:48

Apple. But having the government do this

17:50

and having the government involved, I think sets all these

17:52

data. Yeah, they're representing the people. Somebody

17:54

did it. They asked Jim Cook to do it. They

17:57

found salt biophome for your mom. No, but aren't they

17:59

representing the people? Like how do you expect

18:01

the people to organize like file a class action?

18:04

No, I think if enough companies boycott the Apple

18:06

platform and tell consumers to boycott though I mean

18:09

consumers can make decisions That's not a

18:11

person can make decisions whether or not they want to

18:13

develop for up now that is not realistic You know

18:15

that's not really talking about that's nice. Come on I

18:17

don't think I want the government stepping in to tell

18:19

private companies what to do because customers don't like what

18:21

they're doing Wait, would you think

18:23

that the government's just stepped in to prevent

18:25

Microsoft from taking over the browser the whole

18:28

Netscape case? Bundling

18:31

I don't know the case well enough. I Mean

18:33

I remember it, but I don't want to

18:36

speak I don't want to speak definitively There

18:38

was a substantial chance that if the government

18:40

didn't step in then Microsoft would have extended

18:42

their Windows operating system monopoly They did have

18:44

an absolutely not only they did have a

18:47

monopoly on personal compute OS. I wasn't an

18:49

absolute monopoly I mean, you're the one you're

18:51

using Macintosh At

18:55

the time Macintosh was like 5% of the market was

18:57

under 10% at the time Yeah, it was about four

18:59

or five percent at that time to off Microsoft was

19:01

95% of

19:03

the personal computing market So they did have an

19:05

absolute monopoly on that choice. You had a consumer

19:07

choice It wasn't a very good consumer choice, but

19:09

you sold my choice. You could go

19:11

use Macintosh and you did you were you know I

19:13

don't think I could afford it. Yeah,

19:16

okay But the point is that Microsoft had substantial

19:18

market power and what they were in the process

19:20

of doing Was the smart thing

19:22

from a business standpoint, which is you take your

19:25

operating system monopoly You use that to

19:27

extend into the browser you kill Netscape

19:30

you take over with explore you bake

19:32

explore into Your Windows operating

19:34

system. Yeah, two things are basically the

19:36

same then there hold on from there

19:38

You leverage your control of the browser

19:40

in the home screen to control search

19:43

Okay, so you think about like all the

19:45

dominoes that would have fallen if Microsoft

19:47

had continued maybe if like

19:50

Bill Gates a continuing CEO and

19:52

there would have been no Google there would have been no Yahoo They

19:54

would have controlled the internet they would have control the internet and

19:56

they were actively you left out a big piece of it Sax

19:59

which you're wearing remember the second I

20:01

say it, they were trying to break

20:03

HTML and open standards by creating ActiveX

20:05

and using all those funky colors. And

20:08

so you have to understand these

20:10

companies will, if you give them

20:12

an inch, they'll take the mile.

20:14

And that's exactly what Apple does

20:16

consistently. They consistently try to squeeze.

20:19

I think it's great. Saks is right. Tactics

20:21

is the way to go after these companies. These are

20:23

five recent tactics. I think the

20:26

government's going to win changing three of these. And

20:28

that's win. And for me, the mobile operators tried

20:30

to do the same at that same time. You

20:32

guys remember Verizon? Yes, you couldn't have

20:34

that. You could

20:37

only put their apps if you bought it

20:39

through their store on their operating

20:41

system on the mobile phones that they contracted to

20:43

have made. So they all

20:45

had their own custom version of the

20:47

handset manufacturers OS, which I'll get.

20:50

And then they did red share back with the

20:52

manufacturers. And that's why Android was

20:54

acquired. It wasn't actually at the time to compete with

20:57

Apple. Android was acquired to compete

20:59

with all the mobile phone companies that were trying

21:01

to block access to the internet and apps. Well,

21:03

Google wanted to maintain its search monopoly, let's be

21:05

honest. That's why they bought it. They

21:07

knew users would be there and they wanted to have the

21:10

default browser. They didn't want to have the access to the

21:12

internet block by the handset manufacturer. And

21:14

they wanted to be the default browser in the app,

21:16

which is what you have to agree to, by the

21:18

way. So if you want to talk about perniciousness and

21:20

heavy-handedness, if you have an Android and

21:23

you want to use the Android updates from Google, you have

21:25

to give the bundle of apps. If

21:28

you want the updates and support from Google, you can fork

21:30

it, but then you don't get their

21:32

support, which then breaks your phone. If you

21:34

want their version, that's right. Yeah. So

21:37

there's a little cleverness here, I think. So

21:39

you have two basic ecosystems in mobile. It

21:41

is a duopoly, and both of them advantage

21:43

themselves in ways that

21:47

grow over time. And I think that

21:50

if Apple were completely left unchecked, it

21:52

would figure out ways to boil the

21:54

frog and keep extracting more and more

21:56

value out of downstream applications. Totally. And

21:58

I think we're preparing... 1600

22:00

a phone now. How are we paying 1600 a phone? So

22:03

look, I think it's I think it's good for the government

22:05

to hold their feet to the fire. But here's the thing

22:07

that the government's lacking is, again,

22:09

that example they can point to that's really

22:11

compelling. I mean, with the Microsoft Netscape thing,

22:14

it was really obvious what they were doing,

22:16

right? The browser was the gateway to the

22:18

whole new platform, which was the internet. And

22:21

if Microsoft could make that a

22:23

feature of the operating system, their dominance

22:25

would extend into the new computing platform. But do you

22:27

remember how they did it, Saks? Apple Watch is just

22:29

not the same thing. The Apple Watch is kind of

22:31

an appended manner of your phone that

22:34

most people don't even want. So it's

22:36

hard to point to something. Do you remember Saks

22:38

how they were getting the

22:40

browser built in? They were going

22:42

to the OEMs, the Dell's of the world, the HPs

22:44

and saying, if you want this

22:47

price for Windows, you have to

22:49

include the browser on the desktop, it has to load, it

22:51

has to ask your credentials, all this stuff. Or

22:54

you could pay $150 for Windows

22:57

if you want it without the browser bundled.

23:00

So they use this pricing to

23:02

get the OEMs to bundle it. And

23:04

that was where... Well, isn't that what Google's doing

23:06

with Android effectively? Exactly. Is like...

23:08

It's exactly right. You can have

23:11

your own version without Google search if

23:14

you're willing to put up with all

23:16

these headaches. And maintain an operating system,

23:18

update the operating system. That's exactly what

23:20

they're doing. So anyway, I think

23:22

three out of five of these get settled and it'll

23:24

be good for consumers. Ultimately, I would... Not

23:27

financial advice, but I think that Apple

23:29

is going to be able to manage this. I'd buy the stock. I

23:32

may buy more of the stock. I think that this will be good

23:34

for them long term. All right, listen, going

23:37

even deeper into Apple and

23:40

this Google relationship, big questions emerged this

23:42

week, as Bloomberg reported, that

23:44

Apple was speaking to both

23:46

Google and OpenAI about powering

23:48

certain AI features on

23:50

iOS specifically, according to the article,

23:52

this deal between Apple and Google seems much

23:55

more likely than the OpenAI one and could happen

23:57

this year. As you know, we've talked about it

23:59

on this show. before, Google has been the

24:01

default search engine on iPhone for I think

24:04

over a decade now and Google pays

24:06

Apple 20 billion a year. That's

24:08

pure profit for Apple. It's unclear

24:10

what features they would power with

24:13

Google Gemini. Maybe it's like

24:15

the search field and Google Gemini is on

24:17

there. Maybe it's built in and

24:20

Apple has been building its own LLM. It's

24:22

called Ajax. They've been doing some other open

24:24

source stuff called Maggie. What do you think,

24:26

Chamath? You were talking about this on Twitter.

24:28

You had a pretty good tweet

24:31

and a pretty strong position. If

24:33

Apple is going to use Gemini, what

24:35

does that say to you about Apple

24:38

and their view of the future? I think it's

24:40

them giving up. This is the most consequential new

24:43

development in technology and compute in

24:45

probably 20 years, 30

24:48

years. To be spending

24:51

tens of billions of dollars of R&D and

24:54

to not have enough allocated to this so

24:57

that you have a legitimate path forward to

24:59

do it yourself I think is a little

25:01

inexcusable actually. It's

25:04

akin to IBM in the 70s going to

25:07

Microsoft and basically asking them

25:10

to build the operating system for them. When

25:13

you're such a dominant company and such

25:15

a dominant position but you

25:17

abdicate your responsibility to innovate, I

25:20

think that that's a really bad situation

25:23

for a company to be in. It's like

25:25

Yahoo adding Google search, right? By

25:27

the way, on the heels of turning off

25:30

cars and losing

25:32

these antitrust issues

25:35

in Europe and then being sued by

25:37

the DOJ here to

25:39

then be in a licensing discussion with

25:41

a third party for such a critical

25:43

technology I think just says not

25:46

very good things about the state of the company.

25:49

Freeberg, your thoughts on this potential deal? We

25:51

don't know what features this with power. I

25:53

don't know if it's Siri or it's just

25:56

image editing or search or

25:58

an extension of search maybe giving an answer. Sir,

26:00

I have no idea. It seems to me like

26:02

there's the. If Apple's doing the

26:04

right thing, which I'm sure they are, they're probably

26:06

building their own alternative. Platform.

26:08

Here. They. Realized going to take them

26:11

some time in the interim. They want to have a

26:13

stopgap and my guess is caught by going to Google.

26:15

They're probably going to get paid. Instead of

26:17

having to pay someone else for the technology,

26:20

To. Google will realize some benefit from

26:23

getting users over to search results

26:25

and seeing ads so. Google

26:27

will probably benefit and pay them instead

26:29

of. Them.

26:31

Having to go pay someone for access to some

26:34

technology that consumers might want access to have. Sex.

26:36

Anti Nazi or I'm sure apple going to make an

26:38

investment building the role models here. Yeah.

26:41

I mean who would look at the

26:43

launch of Google Gemini and say I

26:46

want that. That launch

26:48

was a total fiasco. Another well company

26:50

I didn't feel the apple philosophy. Plenty

26:52

of the how look would simply have

26:54

to be selected the launch a Gemini

26:56

and say i don't see the longer

26:58

you got that exactly. What

27:01

I want That. We

27:04

the President of United States I mean

27:06

Tim Cook is the only person on

27:08

Earth who's impressed with the launch of

27:10

Google Gemini. I'm it's I can't believe

27:12

the story is true. It's just a

27:14

can't be true I'm so excited today

27:16

to large stamina. Hats on states each

27:18

year. Heard ico as Twenty Seventh. Oh

27:21

my God can you imagine of my

27:23

siege? two in Search of Blood is

27:25

jeopardized? The story can't be true because

27:27

of be so strategically.like Jamal said, Even

27:29

if Google July were great, you'd snow

27:31

want to invest in having your own

27:33

saying because it's such a strategic technology.

27:35

why would you ever outsource it to

27:37

your my about an hour? But in

27:39

this case you know the Gemini was

27:41

terrible. It was a fiasco, And

27:43

so. Is the story makes us

27:46

as can't believe a stroke is probably people

27:48

knocking on each other's doors. They spent thirty

27:50

billion dollars less your normandy. Thirty.

27:52

Billie. Apples.

27:55

To be clear, They. spent thirty billion

27:57

dollars on are indeed what did they

27:59

spend don't. Catering.

28:01

I mean, you got to think Apple Vision

28:04

is a big piece of that, right?

28:06

Chips. I guarantee you like half of

28:08

that is like the M3, M4 chip. Yeah, but

28:11

I'm not saying 30 billion, like not a couple

28:13

crumbs fall out of somebody's pocket and so

28:15

there's 50 million allocated to just

28:17

like, you know, mucking around with llama or

28:19

mucking around with Mistral. What

28:21

if they took Project Titan, the car

28:24

deal, that was 10 billion supposedly

28:26

they had spent on that? It's

28:28

towards AI? It would be Microsoft. Again,

28:31

it goes back to the people. I don't think

28:33

the person in charge of the windshield wipers for

28:35

Project Titan is going to be the person that

28:38

figures out how to land a really killer LLM.

28:41

That's a different skill set. But

28:43

it's not even that expensive. I mean, you're right. One's

28:45

a mechanical engineer and one's a computer scientist. So these

28:47

are not the same people. They're not fungible that way.

28:50

I mean, I guess the question here is Apple capable

28:52

of building on a culture basis,

28:54

Freiburg, this type of software. They're

28:57

a hardware company. They have it in

28:59

their data. If you're standing on the shoulders of the whole open source

29:01

movement, all you have to do, like Jamasa, all you got to do

29:03

is take the latest Mistral model

29:05

or jump on Hugging Face and start

29:07

working. I mean, you're

29:09

not starting from zero because of open source. Not

29:12

only are you not starting from zero, not only

29:14

do you have the foundational models that are excellent

29:16

and available and open source, you have

29:18

probably the most prolific set of training data that

29:20

has ever been created in the entire world to

29:22

make these models kick ass. Yeah,

29:24

you have all the Apple photos. Okay,

29:27

I wonder if in their terms, that's the thing is their

29:29

terms of service is so privacy based. I

29:31

wonder if they could even use that information. And

29:33

I just think, look, at the end of the day, the Apple

29:35

brand is still exceptional. And so if

29:38

you're given a $10 million a year job

29:40

at Google versus a $10 million a year

29:42

job at Meta versus a $10 million

29:45

a year job at Apple, if you're like

29:47

one of these killer AI people that

29:49

get these kinds of offers, I

29:52

got to presume that Apple gets their fair share of these

29:54

people. Yeah, you would think.

29:56

And even in the worst case, you

29:58

see Microsoft doing like pretty heavy

30:01

handed deals with companies like OpenAI

30:03

and just to speak with Inflection. So

30:06

it's not as if these deals can't be

30:08

done, right? And so

30:10

it just kind of like leaves, I

30:12

don't know, it's just, it's quizzical to

30:15

spend that much money to

30:17

not necessarily be willing to compete on the

30:19

human capital, to not necessarily be in

30:21

the market acquiring these

30:23

businesses. I don't

30:25

know, it's just a question mark, like what's going on in

30:27

there? And Jomath asks, what's going on over there? We have some

30:30

insight footage. Nick, you wanna play tape? What's

30:37

going on here? What

30:43

am I watching? I

30:50

hope we didn't keep you waiting. Mother

30:53

Nature, welcome

30:56

to Apple. How

30:59

is the weather getting in? Oh,

31:06

she's controlling the weather. The weather was however I

31:08

wanted it to be. Yeah, they did this couple

31:10

of. But this reflects their self image. I mean,

31:12

they released this. They care about the plan. No,

31:15

it reflects that they think somehow this

31:17

is an accurate reflection of the way they do meetings. No,

31:21

I think it was the point they were making

31:23

is they were trying to do a skit about

31:25

like how they're doing less packaging and including that

31:28

skit was to tell you that you're not getting

31:30

a charger in your next iPhone because they don't

31:32

wanna say anything. You didn't think it was cringe?

31:35

I thought they did a little cringe. Maybe

31:37

it's a little cringe. Yeah, it

31:39

was super cringe. I do appreciate that

31:42

they are doing sustainable products

31:44

and not putting as much stuff in landfill. So

31:46

I think they should get a lot of credit

31:48

for that. What I do think about this deal

31:50

that's really interesting is it's sort of cementing the

31:52

Google and the Apple alliance

31:55

against Microsoft. And Microsoft has

31:58

big AI ambitions. So this is

32:00

kind of interesting. If you lock

32:02

in the duopoly of Google

32:04

and Apple, and then you lock in Google search

32:06

monopoly, and you start fighting

32:08

Microsoft, I think that's what's probably going on

32:10

here is they're trying to figure

32:12

out how do we keep

32:14

Microsoft away from running away with this.

32:17

All right, we covered that huge NAR

32:19

lawsuit back in December, and there has

32:21

been a settlement, big news for

32:23

the National Association of Realtors. They've agreed to

32:25

pay $418 million in

32:28

a settlement last week. And federal

32:30

jury found that the NAR and

32:32

several large real estate brokerages

32:34

conspired to artificially inflate agent

32:37

commissions. The settlement is pretty

32:39

big deal. People are freaking out

32:41

about it. As you know, the seller

32:44

of a home pays the buyer's

32:47

agent's commission. So you have a buyer and

32:49

a seller, 6% fee typically, sometimes

32:52

it's five, but they split that 3% to

32:54

the buyer and the seller, but the seller

32:56

is paying that 3% to the buyer. Now

32:59

that can't be listed in the MOS anymore,

33:02

and that deal cannot be done ahead of

33:04

time. Buyers are

33:06

responsible for paying their agents commissions going forward. So

33:08

if you bought a million dollar house and

33:11

you were the buyer, you would pay 30,000

33:13

to your buyer's agent, or you would choose

33:15

to not have an agent, or

33:17

you would choose to negotiate it, and

33:20

you have to have a signed contract. This is

33:22

a crazy, just shocking

33:26

shock to the system, according to most people who are in

33:28

it. I've seen a lot of real estate folks

33:30

who are saying this is gonna be healthy because

33:32

you have this, you have to have this conversation

33:34

between the buyer's agent and the buyer, but

33:38

commissions in the United States are a

33:40

hundred billion dollars a year, and

33:43

one analyst projected the lawsuit to lead to a

33:45

30% reduction in commission

33:47

payments, and that would

33:49

eliminate about half of the 1.6 million active

33:51

NRA members from the industry. You had a lot of

33:53

feelings on this. Freidberg, when we talked about it a

33:56

couple of months ago, what are your

33:58

thoughts on this settlement? Is anything gonna change? Is this

34:00

as groundbreaking and shocking as

34:03

people seem to think it is? Well,

34:05

I would take this settlement

34:09

along with a lot of the

34:11

developments and advances in AI as

34:16

being the moment of catalyzing

34:18

real change in the residential real

34:20

estate agency industry. It's

34:23

an industry that's been known to have

34:25

fixed pricing and

34:27

be very expensive to consumers, a

34:30

real tax on the system. And

34:33

it's largely been wrapped around this

34:35

idea of mitigating

34:37

your liability, reducing risk,

34:40

servicing the customer. Many

34:42

of those aspects over the last couple

34:44

of years have been largely standardized

34:46

through forms, digitized because so

34:48

much of this information is no longer going

34:50

to get paperwork from the courthouse, but a

34:52

lot of the information sits digitally and can

34:54

be accessed in a democratized way. And

34:57

the fact that so much of the service

34:59

and discovery, reading through documents, understanding what they

35:01

mean and what they say can be automated

35:03

through AI and LLMs, much of

35:06

this is kind of coalescing around what I hope

35:08

and expect will be a

35:10

more seamless, automated, direct

35:12

marketplace for consumers. The challenge is

35:14

that most consumers put most of

35:17

their personal net worth into their

35:19

home. And so it

35:21

is where most people's net worth is tied up. And

35:23

so because it is such a sensitive

35:25

investment and it is their entire savings,

35:29

they want to have a trusted advisor by them. So

35:32

it's going to take some time before

35:34

that trusted advisor becomes some piece of

35:36

software. But I do think that software is going

35:38

to play more and more of a role in

35:40

providing advisory tools and services to consumers

35:43

in this transaction marketplace. And that's only

35:45

going to catalyze and accelerate

35:48

the fee reduction. I

35:50

do project and I do expect that much of

35:52

what is charged on a commission basis on a

35:54

percent of home value today will change to being

35:57

a fixed fee. And a 5K, 10K. for

36:00

your buyers. Different services. So you can have someone

36:02

do... A la carte. Do

36:04

the disclosure diligence for me for $5K.

36:07

Negotiate the purchase for me for $10K. And

36:10

you as a consumer will start to pick from a

36:12

menu of the services that you want

36:14

to have rendered for you and things that you're comfortable

36:16

doing yourself. I don't need someone to negotiate price. I

36:18

don't need someone to find me a home. I've got

36:20

Redfin. I can go do that. I can arrange open

36:22

houses on my own. The lockbox is there. I'll go

36:24

walk around the property. I don't need someone to point

36:26

out that the color is nice in a room. And

36:29

so I think that there's elements of what will

36:32

happen here, which is a fragmentation and

36:34

then an automation of these services. And

36:36

as a result, significantly reduction. And

36:39

I'm in the middle of doing this myself right now with

36:42

a piece of real estate where I'm not using an agent. And

36:45

I've been using a direct

36:48

listing service. I've used all of

36:50

these standard forms. There's a lot of AI tools you can

36:52

use to kind of read all the disclosures, see and make

36:54

sure everything's scope aesthetic. And

36:56

these escrow agents, they'll handle a lot of what

36:58

a lawyer will handle. And they'll

37:00

get paid a fee, which

37:02

is much less than the agent's fees. So I do think

37:04

that there's a big disruption happening in this industry. I think

37:07

it's really important for consumers. Agents are going to be

37:09

hands in the air telling you, this is ridiculous. You

37:11

need someone to help you. You need an advisor. That

37:13

will continue for a good chunk of the market for

37:15

a very long period of time. But

37:18

I do think that it's for the benefit of

37:20

consumers over time to see these fees come out of

37:22

the system and see those savings go

37:24

back into consumers' pockets and for the value of

37:26

their real estate to go

37:28

in their pockets, not into an agent's pocket. Is it

37:30

going to change the profitability

37:36

of a realtor? Pretty meaningfully, right? Both realtors won't

37:38

be able to be in business. Right. I think

37:40

the sellers will do fine and they might capture

37:43

more of it because my

37:46

understanding is the seller

37:48

will maybe do both sides of the

37:50

transaction. No, that's not what's going to happen. And there's

37:52

also going to be limits on that. But here's what

37:54

I will say. If you

37:57

look at the number of people, there are one... 1.4

38:00

million members of NAR

38:03

today, the National Association of Realtors.

38:06

If you look at the distribution

38:08

of earnings, you guys know this, my

38:10

guess is probably 10% of

38:12

those Realtors make 40% of the fee income or

38:14

50% of the fee income. That

38:16

brings a long tail. So there's

38:18

probably a third of those folks who are

38:21

already kind of sub

38:23

living standards in terms of income. Maybe

38:26

half of them won't be able to make enough

38:28

money in this new, you know, fee

38:30

regime that it'll no longer be an

38:32

attractive proposition to be a real estate agent for

38:35

maybe half a million to a million people

38:38

over time that are agents today. Yeah,

38:41

you know, interesting this, the

38:43

sellers are going to be faced with

38:45

buyers who just show up having seen

38:47

something on Redfin and don't have a

38:49

buyer's representation. And so they think, from

38:52

the stuff I've been reading that the sellers agent

38:55

might be pointing them to, hey,

38:58

go to these services and

39:00

be acting as like one broker, essentially

39:03

representing both sides. That's what

39:05

people are saying is the potential downside. I don't

39:07

think that's not gonna happen for a couple of

39:09

reasons. But let me ask you guys a question.

39:11

If you guys wanted to

39:13

go buy a new house, currently, you just go, you

39:15

know, sign up an agent or buyer's agent, and they

39:17

go walk with you and eventually they'll get paid by

39:20

the sellers agent. So now you have

39:22

to negotiate a fee with them upfront. Would

39:24

you negotiate a fee and say to a buyer's agent, hey,

39:26

I'll pay you 2% of whatever home I buy?

39:29

Would you be comfortable doing that or 3% or 4%? How

39:33

would you have that conversation? No, well,

39:35

you say that I mean, forget I know you've got a

39:37

different situation, because you've got real estate people

39:39

like working for you. But like,

39:41

if you were to go out and if would you go out

39:44

and get an agent and pay and negotiate a few of them?

39:46

It is not worth it to the

39:48

buyer to pay 2 to 3% of

39:50

the purchase price to make appointments. You

39:52

Can see all the houses on MLS

39:54

through Zillow or Redfin or whatever. And

39:56

What about handling clothing and disclosure? No

39:58

Buyer would ever volunteer. Surely agree

40:00

to pay this massive permission is

40:03

not worth it. So. It's

40:05

game over for the Realtors. If.

40:08

Buyers. Are forced to pay their

40:10

own. Brokers, Commission. The.

40:13

Only reason the system works. Is

40:15

because the sellers forced to pay for. You.

40:18

And when you sign the listing remote. With.

40:20

The sellers agent. You. Can

40:22

negotiate a little bit at the margin. Sometimes you

40:24

can get their six percent down to. Forty.

40:27

Five percent for a big listening. But.

40:29

Fifty percent of it will always go to the by

40:31

side. I. Mean I settings guys the by side

40:33

isn't doesn't do anything. why don't you make it. To.

40:36

Percent for yourself, one percent for the buyer.

40:38

They won't do that, They just won't take

40:40

your business. The seller will not represent you

40:42

fab like offers a rules against it. So

40:44

as a whole thing is like protect It's

40:46

like a rocket does protect and now interact

40:49

in well I saw on like a little

40:51

bit skeptical that just gonna work out exactly

40:53

the way were saying because it is just.

40:55

Such. As a death blow to the industry

40:57

of buyers are forced to pay their own to

41:00

missions. Either. Or half by side,

41:02

brokers, missions and in the articles they're saying

41:04

they're stole i some gray area by was

41:06

gonna happen. But. That.

41:08

Is what should happen. Buyers. Be

41:10

responsible for paying their own brokers and if you

41:13

do that either your knockout half the fees. The

41:15

sensory peers made him one thirty sharp. charge an

41:17

hourly fee as buyers each and like even when

41:19

people you know jake out two hundred dollars an

41:21

hour. three dollars an hour to get understand fact

41:23

home prices as one like if you if the

41:25

buyer had to pay all of a sudden they're

41:28

affordability. Effectively. Goes down because

41:30

if they have to pay an extra hundred thousand

41:32

dollars for home than that's a hundred thousand dollars.

41:34

they can't pay to that less than they can

41:36

pay. For. The house itself because have

41:38

to be an agent so but lead author to

41:41

home prices. Not a nets out because the sellers

41:43

agents no longer paying six percent they're paying three

41:45

percent. And so now the

41:47

sellers agent at three percent more. That.

41:51

Matter. your natural thing that's the city and

41:53

but net net on a it's good for

41:55

buyers and sellers because transaction costs of go

41:57

down exactly how many goes back and consumers

41:59

that I think it's a great opportunity for

42:01

startups. I'll say this right now. Like I think there's going

42:03

to be a lot of startups that are going to come out of this ruling

42:07

that are going to launch a la carte

42:09

services, leveraging AI to make these services available

42:12

direct to consumers without needing an agent. And

42:15

they're going to be pretty compelling services and they're

42:17

going to show up real fast. I mean,

42:19

a lot of people will not list their

42:21

home or sell it because that 6%

42:23

might put them underwater. So

42:26

if it's now down to two or three, yeah, people are going to be

42:28

able to sell it. Totally. Yeah, people might

42:30

be like, you know what? It's always been crazy to

42:32

me that anyone would pay a percent of absolute value.

42:35

It makes no sense. I bought a home for a million.

42:37

I'm selling it for a million one. So

42:39

I've made $100,000 profit, but I got to pay $60,000 of my total

42:41

gross to an agent to

42:46

sell it for me. 60% of the profit goes to the name. I

42:49

just gave them 60% of my profit on my home. It

42:51

makes no sense. And the crazy

42:53

thing is that a bad sale

42:56

of your home would be, let's call

42:58

it $900,000 or a million and

43:00

a great sale would be $1.2. So

43:02

it's a very small margin

43:04

where they actually have an impact based on the

43:06

quality of their effort, but they get

43:09

compensated for the whole thing. Yeah,

43:11

for the whole thing. They should get compensated. No matter

43:13

the outcome. It should be that you get a flat

43:15

rate, $10,000 out of a

43:18

million, and then you get 10% over a

43:20

million or something. Yeah,

43:24

the incentive comp should be

43:26

variable based on performance, whereas

43:28

the guaranteed part of the comp should be, like

43:30

you said, a flat fee. It's kind of like

43:33

the salesperson has an OTE. Exactly, like a staff. Like

43:35

a base variable. Yeah. I

43:37

mean, you could say, if you said 1% up to

43:39

a million, so that's 10K, and then 10% for the

43:43

next 100K, and then 20% for the next

43:45

100K. So that $1.2 million is really hard to get. Yeah,

43:49

I'll give you 20% of that incremental 100. That would

43:51

be a much better deal if you

43:53

would be getting paid for the actual performance. So

43:56

really interesting. But you know, J.K., I have an

43:58

idea for the National Association of realtors. Oh, here

44:00

we go. If they

44:02

want to stop being perceived as an evil monopolist, all

44:04

they got to do is put out a cringe ad

44:07

talking about the environment. Everyone's going to love them again.

44:09

Absolutely. They should do an ad where they

44:11

have a diverse cast of people and that

44:13

diverse cast could talk about

44:16

like putting sustainable forestry around it.

44:19

All I gotta do is say something, something, landfills.

44:21

Everyone will love them again. They can charge. Absolutely.

44:23

Absolutely. Virtually signaling, right? They can genuinely. Yes,

44:27

absolutely. All right. Let's

44:29

keep the train moving here at the

44:31

AI landscape shifting yet again. Microsoft is

44:33

just another one of these shadow

44:36

acquires this time of inflection AI bizarre

44:39

deal. Microsoft has hired

44:41

most of the team at inflection

44:43

AI, including the CEO Mustafa, who

44:46

everybody knows. I just actually had him on

44:48

this weekend startups. He was the co-founder of

44:50

deep mind. He worked at Google for years,

44:52

and now he's going to be the CEO

44:55

of a new company called Microsoft AI.

44:58

It's essentially the consumer AI division of Microsoft,

45:00

but they did give him the CEO title

45:03

for background inflection have raised 1.5 billion

45:06

over the last two years. It was one of these

45:08

giant fundings that occurred

45:10

to build a foundational model. Like

45:13

open AI is doing like in prop is doing,

45:16

they had a chat bot called PI. Very similar

45:18

to chat GPT. It was supposed to remember

45:20

your history and build a relationship with you.

45:22

That's all getting shut down. Reed

45:25

Hoffman, who is a major investor in

45:27

this company and who's on the board

45:29

of Microsoft and who's so linked into

45:31

Microsoft, played an important

45:33

role in this deal according to reports

45:35

and the inflection investors included Bill Gates,

45:37

Eric Schmidt, and a bunch of other

45:40

interesting folks. But this was

45:42

the aqua hire, which is the weird thing. Trmup they

45:44

hired all the employees. They leave the shell of

45:47

a company. The company is going to go do

45:49

some enterprise stuff. And

45:51

investors get to keep their equity inflection, but

45:53

I guess that might be worthless.

45:56

There's something going on here that we don't know

45:58

about this deal structure. When you

46:00

saw this deal and you see Satya taking

46:04

the entire team, like he threatened to do with

46:06

opening eye, if you remember the same exact thing,

46:08

Aquire everybody, if you don't take the deal, what

46:10

is your take of what's going on here?

46:13

Why did this occur like this instead

46:15

of just buying the company? I

46:17

mean, it occurred like this because Reed and Bill

46:21

are inexorably tied to Microsoft, so they were able

46:23

to get a deal for investors that would have

46:25

never happened otherwise, and so good on them. I

46:27

think they did a very

46:30

good job protecting the fiduciary interests

46:32

of the

46:35

investors of this startup. Saks, what do you think?

46:37

No conflict, no interest. Yeah. I mean, we say

46:39

that a lot. This is so weird though. Why

46:41

not buy the company, Saks? Is

46:43

it because of antitrust? Actually, that's a

46:45

good theory. It's just as hard for

46:47

Microsoft to get anything through at this point. Probably,

46:50

they're just like, why even deal with the

46:52

antitrust? They don't really need the assets. I

46:55

think they licensed the core tech from

46:57

the Inflection C Corp, and then they

46:59

hire away all the talent, and then

47:01

the investors get made whole. I

47:03

think it's like an aqua hire

47:06

with a little bit of tech along

47:08

with it that they get through the licensing deal. Maybe that's just

47:10

to protect them from an IP

47:12

standpoint. My take on this is that this

47:14

was a bailout. This was a

47:16

bailout of the investors. I don't think the investors got ripped

47:19

off here. I think the investors were

47:21

thrilled to get their money back for whatever reason.

47:23

This company wasn't going anywhere, and it raised hundreds

47:25

of millions of dollars. Reed and

47:27

Bill obviously are wired in. They're on the

47:29

Microsoft board, and this company

47:31

did have some talent that Microsoft wanted.

47:34

They basically did a giant aqua hire, and

47:36

it bailed out the investors. Freiburg,

47:38

is that your take? Is it

47:40

a bailout, or do you think this is a

47:43

new interesting end run around antitrust where, hey,

47:45

if Adobe wants to buy the next

47:47

Figma, just buy the team, do

47:51

a license of Figma, some

47:53

Fakaka lunacy, or do you

47:55

think this is a bailout? What's your theory here, Freiburg?

47:58

What does it say about Microsoft? approach

48:00

to the AI space? I don't think this is

48:02

some run around anti-trust.

48:05

I think, obviously, like a lot of

48:07

companies, there's

48:09

been a realization on how

48:11

quickly foundational model development is

48:14

commoditizing and how quickly costs

48:16

are escalating and how many folks are

48:18

chasing it. So having some

48:20

unique advantage in the particular plane of

48:23

the market where they were

48:25

participating as a startup maybe

48:29

became difficult and untenable and negotiations

48:31

and conversations between all these parties who

48:33

all know each other very, very well

48:36

and are all very friendly, this ended

48:38

up being the best way to move forward. All

48:41

right, there you have it, folks. In other

48:43

related news, the Saudis are planning a $40

48:46

billion AI fund, according to the New York

48:48

Times. Reps from the Public

48:50

Investment Fund in Saudi have

48:53

spoken to a number of firms about partnering

48:56

on it. This would be the second largest

48:58

venture fund of all time behind SoftBank's $100

49:00

billion Vision Fund 1, which you

49:02

remember was also backed by the Saudis

49:04

and some other folks in the Middle

49:06

East region. This new fund would reportedly

49:08

invest in AI startups, ship companies, and

49:10

data centers. So

49:12

we thought we'd do a little quiz here.

49:14

If you were given the $40 billion, David

49:16

Sachs, how would you

49:18

allocate the $40 billion in AI

49:21

in today's market, 2024, going on? If

49:23

they put you in charge of this $40 billion

49:26

AI fund, where

49:28

would you deploy it? Same question would come

49:30

around the horn to you, Freeberg, and then you,

49:32

Chamor. Well,

49:34

the first thing is I wouldn't be in a

49:36

rush to deploy all $40 billion at once, because

49:38

that's a recipe for spraying a lot of money

49:40

into unproductive or overhyped things. So I would take

49:42

my time, first of all. But

49:45

second, in terms of the framework, I would think

49:47

about the different levels of the stack of AI

49:49

and try to figure out where the value cap

49:51

is going to be. And I think there's maybe

49:54

four different layers of the stack. First, you've got

49:56

the silicon. You've got the chips, where Nvidia is

49:58

dominant. Then you've got the

50:01

foundation models where it's open AI and

50:03

then there's some open source models. And

50:05

then you've got kind of infrastructure, dev tools,

50:07

vector databases, things like that. And

50:09

then finally you have the applications sitting on

50:11

top of that, which are just getting started.

50:13

I think it's really hard to know from what we're seeing

50:16

today exactly who's going to capture the most value in the

50:18

stack. I mean, you could make an argument

50:20

for or against pretty much any layer of the

50:22

stack. I guess if I had 40

50:24

billion to deploy, what I would do

50:26

is try to identify who are the leading companies

50:29

at each level of the stack and then who are the

50:31

most promising challengers. And I would

50:34

make a bet at every layer. So I'm

50:36

covered. That's not what I personally

50:38

do because I'm not, you know,

50:40

I'm just not a hardware investor. I'm not really an

50:43

infrastructure investor. I'm more of an application investor. So

50:45

I'm going to focus on that fourth layer

50:47

of the stack and distrust that there's going

50:49

to be enough value there. But again, if I was managing

50:52

a $40 billion sovereign wealth fund,

50:55

then I would play at every single level. And I

50:57

would hire the best people who know each

50:59

layer of that stack. SAC can do a

51:01

good job of sort of showing the

51:03

four layers of the stack. I

51:05

think open source, the application layer and

51:08

specifically robotics are huge opportunities that are

51:10

under invested in right now. So

51:12

I take the top open source projects. I'd

51:14

find those top contributors, take the top 20

51:16

or so open source projects and back them

51:18

to the tune of, you know, significant

51:21

money, 50 million, 100 million, whatever it takes

51:24

and try to own the top 20 or

51:26

have insights into those top 20 open source

51:28

models and own those teams. Look

51:30

for the top contributors, really easy to do

51:33

on GitHub and Hugging Face and Replit

51:35

and other places where they're active and

51:37

empower those. Then

51:39

obviously there's verticals. We're working on

51:42

people who are doing screenplays and tax just

51:44

like you are SACS and that's a

51:46

fantastic place to deploy money. But then I

51:48

think robotics is going to be super, super

51:51

compelling here. And that

51:53

takes a lot of money. And so

51:55

you do have the capital as a

51:57

weapon and the 20 year view of

51:59

this and making a general purpose

52:02

robot like Elon's doing with Optimus

52:04

or figure AIs doing with

52:06

their robot. I think robotics is a major

52:08

place. You can probably buy, I don't know

52:10

who owns Boston Dynamics now, I know Google

52:12

sold it. So you can probably buy

52:14

some robotic companies and then put the AI on them

52:17

and really take a 20-year view of

52:20

robotics. Freiburg, your thoughts?

52:22

How would you deploy the 40 million? The

52:24

bottom three stacks are very

52:26

difficult right now to find a

52:31

footing. Obviously, I think

52:34

the generally accepted framework for how to

52:36

think about how the market is

52:39

broken apart. But on

52:41

the application side, I think is where you could

52:43

think about finding more traditional

52:47

business model advantages. So

52:50

I think your point about robotics

52:52

is a really interesting one. Is

52:55

there an enterprise-like sale of robotics

52:57

tools that have positive ROI for

52:59

enterprises? And is there a business that's

53:01

working there and that is scalable? Vertical solutions

53:05

in labor and automation and productivity

53:07

gains are probably the best

53:09

sharp ratio. Good alpha,

53:11

lower beta and a

53:13

good place to kind of deploy. Competing in

53:15

chips, competing in infrastructure,

53:18

competing in models. As we

53:21

just talked about with respect to what happened with

53:23

Inflection and Microsoft this week, it's

53:25

such a rapidly changing environment. It's

53:28

hard to have high confidence in where things are headed

53:30

there. So I do think that

53:32

we do know that there are enterprise markets. We know

53:34

that there are segments of things like food, medicine,

53:38

manufacturing. These are markets that aren't going

53:40

anywhere. And they can all certainly

53:43

benefit from unlocks in

53:45

software or in robotics and automation and

53:47

hardware. So that's probably where I would think

53:50

about concentrating capital. But 40

53:53

billion is a lot of money. So you'd probably

53:55

think about deploying it over what period of time?

53:57

Is it 10 billion a year over four years?

54:00

And then is it broken apart in what way

54:02

over those sectors and over what geographies and you

54:05

know then find good managers to help you get it deployed.

54:09

Trim off, you're batting clean up here. You got to hear

54:11

everybody's answers before giving your

54:13

own. What are your thoughts on your other

54:15

besties answers and what's your plan to deploy

54:18

40 billion for the

54:21

kingdom of Saudi Arabia's public investment fund?

54:23

The most important thing that fund managers

54:25

get wrong is not having

54:27

appropriate reserves for your winners. And

54:30

if I look back, the real

54:33

profit dollars that I leaked was not

54:35

investing enough up front but when I didn't have

54:38

enough reserve to do the full

54:40

prorata or even super prorata and the ones that worked.

54:42

So the first thing you have to do and when

54:44

I do the regression on all of my funds, you

54:47

really need to reserve between 40 and

54:49

50 percent of the total fund size for

54:51

reserves. So take 20 billion off the table

54:53

and now you have a smaller problem

54:55

which is how do you deploy 20 billion because the

54:57

other 20 billion is purely meant for the winners where

55:00

you cram the money into the few that are winning so

55:02

that you can make the most money. Of

55:05

that 20 billion, I would

55:07

probably take two thirds of it and

55:10

I would go to all the hyperscalers and

55:13

anybody that's providing cloud compute and

55:15

essentially buy out all of the compute

55:18

credits on GPUs so

55:20

that I could tell any startup in the world, you

55:24

will do a safe with me,

55:26

the kingdom of Saudi Arabia. I

55:28

will give you free compute on

55:31

pick your cloud provider, pick your model, I

55:33

don't care. So if you want to run

55:35

llama on grok, great. If you want

55:37

to run Mistral on GCP, great.

55:39

If you want to run OpenAI

55:42

on OpenAI, fantastic. But

55:44

we will pay for all the compute. In return, we

55:46

get 7 percent up front and

55:48

you have to tell us some benchmark of

55:50

how this model is improving. And

55:54

then you are hiring a team of people whose job

55:56

it is and you can probably train a model to

55:58

do this as well. to

56:00

ingest the reporting in a systematic

56:03

way to understand if

56:05

it is, as you guys said, a robotics company,

56:07

there are measures of the quality

56:09

of a robot's dexterity

56:12

or vision or manipulation or

56:14

tax completion rate. If

56:16

it's a search engine, it's a different thing. If it's

56:18

a consumer-facing app, it could be users. If it's a

56:20

drug discovery app, it's the number

56:23

of legitimate protein synthesis passes

56:25

that pass, whatever it is.

56:29

Now you've scoped the problem to 20

56:32

billion is in reserves, 15 billion of it

56:34

is tied up in credits, and then the 5 billion,

56:36

I do think it's what David said, what SAC said,

56:38

which is a billion and

56:40

a half to the hardware, a billion and a half

56:42

to infrastructure, a billion and a half to some of

56:44

these discrete ideas. You

56:46

probably need a 30 or 40 person team

56:48

total, no more. But

56:51

that business could make a trillion dollars if it was

56:53

set up that way. But the key is to get

56:55

the credits out to people so that

56:58

the reality is what the commonality

57:01

across every single AI startup is

57:03

that they will have to run on a model, and

57:06

that model will be hosted somewhere in a cloud, and

57:08

all of that compute will need to be paid for. So

57:11

Saudi Arabia should pay for that compute,

57:13

get 6% or 7% upfront,

57:16

and retitle the safe, not

57:19

as the, what is the safe call, Jason?

57:22

Hello, agreement for future equity. This

57:25

should be the Saudi agreement for

57:27

future equity. I love it.

57:29

That's what they should be. All right, well, there you have

57:31

it, MBS, the all in KSA $40

57:33

billion fund. Give

57:36

us a call. We'll fire it up and we'll deploy

57:38

the $40 billion for you. Hey,

57:40

kid, maybe. Actually, I'm curious, you

57:42

guys, since everyone kind of seems to

57:45

agree with the framework of the four levels of

57:47

the stack, where do you guys think

57:49

the value cap is going to be? Do you think it's

57:51

going to be by the chip companies, the

57:53

foundation model companies, the infrastructure

57:56

companies, or the applications? I'll actually

57:58

change my answer. in a little

58:00

bit. I don't think it's necessarily new technology companies

58:03

per se that are going to

58:05

benefit most, but I think that it's businesses

58:07

in traditional industries that are going to

58:09

be able to leverage these tools, whether they're using open

58:11

source tools or third-party technology

58:13

capabilities that they're buying as a service

58:16

or as a piece of software. There's

58:18

an incredible set of advances that are

58:21

going to be realized in things like

58:23

chemical manufacturing, drugs, general manufacturing

58:25

that have existing scale and

58:27

infrastructure that are the fastest

58:29

to move to adopt these technologies are

58:32

going to benefit the most. I think that's

58:34

probably where most of the value will accrue

58:36

is not necessarily a chip company, but in

58:38

other businesses in traditional industries that

58:40

can transform themselves using

58:43

AI. So your answer is at

58:45

the application level. And then within

58:47

application level, you could divide it up

58:49

between existing companies and startups. Yeah. And

58:52

existing companies are able to leverage AI

58:55

innovations that are lower in the stack. Another

58:57

way to answer your question, Sacks, is

59:00

which spaces are the most crowded already?

59:02

And so obviously, chips are super crowded,

59:04

lots of incumbents, lots of people going

59:06

after it. And the foundation models are

59:08

very crowded as well. And the application

59:10

level is yet to sort

59:13

of be built out, right? Because we're in year zero or

59:15

year one of that. So I do think

59:17

application levels where there's a lot of value. But I

59:19

also think the hosting and

59:21

the development of these various niche models

59:23

is going to be a huge opportunity.

59:26

So just, once again, those open

59:28

source models being put into

59:30

cloud computing environments and optimized, that

59:33

could be a whole new level of

59:35

AWSs and Google Cloud and

59:38

Azure. So there's an opportunity for somebody to

59:40

compete with those incumbents by just

59:42

being better at those models that are open

59:44

source and having the team that's doing those

59:47

updates, and then optimizing that cloud. So I

59:49

would go with the cloud level for

59:51

those open source models and providing that as a

59:54

service like VMware does or other

59:56

folks do. WordPress will be a

59:58

tiny example of an open source project. like that. And

1:00:01

then I just something tells me this robotic space,

1:00:04

which has been a false start over and over

1:00:06

and over again. I think this

1:00:08

is the time where actually it's going to work. And

1:00:11

so I'm, I love that hardware robotics

1:00:13

space. I think you're right, Sacks. I think

1:00:15

it's kind of bookended in my mind. I

1:00:17

think the folks that are

1:00:19

building fundamental hardware will

1:00:21

make a lot of money over the next

1:00:23

five to 10 years. And then the folks

1:00:25

that are building the fundamental application level experiences

1:00:27

will make a lot of money as well.

1:00:29

And then who knows? The thing

1:00:32

that we're not talking about enough, and I don't

1:00:34

understand it well enough is the

1:00:37

entire way in which apps are built needs

1:00:39

to fundamentally be rethought

1:00:41

from first principles, because you essentially

1:00:43

basically have these client server

1:00:46

apps where you have all of

1:00:48

this business logic that's

1:00:50

sitting back in some server somewhere.

1:00:53

And I don't think that that's how apps

1:00:55

get built today in a world of

1:00:57

AI. And so it doesn't need to work that way. So

1:01:00

I don't exactly know what that means. But it just

1:01:02

seems like most of the architecture of how the internet

1:01:04

works today doesn't make much sense. So

1:01:06

that would then support free Brooks point that

1:01:08

all of the kind of the legacy things that

1:01:10

we all know and trust also get rebuilt.

1:01:13

And so that could create an entire wave where

1:01:15

networking value gets recreated and

1:01:17

everything. So my

1:01:20

gosh, I mean, it could be very, very transformation. I

1:01:23

mean, the idea that you would just talk to a

1:01:25

computer or in a chat interface, and you wouldn't use

1:01:27

an app interface, you would just ask questions that would

1:01:29

give you answers, ask your questions, it gives you answers,

1:01:31

give it a task, it gives you a result. This

1:01:34

is why this is why like Saudi needs

1:01:36

to basically make a bet that

1:01:38

there's just developers that are swarming around ideas,

1:01:40

let them show you

1:01:43

what's happening. Because paying 800 million

1:01:45

a year of fees is

1:01:47

not the best way to generate a return on $40

1:01:49

billion if this is where you want to be investing,

1:01:51

I don't think. Yeah, 800 million could pay for a

1:01:53

lot of credits, which probably gets you one

1:01:55

step closer to the To the Answer. It

1:02:00

hundred million in revenue. Company is worth

1:02:02

over. Eight billion

1:02:04

Dollars. In. Trading is a

1:02:06

fifty percent where do you think term

1:02:08

offices? The opening of the ip our

1:02:10

window. again someone off the tribe coming

1:02:12

in from what are you take from,

1:02:15

it's popping fifty percent on the first

1:02:17

day. I. Don't

1:02:19

know. I. Really haven't looked

1:02:21

at the company, I haven't looked at the

1:02:23

the financials. But

1:02:26

like. This last three or four weeks I

1:02:28

think people have been so excited and rearing

1:02:30

to go. Were.

1:02:32

In the middle of another mean coin craze.

1:02:35

I don't like it. I just think if we are

1:02:37

also pulled into that nonsense. I think the

1:02:40

point is that I think there is a. Speculative.

1:02:43

Party going on right now and.

1:02:46

I'm not saying that. read it as part of. But. whenever.

1:02:49

These things are so Miss Price. Just

1:02:52

means of people are are ready to gamble little bit.

1:02:55

Their stuff A lot of gambling in the system

1:02:57

as we've seen and were recording this on Thursday,

1:02:59

so who knows what the stock price will be

1:03:01

tomorrow. Generally I mean to get power

1:03:03

source get. The other thing is I. Our

1:03:06

did say finally, it looks pretty likely we're

1:03:08

going to get these three cats. So. We're.

1:03:11

Going to be down to four and a

1:03:13

half to four, seven, five, On.

1:03:15

Said son's by the end of the year. It.

1:03:18

Probably means that will get another fifty to

1:03:20

seventy five basis. points to twenty twenty five

1:03:22

so people will look out. To

1:03:24

the under twenty twenty five and look at it said

1:03:26

funds rate that sort of like three seven five to

1:03:28

four percent. So they're getting pretty excited and. Frost.

1:03:32

Done. So. This is the

1:03:34

beginning of the beginning in terms of that kind

1:03:36

of speculation. Them So Crypto

1:03:38

Go bonkers. Last couple of weeks. Yeah,

1:03:41

but what's your take on all the sachs? Is

1:03:43

this feel like. People. Are.

1:03:46

In i gamble mode again and

1:03:48

de sac concerned? You guess after

1:03:50

what we just experienced. Well.

1:03:53

i think their pricing in these right

1:03:55

thoughts as if they're doubling of happen

1:03:57

by my side i think the x

1:03:59

markets expectation is with,

1:04:01

let's call it 70% certainty that you're

1:04:03

going to get three rate cuts this year. And

1:04:06

it still seems to me that that's a little bit

1:04:08

up in the air because

1:04:10

inflation has not come down to the Fed's

1:04:12

target of 2%. It's still right around 3%,

1:04:15

3.1%. So it's been

1:04:18

sticky around 3%. It

1:04:20

was coming down pretty fast last year, but it's

1:04:22

still there. So people I

1:04:24

think are maybe gambling in the sense that they're

1:04:26

counting on rate cuts that haven't happened yet, but

1:04:29

they liked Powell's dovish

1:04:32

rhetoric yesterday. And they

1:04:34

seem to think that that means that the rate

1:04:37

cuts are definitely coming. I

1:04:39

don't know how Powell can promise that without inflation

1:04:41

coming down unless he's trying to create

1:04:44

a Biden put. Hmm, there it

1:04:46

is, folks. Which was my theory going into

1:04:48

this year is that, you know, if

1:04:51

you took the Biden put from the Fed. Absolutely.

1:04:53

And it's looking like Biden has a

1:04:56

clear path to victory based on the

1:04:58

lowest unemployment of our lifetimes and this

1:05:00

roaring stock market records. Yeah,

1:05:03

so easy Biden victory. Freeberg,

1:05:05

what do you think about

1:05:08

the risk on? I

1:05:11

don't know. Biden is incredibly

1:05:13

unpopular. If you look at his actual polling,

1:05:15

he should not win the election. But

1:05:17

that's in spite of, you're right, the economy is seems

1:05:20

to be doing pretty well. The stock market is certainly

1:05:22

doing pretty well. A president

1:05:25

with these fundamentals would

1:05:27

normally get reelected. But Biden currently is not

1:05:29

his polling is not the polling of a

1:05:32

president who's going to get reelected. It is

1:05:34

pretty amazing. If Powell comes through

1:05:36

with the rate cuts, that will help him. It's

1:05:39

pretty amazing how horrible a job he's doing that

1:05:41

he's made Trump look like a better candidate to

1:05:43

some people. That is an extraordinary achievement by Biden

1:05:46

and we'll have to agree there. Look,

1:05:49

I think the bigger economic issue we should be

1:05:52

talking about was that

1:05:54

buried article in the Wall Street Journal

1:05:56

about commercial real estate. Do

1:05:58

you see that article? Crazy! So

1:06:00

this was your journal. reported that of

1:06:03

the thirty six billion of office loans

1:06:05

that came, do. In the

1:06:07

commercial mortgage backed securities market last

1:06:10

year. Only. One quarter were paid

1:06:12

off in full. That's.

1:06:14

According to data from real see alex from

1:06:16

Credit Ten. Is that extended

1:06:18

pretend Travis You that you see their oh

1:06:20

yeah mean only One quarter. Of

1:06:22

the loans got paid off. For thirty

1:06:25

five percent didn't get paid off in full. So

1:06:27

there's a hunter buildings that are

1:06:29

underwater right now, but nobody wants

1:06:31

to. Foreclose. And then

1:06:33

sell them at a fire sale price. So.

1:06:36

Yeah, everyone is his paternal accent.

1:06:38

Yeah, Sigma can down the road to be wants

1:06:40

to take the medicine so humans are hoping they're

1:06:42

hoping that rates will come down fast enough and

1:06:44

occupancy rates will go up. Fast. Enough

1:06:47

that no one has to. Foreclose.

1:06:49

But. If rates don't come down. Then.

1:06:52

You have a real problem. Me Again, everyone is

1:06:54

just assuming. That. Inflation will

1:06:56

continue. It's path from three percent. Two

1:06:58

percent of them will get these rate

1:07:00

outs and they're hoping they can cling

1:07:02

on. Landlords are. Clinging. On I

1:07:04

think the banks that one him all this money. Or

1:07:07

play on for dear life. But.

1:07:09

If rates they. Fire.

1:07:11

Longer. Than you're to

1:07:13

see some real distress. Signal

1:07:15

to his original banking system Freiburg you had

1:07:17

a a science can hear. couple of different

1:07:20

things came up the Hubble telescope. Pretty.

1:07:22

Printed. Organs.

1:07:25

What would you want to go with? this? A. Media

1:07:27

going to talk about. That. How the universe

1:07:29

is expanding at an accelerating rate and. Expanding

1:07:32

differentially everywhere we look and we don't

1:07:34

understand how or why. I hundred

1:07:36

percent in Saxony to use the lives are yeah.

1:07:38

definitely. He

1:07:43

drives love Side quarter lead now is

1:07:45

free markets your time to size to

1:07:47

to shine All the social science alice

1:07:49

what's going on have been of my

1:07:52

favorite topic. Astrophysics,

1:07:54

I remember the she might have some back of

1:07:57

the sexes that are we headed for your anus.

1:08:00

Dead set on it. We're dead. It's that collision

1:08:02

course over the vein is he disappeared

1:08:04

again It's just gonna drop in and do a

1:08:06

new readiness joke. I hear him giggling with his

1:08:09

camera off. He's with his writers He's a teacher

1:08:11

to the writers room to get an update on

1:08:13

the Uranus jokes It's long

1:08:15

been known that our

1:08:18

universe is expanding we've all heard about

1:08:20

the Big Bang Theory and Couple

1:08:26

decades ago scientists began to

1:08:28

observe the

1:08:30

brightness of supernovae

1:08:33

or exploding stars and

1:08:36

Then by looking at their brightness We could

1:08:38

tell how far away they were and then

1:08:40

looking at the shift of the light whether it's

1:08:42

getting red or blue you can tell whether the

1:08:44

light the the supernovae is moving away

1:08:46

from us or towards us and how

1:08:49

quickly and so what

1:08:51

scientists have Used

1:08:53

those observations to deduce several

1:08:56

decades ago and for which a group won the

1:08:58

Nobel Prize is that our

1:09:00

universe is Expanding

1:09:04

Meaning that everything is moving away from itself. It's

1:09:06

almost as if we're all in a cake the

1:09:08

best analogy I've heard is that there's raisins in the

1:09:10

cake and as the cake Expands

1:09:13

in the oven all of those

1:09:15

raisins look around and they're all moving apart from

1:09:17

each other. So the distance between everything is Getting

1:09:21

wider space is expanding

1:09:23

So we can actually throw a moment of the Big Bang That's

1:09:27

the theory is right This all started in the Big Bang

1:09:29

and it's not like you know We're all

1:09:31

in the middle of the universe and the universe is moving away

1:09:33

from us It's that the entire entirety

1:09:35

of space is expanding itself And so

1:09:37

everything is moving away from everything else

1:09:39

the challenge in the data was that

1:09:41

recently we saw the different types of

1:09:43

stars That

1:09:45

were observed with the Hubble Space Telescope Had

1:09:49

actually a different Rate

1:09:52

of expansion of the universe than what was

1:09:54

previously thought from the supernova that we saw

1:09:56

and everyone thought there was something wrong with

1:09:58

the Hubble Space Telescope data So,

1:10:00

you know, in a final print published

1:10:03

last month, the

1:10:05

James Webb Space Telescope, which

1:10:08

showed much higher resolution of imaging and

1:10:10

as a result much better data. And

1:10:13

here's an image of it. You can just see the difference between

1:10:15

what came out of James Webb and

1:10:17

what was used in Hubble. So

1:10:19

that shows one star that was used to calculate the

1:10:22

rate of expansion of the universe. And

1:10:25

what this star showed is that the rate

1:10:27

of expansion of the universe

1:10:29

as measured across a thousand of these

1:10:31

stars called cepheid stars in five different

1:10:33

galaxies had a value that didn't

1:10:35

match the value that we see when we look

1:10:37

at exploding stars very far away.

1:10:40

And so it once again confirms what

1:10:42

the original Hubble data showed, which is

1:10:44

that the universe is expanding at

1:10:47

different rates in different parts

1:10:49

of the universe, which totally doesn't make sense.

1:10:52

And there's no really clear answer as to the physics

1:10:54

of why. If there was a big bang and

1:10:56

the universe started to expand, it should be

1:10:58

expanding the same way everywhere. There's no reason

1:11:00

that different parts of the universe should be

1:11:02

expanding at a different rate than other parts

1:11:04

of the universe, that there's differential expansion happening

1:11:07

across the universe. And it really

1:11:09

tells us that there's very little we know

1:11:11

or understand about this large-scale structure of the

1:11:13

universe and why these things are happening. There

1:11:15

was one group that put out a theory

1:11:18

where they said that the local universe

1:11:20

where we see these cepheid stars that

1:11:23

this data is based on, there's almost like a little

1:11:25

bubble. Outside of the bubble, there's

1:11:27

galaxies and they're pulling stuff out faster than

1:11:29

what's happening where there's not as much density.

1:11:31

So there's a lot of different reasons and

1:11:34

explanations for why this might be happening. But

1:11:37

it's kind of a big story because it

1:11:39

basically confirms that we see a lot of

1:11:41

differential expansion happening across the universe and we

1:11:43

don't know why. Just

1:11:45

another big mystery of the universe. So super

1:11:47

kind of interesting confirmatory data from James Webb

1:11:49

this week. Joao, do you want

1:11:51

to talk about the pig kidney? I

1:11:55

mean, I think it's really incredible. They

1:11:57

took a kidney from

1:11:59

a pig. pig and then they use CRISPR-Cas9

1:12:02

to edit out certain

1:12:05

subset of the pig genes and edit in

1:12:07

a certain useful subset of human genes and

1:12:09

then they transplanted that kidney into a 64-year-old

1:12:11

man. And

1:12:15

he apparently seems to be doing well. It's

1:12:17

really incredible. And this is a guy that

1:12:20

was at end-stage kidney disease and

1:12:23

my father went through this. So when you're at that

1:12:26

end-stage renal disease, you're

1:12:28

getting dialyzed every three days for

1:12:30

up to four, five, six

1:12:33

hours trying to basically

1:12:35

have a machine that does the job of the kidneys for you

1:12:37

to keep your blood clean. But

1:12:39

it's just kind of a death sentence and it's like a

1:12:42

slow-moving death sentence at that. So

1:12:45

the idea that you could now use

1:12:48

this effectively infinite source

1:12:50

of organs

1:12:53

seems really important. It's

1:12:55

really compelling. Pretty

1:12:58

amazing. I don't know. What do you think? Definitely

1:13:01

where we're headed. I don't even think they're going

1:13:03

to need to. Yeah. Speaking

1:13:06

of medical breakthroughs, did you

1:13:08

guys see the guy

1:13:10

who got the neurolink chip? I think I'm

1:13:12

going to check with his brain. He's using the

1:13:14

force. Yeah. That was awesome.

1:13:16

He's using the force. Yeah, he described it as

1:13:18

he was playing chess and he said like, I

1:13:20

love playing chess and I've learned to use my

1:13:22

mind to move pieces around and he

1:13:25

said like, it's like having the force. Totally.

1:13:28

I mean, I've watched the live stream and he

1:13:30

said that he was a quadriplegic from the shoulders

1:13:32

down but he loves chess.

1:13:35

He wasn't able to play chess. Now he's able to play

1:13:37

chess. He was able to play Civilization, the game, which I

1:13:39

guess was really hard in

1:13:42

the absence of having this chip and now he can do

1:13:44

that. It's incredible. I

1:13:46

mean, it's really incredible. It's so cool that they kind

1:13:49

of went public with it. I mean, obviously he agreed

1:13:51

to go public with it. I think Trevor

1:13:53

is his name. It's pretty amazing

1:13:55

and there's more amazing stuff coming.

1:13:57

They're making significant progress over there.

1:14:00

talking to somebody who works there and you

1:14:02

know it's it's it's gonna be yeah

1:14:08

it's gonna be a process but they're

1:14:10

making significant progress and they're gonna have

1:14:13

I think even bigger announcements in

1:14:15

the coming year they're getting it done. Yeah

1:14:18

my kids that were really blown

1:14:20

away by it you know it's it's

1:14:22

interesting to me what like penetrates their

1:14:24

cynicism bubble and like what they get

1:14:26

actually impressed or excited by and

1:14:28

they thought this was really cool. You

1:14:31

know so funny it's like people are breathlessly

1:14:33

like looking at the Elon

1:14:35

Don Lemon video and they're

1:14:37

trying to opine whether he had a good interview

1:14:39

or a bad interview and it's like he actually

1:14:41

had one of the best weeks of his life.

1:14:45

Yeah Starship 3 made it out. Yeah.

1:14:47

Starship had an incredible performance last week

1:14:49

and then this thing this

1:14:52

week I mean what an incredible seven-day

1:14:54

run it's more than most of us

1:14:56

will have in our lifetime. It's also

1:14:58

nuts. That down-lum an interview like there

1:15:00

was actually some really interesting topics discussed

1:15:02

but I felt like this was like

1:15:04

this was the perfect like baton

1:15:07

passing between old media and new

1:15:09

media where he was constantly trying

1:15:11

to do a gotcha journalism thing

1:15:13

and Elon was trying to engage in good

1:15:15

faith and like an interesting discussion and then

1:15:18

every time they discussion that interesting he

1:15:20

kept trying to go back to like sensational gotcha

1:15:22

journalism and it was a totally wasted opportunity

1:15:24

if you just don't let me just stayed

1:15:26

in the moment and like engaged

1:15:28

in good faith instead of trying to get this

1:15:31

like clip it was it really

1:15:33

bad performance by Don Lemon I thought it

1:15:36

was really getting interesting at moments in time. He's

1:15:38

obviously trying to get a job back at some

1:15:40

mainstream media company. Maybe. This

1:15:44

is his resume. If he can basically get

1:15:46

some clip of Elon saying something

1:15:48

that he can twist into something that it doesn't

1:15:50

mean and he can say I got Elon

1:15:52

Musk then he can get his job back at CNN

1:15:55

or one of them. He got paid out in

1:15:57

full like 20-30 million bucks and by the way

1:15:59

though those jobs are not, they don't exist

1:16:01

anymore. Like Tucker and him were like the

1:16:03

last of that generation to

1:16:06

get those huge pay days. That's over.

1:16:08

Don Lemon got paid $30 million. Like,

1:16:10

hmm. I think CNN,

1:16:12

the report was CNN paid out his

1:16:14

full contract. The same thing happened,

1:16:16

I think with Tucker, they both got paid their

1:16:19

full contracts. Whoever negotiates those contracts with the

1:16:21

networks over the last couple of decades is

1:16:24

awesome because it's pay for play.

1:16:26

Like, whether you are on air or

1:16:28

not, you get paid. And

1:16:30

then they don't let you go to a competitor. That's

1:16:33

why both of them are doing internet stuff, X

1:16:36

or whatever, carved out stuff. Just wanted to give a shout

1:16:38

out here at the end of the pod for

1:16:40

the guys at all in talk. Chris and

1:16:42

Spencer did an amazing job, the team over

1:16:44

there from Good Future Media of creating

1:16:47

just that amazing fan account. Now that we're

1:16:49

growing up here, we're all in and we'll

1:16:51

have a CEO announcement at some point. We

1:16:54

have taken over the accounts and we're starting

1:16:56

to make our own clips. So thanks to

1:16:58

them for supporting us and helping grow it.

1:17:01

If you want to see the four of

1:17:03

us, go to YouTube and type

1:17:05

in all in podcast and then subscribe. We're

1:17:08

almost at 500,000 subscribers. And

1:17:10

we decided that if we hit 1 million subscribers by

1:17:12

the end of the year, we're gonna throw a party

1:17:15

and y'all will be invited. Love

1:17:18

as many people at the party as possible. So

1:17:21

go ahead and find us on TikTok, LinkedIn, and

1:17:23

Instagram. Just do a search for all in and

1:17:25

you can search for all of us. Chamath, David

1:17:28

Friedberg, David Sachs, and Jason Halikhanis.

1:17:30

Thanks again to the super fans

1:17:32

for helping us build up. Amazing

1:17:35

audience. Couldn't have done it

1:17:37

without you. And yeah, we're just posting clips,

1:17:39

I think, every other day on

1:17:41

the YouTube channel. And so if you

1:17:43

go in there, you can subscribe and then right next to

1:17:45

it, there's a little alarm bell. You hit the bell, you'll

1:17:47

get an alert and you get to see our faces and

1:17:49

know what we look like. We also put

1:17:51

a lot of graphics. So if you hear us

1:17:53

referring to charts, that's all on the YouTube channel.

1:17:56

For David Sachs, I'm Paul Palapatiya, David

1:17:58

Friedberg, I am. Jason

1:18:00

Calacanis and we will see

1:18:02

you boy next month as moderator you forgot your

1:18:04

timeline and the host of this week in startups

1:18:07

So search for my pocket this week startups when

1:18:09

you review startup I'm just gonna get plug in

1:18:11

for that. No, some people don't know I have

1:18:13

another podcast So please search for this week in

1:18:15

startups and that as well. See you all next

1:18:17

time. Bye. Bye I

1:19:00

Be one of your feet