back to indexE108: Doxing debate, Nuclear fusion breakthrough, state of the markets & more
Chapters
0:0 Jason's new gig!
1:5 Twitter's new privacy rules, notable suspension, doxing dynamics
20:48 Nuclear fusion breakthrough and geopolitical ramifications
42:58 Jason and Sacks's big night
51:11 State of the markets: Coupa acquired by Thoma Bravo, startups at all stages seeing heavy valuation reductions, what LPs are thinking
82:8 TV catch up, worst person in tech bracket
00:00:00.000 |
Captain Calacanis is here reporting for duty. 00:00:06.800 |
I just want to say all in podcast now sponsored by Moncler 00:00:11.000 |
and Spirit Airlines or now sponsored by the village people. 00:00:22.000 |
I don't think it's thin enough to be a flight attendant. 00:00:30.700 |
Like fat shaming would be number 72 on the list. 00:00:33.800 |
He can't get canceled because all the libs have left Twitter. 00:00:36.700 |
There's nobody to there's no home monitors left. 00:00:43.700 |
You said they moved to Canada when Trump got elected. 00:00:47.800 |
A hacker figured out a way to take all this data and track, 00:01:20.000 |
The fact that the law allows for people to do this persistent 00:01:27.100 |
tracking of planes, which then becomes persistent tracking of 00:01:31.400 |
a person and what really is at stake here is how we define the 00:01:35.000 |
term doxing for people who don't know the term doxing. 00:01:37.100 |
It means giving a person's location that could be your home. 00:01:39.500 |
That could also be your at a location for some period of time. 00:01:41.900 |
You're you're at a hotel, you know for a basketball game and 00:01:47.000 |
You can take a picture of a celebrity say there's a celebrity 00:01:53.000 |
We all understand doxing is dangerous and it in fact is against 00:01:56.600 |
the law to just give people's addresses and stuff like that. 00:01:59.200 |
The issue here is a new type of doxing which I'll call, you know, 00:02:05.300 |
persistent coordinated doxing where dozens of times a month, 00:02:10.300 |
It may not be against the first amendment sacks. 00:02:14.500 |
I think you would agree but we have to ask ourselves do we want 00:02:17.700 |
to live in a world where whether a person's on an electric bicycle 00:02:21.000 |
or an airplane or any device in between somebody should be 00:02:24.200 |
releasing dozens of times a month a specific dedicated feet of 00:02:29.000 |
It is terrorizing as a parent when this happens. 00:02:36.900 |
We don't want to live in a world with de facto doxing. 00:02:40.700 |
What these sites were doing was de facto doxing. 00:02:43.200 |
I think it was a bad decision and I think that it represented 00:02:49.000 |
you on the least generous statement would be that it represents 00:02:53.500 |
deep hypocrisy in that not just a few weeks ago. 00:02:56.600 |
Did he say he would never delete that account? 00:02:58.400 |
But he also said he was buying Twitter to enable freedom of speech 00:03:02.400 |
and freedom of expression and that he wouldn't come in and do the 00:03:04.700 |
same sort of content moderation that was done by the old regime. 00:03:07.700 |
And then he came in and did exactly what the old regime did, which 00:03:11.400 |
is that he took the rules and he took the quote moderation policies 00:03:15.000 |
and he found a way to use them to make some editorialized decisions 00:03:19.900 |
Now, the more generous thing is what you guys are saying, which I 00:03:23.500 |
don't think is necessarily wrong, which is that he's trying to 00:03:26.000 |
protect people where there's some loophole or some law that doesn't 00:03:29.900 |
seem right morally, but it is the law and it is what it is. 00:03:33.100 |
In those cases, I think you run into the exact same issue that the 00:03:38.300 |
old guard at Twitter had that the moderators and the executives at 00:03:41.900 |
YouTube have dealt with and that the executives at Google have dealt 00:03:44.600 |
with and that we sit here and we criticize until you're on that 00:03:47.600 |
side of the table and you're forced to make these moderation 00:03:51.000 |
You're forced to make these policy decisions and you're forced to 00:03:53.300 |
implement these policy decisions because of some moral framework 00:03:59.700 |
Some people will say that's not freedom of expression. 00:04:05.300 |
You're taking this particular case away from a 15 or 16 year old 00:04:11.000 |
And so I think what it shows is just how hard it is to moderate 00:04:15.100 |
these sites, these platforms, and that there is no simple, easy, 00:04:18.600 |
idealistic ideologue of, hey, all these things are open. 00:04:22.500 |
All these things can be used by anyone all the time, because as 00:04:25.200 |
soon as one of these edge cases start to happen, you want to come 00:04:30.700 |
So I have had these issues happen to me multiple times. 00:04:35.300 |
I'm not nearly as important as Elon is, but it feels the same 00:04:44.700 |
That being said, I think the real decision for somebody like me 00:04:50.400 |
is that if it's too much is frankly just to get rid of it and 00:04:54.200 |
you know to find a different mode of transportation that's a 00:04:59.500 |
And the reason I say that is that I just think that you would 00:05:02.400 |
have to go and get the government to basically change the law 00:05:06.100 |
And so then as a result, your reaction will seem somewhat 00:05:14.800 |
Let me just summarize this and be the first one to just state 00:05:17.400 |
this. I think that if there's any person in the world that can 00:05:24.400 |
But man, has he taken on just a gargantuan battle and increasingly 00:05:31.600 |
I am not a fan of this battle and I'll tell you why this is a 00:05:35.200 |
man who is essentially proven that he can bend the laws of 00:05:41.300 |
He's done it twice once in electric cars and once in rocketry. 00:05:45.900 |
The problem is that the realm of decision-making at Twitter 00:05:49.400 |
has nothing to do with the laws of physics and is governed by 00:05:53.000 |
emotions and psychology in which there is no canonically 00:05:58.500 |
And so he's quickly finding out that half the population will 00:06:01.800 |
always find fault with him no matter what he does. 00:06:04.900 |
And now the implication of that becomes very important. 00:06:09.100 |
We saw yesterday that he had to sell another 3.8 billion dollars 00:06:16.300 |
It's because this transaction which was, you know, very tight 00:06:20.800 |
to get done probably required lots of margin, you know, there 00:06:24.800 |
are I look I have a margin loan at Credit Suisse. 00:06:27.800 |
So I know how these things work and you can very quickly get 00:06:31.600 |
You have to sell down things that you own in order to maintain 00:06:37.000 |
He's had to do this twice now in the last few weeks and that's 00:06:40.000 |
because again not because of the demand at Tesla as far as 00:06:43.600 |
we can tell but because people believe he's distracted and 00:06:47.100 |
so people are anticipating weakness at Tesla. 00:06:59.900 |
Is it just putting himself under an enormous amount of pressure 00:07:07.100 |
Yes, and I think that this is sort of where we're at six weeks 00:07:13.900 |
I mean, I was saying this guy learned in six weeks what it 00:07:17.700 |
took YouTube seven years to learn how hard it is to moderate 00:07:20.800 |
content and you know, I think this is where I disagree is 00:07:24.600 |
you're you're attributing so much good faith to these content 00:07:29.500 |
moderators at YouTube and Twitter when the Twitter files 00:07:32.900 |
reveal that they made no effort to suppress their bias. 00:07:37.100 |
In fact, they were like pretty much dancing in the streets 00:07:43.500 |
every time they booted off someone they didn't like. 00:07:46.100 |
Before you react to what Freeberg just said at the end that 00:07:50.600 |
Like it's like well look, I mean if you define what Elon 00:07:54.400 |
is, you know doing there as you know, acting as a judge 00:08:00.200 |
arbitrating on every little content moderation decision. 00:08:03.700 |
Is that a great use of his time relative to what he could be 00:08:06.400 |
building at Tesla and SpaceX and doing on behalf of humanity 00:08:10.900 |
But if you define what he's doing in the larger sense as 00:08:13.700 |
restoring free speech to the most important Town Square 00:08:17.800 |
social network, hopefully thereby inspiring other tech 00:08:21.700 |
companies to move in the direction of opening things up. 00:08:24.900 |
Then I actually think it's a pretty good use of his time. 00:08:27.000 |
So look, I think we can quibble about this or that decision 00:08:30.500 |
that he makes or this or that tweet, but I think the overall 00:08:34.100 |
thrust of what he's doing is very important for the country 00:08:41.100 |
Hopefully he'll find some people at Twitter who he can empower 00:08:44.900 |
and trust to make these content moderation decisions. 00:08:47.400 |
So he's not drawn into every single little battle, right? 00:08:50.500 |
We do want him focused on the highest priority problems. 00:08:55.500 |
I just think that what he's learning and what we're living 00:08:58.800 |
and seeing in real time is that there is no canonically right 00:09:04.900 |
There's only a decision where some percentage will support 00:09:11.800 |
He did say when he took over he knew that would be the case. 00:09:14.500 |
He said you will know I'm doing the good job when both sides 00:09:20.400 |
I think it's important for people to understand what the new 00:09:22.900 |
So I'm just going to quickly read it just to hang on. 00:09:26.600 |
Sorry, hang on one sec before you get to because I think the 00:09:28.900 |
philosophical point rather than the specific one is an important 00:09:32.400 |
one and I just want to respond to what you said and have sacks 00:09:35.800 |
respond to this in the case of the points you make around the 00:09:39.500 |
And by the way, I don't agree with any of the moderation 00:09:43.500 |
So I don't think that someone should be suspended for posting 00:09:48.500 |
I don't think someone should be suspended for saying controversial 00:09:52.200 |
Just so I'm clear on that because I know that you know that 00:09:55.000 |
we describe yourself as a libertarian speech. 00:10:00.800 |
And and so in this particular case, I think what really irked 00:10:05.400 |
me I was trying to identify why it made me so angry yesterday. 00:10:09.900 |
And I think the reason was that in the case of the Twitter file 00:10:17.000 |
It was one person that was affected because the majority wanted 00:10:22.400 |
And I think in this case, it's that the minority wants to affect 00:10:26.900 |
the majority in the sense that Elon has aggregated this control 00:10:30.600 |
and this power over moderation and he's benefiting himself and 00:10:34.300 |
a few people that have private planes and he's shutting down 00:10:37.400 |
hundreds of tweet feeds Twitter feeds that are using publicly 00:10:41.800 |
And so it feels even more onerous of a use of power and influence 00:10:48.300 |
because he's taking he's doing something that benefits a small 00:10:54.500 |
Whereas the alt whereas the other one was affecting a small 00:10:57.800 |
number that benefited a large number because that's what a lot 00:11:01.900 |
A lot of people wanted to see Trump suspended and it wasn't 00:11:05.200 |
Okay, I don't know if that if that makes sense. 00:11:06.900 |
Yeah, we understand your position completely. 00:11:11.800 |
I think it's very important to understand what he is saying 00:11:14.100 |
about this accounts dedicated to sharing someone else's live 00:11:17.500 |
location are going to be suspended going forward. 00:11:19.500 |
You can still share your own location obviously content 00:11:22.600 |
required, you know content for public engagements, you know, 00:11:26.300 |
the president is speaking somewhere whatever you just really 00:11:28.900 |
can't be persistently consistently tracking an individual 00:11:32.700 |
otherwise known in you know, sorry, Jason, if any are talking 00:11:36.900 |
but Jason if NPR is live tweeting sure Jerome Powell speech 00:11:42.100 |
perfect problem XYZ location not a problem if they do Jerome 00:11:46.700 |
Powell's location for the next year for the next year 10 times 00:11:56.800 |
I'm just saying like let's just say he gives a speech every 00:12:01.200 |
No, if you're really funny if you're giving a speech at a 00:12:04.000 |
public place where you've announced that you're going to 00:12:06.600 |
be appearing at a certain time and place you've already made 00:12:09.000 |
a problem where you're going to be what we're talking about 00:12:11.700 |
is and what what what Elon Jett was showing was a live stream 00:12:15.300 |
of precision GPS coordinates over a sustained period of time. 00:12:21.600 |
But if you look at like the weapons are so successfully being 00:12:24.400 |
used in Ukraine right now, they're all precision GPS guided 00:12:27.500 |
now right now you have to be a state actor to get a hold of 00:12:29.700 |
those weapons, but you could imagine over the next decade 00:12:33.200 |
that having someone's precise GPS coordinates over a sustained 00:12:37.900 |
It would be pretty easy to target them for and not to be 00:12:47.100 |
I'm scared that Duke come at me anytime when I get my jet. 00:12:58.000 |
I'll be on my first flight and he's just going to send a drone 00:13:00.800 |
in but look, let's talk about hypocrisy for a second. 00:13:04.900 |
Let's let's talk about CNN's hypocrisy in the media's hypocrisy 00:13:08.300 |
because earlier in the week they were saying that any criticism 00:13:11.800 |
of Yoel Roth who is Twitter's former head of trust and safety 00:13:15.800 |
amounted to a threat to his safety and they had this like 00:13:19.400 |
theatrical tweet where they claimed he was having to flee 00:13:22.000 |
his house, which a lot of people found pretty preposterous. 00:13:25.600 |
They were basically saying that public criticism of someone 00:13:29.400 |
who has put themselves out there to engage in a public debate 00:13:35.400 |
However, publishing someone's real-time location on a continuous 00:13:45.200 |
If one of those two things is a threat to safety, it's the real-time 00:13:49.900 |
I think we now understand why Elon did what he did. 00:13:52.600 |
He basically had an incident in LA in which the safety of his 00:13:55.800 |
kid was threatened because he's got stalkers coming after him. 00:14:03.400 |
Yeah, why should his personal experience affect the usage of 00:14:07.600 |
the service that hundreds of millions of people use and that's 00:14:10.700 |
The decision should not be based on what affects him personally. 00:14:15.100 |
Yeah, it's be a principle basis for any decision about content 00:14:19.800 |
Maybe in the first few hours that decision it wasn't handled 00:14:22.600 |
perfectly because there wasn't a principle basis. 00:14:27.200 |
The principle basis is what J Cal showed and this applies to 00:14:30.900 |
And so, you know now it's a debate about whether that policy 00:14:34.800 |
Now is Elon just as arbitrary and capricious as the former 00:14:38.800 |
executives who are running trust and safety at Twitter. 00:14:46.200 |
He said that when we ban or shadow ban an account there has 00:14:49.300 |
to be a reason for it and you have to be alerted to it. 00:14:57.300 |
It's there right no more shadow that is different. 00:14:59.800 |
And then the second thing is that and again, I think you 00:15:02.200 |
could say they didn't do this perfectly in the first few 00:15:04.400 |
hours, but there needs to be a principle basis for a censorship 00:15:07.300 |
decision and it needs to be applied to everyone equally. 00:15:10.000 |
And so far we haven't seen any basis for believing that he's 00:15:15.900 |
Whereas the former rulers at Twitter were indulging their 00:15:19.400 |
personal bias and personal preferences and who were they 00:15:24.900 |
If you were someone who is allied with them, it was almost 00:15:27.300 |
impossible to get censored no matter how hateful your tweets 00:15:31.000 |
were. But if you were somebody on the other side of the 00:15:33.000 |
political debate, they were eager to suppress you. 00:15:35.200 |
And I think that at least so far Elon has not shown that type 00:15:40.800 |
He selected against someone that put him at personal risk. 00:15:47.800 |
Yeah, if that's where the decision had stayed, yeah, then 00:15:53.000 |
I would agree with you, but I think that since then they've 00:15:55.900 |
put in place they've undergirded that decision. 00:15:59.800 |
I think those sites are I think those tweet streams are cool. 00:16:03.000 |
I think there's some cool tweets streams that some of these 00:16:05.500 |
people run and there are hundreds of them and they're actually 00:16:08.100 |
You can see where these different labor of people tracking 00:16:10.500 |
people's planes like where Air Force One is they show all these 00:16:12.700 |
different planes it look and whether or not the FAA should 00:16:15.200 |
be publishing the state as a separate question, but it's on 00:16:19.900 |
It's like turning off the RSS feed from the open internet 00:16:25.500 |
So here's the part I agree with which is I think this policy 00:16:29.000 |
with regard to plane specifically is going to be futile. 00:16:31.800 |
Yeah, it's going to be at best harm reduction because as long 00:16:35.100 |
as there's many ways to publish this information on Reddit, 00:16:39.200 |
So listen, I think this whole I think this whole policy on 00:16:43.900 |
I think the real issue the real underlying issue is that the 00:16:46.600 |
FAA is publishing these I count numbers thereby making every 00:16:51.700 |
I don't think there's I have a counter to it. 00:16:54.200 |
If I'm necessary, I have a counter to it actually sacks if 00:16:57.000 |
I may what we saw whether you agree with it or not with the 00:17:03.000 |
mass banning of certain individuals did actually silence 00:17:08.700 |
One of the reasons in fact, you'll I'm wanting to buy Twitter. 00:17:11.400 |
So if you look at certain individuals, whether it's Milo 00:17:14.000 |
Alex Jones Trump himself right on down the line when they 00:17:17.700 |
got banned across all systems, it was dramatic in terms of 00:17:23.200 |
So because of the size and scale of YouTube Facebook Twitter, 00:17:27.600 |
When they act in coordination, they can have a dramatic impact 00:17:32.400 |
not a perfect impact but a dramatic which is why we have 00:17:35.000 |
this issue of hey should 230 be rethought because when they 00:17:37.800 |
act en masse, it is extraordinary what they can do to an 00:17:51.200 |
And then we're moving on to what could be the greatest science 00:17:55.300 |
corner ever in the history of all in pod final word. 00:18:02.700 |
And my point is I don't care about any of this stuff. 00:18:05.400 |
Like I said, like this is my point is like the if you if you 00:18:10.000 |
take like an average person, okay, you know, we are let's 00:18:16.500 |
And you know, if you take out the time with our family, David 00:18:20.900 |
the family is people that are related to you. 00:18:22.900 |
We'll get to can we DM sacks those people one more time. 00:18:28.500 |
But if you take that out and you take out, you know, exercising 00:18:35.500 |
That's when you increase your heart rate and sweat sacks. 00:18:38.600 |
The point is that you have let's just call it 12 hours, you 00:18:42.100 |
know, a functional executive time that you can apply to a 00:18:44.700 |
problem and you can break that down into these blocks, right? 00:18:49.800 |
I would really love what is basically the smartest human and 00:18:53.900 |
the most productive human of our generation to be filling those 00:18:57.900 |
blocks with things that sort of like really transcend and 00:19:04.100 |
increasingly and I agree that freedom of speech is important 00:19:08.600 |
increasingly those buckets are being filled with things that 00:19:12.700 |
are very low level and hyper tactical and are distractions 00:19:21.100 |
And so I think that hopefully he gets all this shit under 00:19:29.200 |
I would like to see him get back to landing rockets on barges 00:19:41.500 |
But I just say one of the reasons why we don't care that 00:19:43.600 |
much about this issue is because I think something to understand 00:19:48.200 |
that's important is there are different kinds of speech and 00:19:50.100 |
different kinds of speeches are different levels protection. 00:19:52.300 |
The fact of the matter is like business advertising is not as 00:19:54.900 |
protected as political speech porn is not as protected as 00:19:57.500 |
political speech political speech speech criticizing the 00:20:00.700 |
people in power is the most protected category of speech 00:20:03.400 |
because the founders of the country understood that the 00:20:06.300 |
people in power will always try to insulate themselves from 00:20:09.700 |
accountability by limiting that kind of speech, but that is 00:20:12.700 |
precisely the kind of speech that the former rulers of Twitter 00:20:15.800 |
suppress the most and show the least sensitivity to so listen. 00:20:19.800 |
I mean is Ilan going to be the perfect content moderator? 00:20:22.200 |
No, I mean nobody is nobody nobody is but I do not believe 00:20:26.900 |
that puts him in the same category as you know, Vijaya Gaudi 00:20:31.200 |
or Yoel Roth who showed no sensitivity for political speech. 00:20:35.200 |
He has indicated a desire to restore freedom of speech and 00:20:38.000 |
I think they ultimately ended up in a good place. 00:20:43.100 |
I don't let's move on to Twitter handles best science corner 00:20:46.600 |
ever. So according to sources scientists work for the US 00:20:50.900 |
government have achieved a net energy gain in a fusion reaction. 00:20:57.300 |
We had ignition energy, which is very different from net energy 00:21:03.200 |
Please let the science nerd have you you have to you have to 00:21:06.000 |
say it correctly so that people understand what you're talking 00:21:08.400 |
about. Let me just make it even simpler than explain to us. 00:21:13.700 |
Friedberg and explain to us why this could potentially change 00:21:17.000 |
everything we did this on a on a show once before but I'll 00:21:20.700 |
kind of do a quick kind of summary again, basically, if you 00:21:25.300 |
take atomic nuclei, which are made of protons and neutrons, 00:21:28.200 |
and they repel one another right because protons are positively 00:21:32.700 |
charged. So they want to push apart from each other. 00:21:35.100 |
So with enough energy and enough density meaning that they're 00:21:40.100 |
They'll overcome their repulsion and jam into each other when 00:21:44.100 |
Some energy is released because the total mass of the fusion of 00:21:48.200 |
those those nuclei is actually less than those nuclei when 00:21:53.800 |
And so some energy is released and that energy drives a chain 00:21:59.500 |
And so fusion is this concept that is fundamental to physics 00:22:05.800 |
and fundamental to the energy driver of our universe. 00:22:08.100 |
So the star in our in our sky, the sun is driven by fusion and 00:22:15.000 |
only about 15% of the mass of the sun at the center is dense 00:22:20.900 |
So the big challenge with fusion is how do you get these these 00:22:25.000 |
atomic nuclei close enough together and moving fast enough 00:22:28.900 |
that they'll actually fuse and release energy and that's super 00:22:32.100 |
The reason it happens in the sun is because the sun has so much 00:22:35.400 |
mass that the gravity pulls all those particles together, they 00:22:39.600 |
get close enough, they get hot enough, they move fast enough, 00:22:41.800 |
and fusion happens, boom, all this energy comes out and every 00:22:48.600 |
But if we can do it, what happens when you fuse nuclei 00:22:51.700 |
together is you don't release any this isn't like a radioactive 00:22:56.600 |
You release energy that can be harnessed to drive our systems 00:23:03.400 |
So in the 1950s, you know, this was theorized, hey, we could 00:23:06.200 |
do fusion on Earth, we got to get a really, really dense plasma 00:23:09.500 |
meaning the atomic nuclei and the electrons have kind of gone 00:23:14.800 |
You got to get them to move super fast, like 10s of millions 00:23:17.600 |
of degrees Celsius, and you got to get them really close 00:23:22.100 |
So there's a couple of concepts to do this, one of which is 00:23:26.100 |
called inertial confinement, which is where you basically 00:23:29.200 |
create a little pellet of the material, you're going to try 00:23:33.700 |
and get to fuse, and you put a ton of energy on the outside, 00:23:39.300 |
And when you compress it really hard, really fast, and you 00:23:41.300 |
can get it to be done in a perfect sphere, and you can get 00:23:44.400 |
it to collapse on itself very quickly without, you know, kind 00:23:47.900 |
of shooting all over the place, enough of those particles 00:23:50.500 |
will come close enough together fast enough hot enough, and 00:23:53.900 |
Another way is through magnetic confinement, where you use 00:23:57.600 |
magnetic fields to create a really hot plasma, get it to spin 00:24:02.200 |
And then the magnet brings that super hot plasma closer and 00:24:05.500 |
closer and closer together until all those particles are moving 00:24:08.300 |
fast, and they're dense enough that they start to fuse. 00:24:10.800 |
So you know, one is called magnetic confinement, the other 00:24:14.700 |
And so what we saw happen this week is at the National Ignition 00:24:19.000 |
Facility, which is a facility that was built starting in 1997. 00:24:23.000 |
And they've spent about three and a half billion dollars to 00:24:25.500 |
date, they demonstrated a net energy output from the fusion 00:24:31.800 |
And what that means is they took a little pellet, and that 00:24:34.700 |
pellet was made up of deuterium and tritium, the atomic nuclei 00:24:39.000 |
that they use the particles that they use are deuterium, which 00:24:42.400 |
is a proton and a neutron stuck together, and then tritium, 00:24:45.400 |
which is a proton and two neutrons stuck together. 00:24:48.100 |
And the reason they use those two combinations is of all the 00:24:51.300 |
different ways you could fuse nuclei together. 00:24:55.000 |
This has the best energy output of any kind of reaction. 00:24:58.000 |
Freeberg, what actually happened this time that made this work 00:25:02.000 |
for what apparently only $3 billion. You said you didn't say 00:25:05.800 |
3 trillion, 3 billion, three, about a third of Yeah, about a 00:25:08.800 |
third of what Sam Bankman fraud stole. We have done something 00:25:12.500 |
here. Allegedly. So what actually happened that is so dramatic 00:25:17.300 |
that we have a press conference, everybody's losing their mind. 00:25:19.500 |
So yeah, I just want to highlight one more thing about 00:25:21.900 |
why this is so hard. You have to get such an incredible density, 00:25:26.300 |
you have to get an incredible energy. So high temperature and 00:25:29.600 |
high density, that confining those atoms and not letting them 00:25:34.100 |
escape and, you know, basically dissolve before they fuse is 00:25:39.100 |
super difficult. It requires so much energy in such a controlled 00:25:42.700 |
way in such a perfect and precise way that all of the digital 00:25:46.900 |
technology, the magnets, all the measurement systems, all the 00:25:50.700 |
software, it's taken us decades to get everything that allows us 00:25:54.500 |
to do this today. And now we're at the point that we may be able 00:25:58.100 |
to start to realize production scale, kind of versions of this. 00:26:02.300 |
So what they did is they had a small deuterium and tritium 00:26:05.700 |
pellet. And they shown 192 lasers onto this container that 00:26:09.800 |
held that little fuel. 192 lasers, the whole thing happened 00:26:14.300 |
in a billionth of a second, the lasers pulsed, boom, here's an 00:26:17.700 |
image of it. And as they did that, they, you know, basically 00:26:21.400 |
x rays kind of hit the sphere, this little BB, if you will, BB 00:26:26.400 |
kind of thing and compressed it and it compressed so quickly. 00:26:29.700 |
And with such heat, and it didn't dissipate because it was 00:26:32.100 |
done so precisely, all the lasers hit at the exact right 00:26:35.600 |
time, boom, this thing compressed, and then energy came 00:26:38.800 |
out. And the energy that came out that was measured was one 00:26:41.400 |
and a half times the energy that kind of went into that reaction. 00:26:44.500 |
And here's a chart that that I'll show you from the National 00:26:46.900 |
Ignition Facility, which shows just how inefficient the system 00:26:50.100 |
still is. And this isn't even speaking to to Chamath's point, 00:26:54.900 |
but basically these guys lose 90% of the energy that they put 00:27:00.700 |
into the center of the system, only 10% is actually used to 00:27:04.900 |
drive the compression, the rest of it is lost. And there's a 00:27:07.400 |
lot of ways to improve the efficiency of the system from 00:27:10.400 |
here. But basically, they put two megajoules in, they got 00:27:13.700 |
three megajoules out. And so it was the first proof point 00:27:18.000 |
production proof point in the 70 years that we've been 00:27:20.800 |
theorizing about nuclear fusion here on planet Earth, that this 00:27:24.200 |
is possible, and it's real. Now, this is these kind of inertial 00:27:27.700 |
confinement systems. There are 33 private technology companies 00:27:32.600 |
today that have raised about three to $4 billion so far this 00:27:37.100 |
year, to pursue several other technologies, besides what the 00:27:41.400 |
National Ignition Facility is showing to try and build 00:27:44.100 |
production ready versions of nuclear fusion. And so these 33 00:27:48.700 |
companies are using a bunch of different types of tools, one of 00:27:51.700 |
which is the tokamak. If you show the image, I'll show you 00:27:53.700 |
this one. There it is. Yeah, tokamak. This is what we talked 00:27:56.000 |
about. That is the magnetic spinning thing that looks like 00:27:59.540 |
Iron Man's arc reactor. Yeah, based it on Yeah, yeah, you 00:28:04.300 |
create a plasma, you basically speed up the hydrogen nuclei 00:28:07.580 |
super, super fast, these uterium and tritium nuclei, super, super 00:28:11.020 |
fast. And then you use magnets and the magnetic field has to 00:28:14.440 |
precisely squeeze the plasma, squeezing and squeezing and 00:28:18.160 |
squeezing it. And if it's slightly off in even the tiniest 00:28:21.120 |
way, think about a balloon, right? If you put a pinhole in a 00:28:24.200 |
balloon, everything escapes from the balloon. You got your 00:28:26.480 |
account. That's how technically hard this is. You're basically 00:28:30.040 |
trying to create a balloon with a magnetic field and you're 00:28:33.200 |
trying to keep the gas and you're trying to make it smaller 00:28:35.440 |
and smaller and smaller. And if any tiny hole emerges, the 00:28:41.580 |
Like when you're trying to hold in the Wagyu? Anyway, let me ask 00:28:47.700 |
this question then about the consistency of this and then 00:28:52.220 |
we'll go to you, Chamath. Can they do it consistently? Or do 00:28:56.120 |
you think this is like they got lucky once? Or are we going to 00:28:59.700 |
be sitting here a year from now and they're like we put in two 00:29:01.700 |
when we got out six. So we did it five times. 00:29:04.260 |
Yeah. So now we've proven that humans can do this. Okay. Which 00:29:08.220 |
is look, I mean, I want to give you guys some and I know kind of 00:29:11.820 |
some of Chamath concerns, which is how humans can recreate the 00:29:14.280 |
sun is what we this comes down to. No, no, no, I want to just 00:29:18.960 |
say one like, even close. I want to say one important thing from 00:29:23.840 |
a historical context. All breakthrough technology starts 00:29:28.640 |
out seeming impossibly large, impossibly expensive and 00:29:33.120 |
impossibly slow. The human genome project 20 years ago cost 00:29:37.140 |
$100 million to sequence the human genome. Today, we can do 00:29:40.600 |
it in a couple of minutes for $100. Okay, credible. The first 00:29:44.920 |
computer, the ENIAC computer had 500 flops of compute capacity, 00:29:49.320 |
it filled a room, it cost $8 million to build. 20 years 00:29:54.400 |
20 years later, all the wrong No, this is all this emotional 00:29:57.460 |
bullshit. You're using the wrong examples. Okay, let him finish. 00:30:00.560 |
Then you go to finish your sentence freeberg. And then we 00:30:02.640 |
go. And today we have an iPhone that can do 2 trillion flops of 00:30:07.040 |
compute in your pocket. I think that what we're seeing with 00:30:10.600 |
fusion today is similar to what we saw with the ENIAC computer 00:30:13.680 |
in the 1950s, which is the demonstration that compute is 00:30:16.800 |
possible. And now we're seeing a demonstration that fusion is 00:30:19.760 |
possible. And a lot of folks have anticipated this moment. 00:30:23.400 |
And they've invested ahead of this. Now, I don't know if any 00:30:26.000 |
of these companies that are currently kind of being built 00:30:28.080 |
are going to be production ready anytime soon. My estimate is 00:30:31.720 |
that we will see production demonstration. Okay, fusion, 00:30:35.200 |
here we go. 23 in the 2030s. So call it eight years from now 00:30:39.200 |
plus, and then you'll see grid scale scale up in the 2040s. So 00:30:43.400 |
this isn't something that's going to happen next year or two 00:30:45.400 |
years for already happening. What are you talking about? Okay, 00:30:48.120 |
now, Chamath your rebuttal. Oh my god, knowing you're a huge 00:30:51.360 |
fan of solar. This is the most navel gazing head up your ass 00:30:54.840 |
scientific bullshit. I've ever heard. Okay, couple points. 00:30:57.760 |
Let's start with the basics. The first is that there was no 00:31:01.880 |
previous technical. Why are you being like, why are you angry? 00:31:05.180 |
At me? I'm not angry. I find it so tiring hearing this. It's all 00:31:09.680 |
syrup. You're a little triggered. He's right. Yeah. 00:31:11.900 |
Why? Like, I don't get it. Because I don't find this 00:31:15.380 |
intellectually honest. Okay, I find it intellectually 00:31:19.160 |
on. Let me finish. Let me finish. Okay. When you talk 00:31:24.860 |
about sequencing the genome, there was no alternative. So 00:31:28.460 |
you're right. It was an enormous technical leap forward. When we 00:31:32.640 |
built a computer, there was no analog. It was an enormous 00:31:36.840 |
technical leap. And so you're right, we have a cost curve we 00:31:40.320 |
don't understand. And then we iterate as rapidly as possible. 00:31:44.200 |
And all these innovations where we built an entire 00:31:46.260 |
infrastructure to ride down the cost curve. The thing is, fusion 00:31:50.520 |
energy exists today. It's called the sun, we actually know how to 00:31:56.360 |
capture it at virtually no cost right now. So according to the 00:32:00.280 |
IAEA, today, you can capture grid level solar energy for 00:32:04.960 |
about 3 cents a kilowatt hour. That's as close to zero as we've 00:32:08.960 |
ever been. And over the next 10 years, their forecast is it's 00:32:12.400 |
going to get to one and a half cents. If you then want to store 00:32:15.880 |
it, and you layer in storage costs will be at a whopping 3 00:32:20.040 |
cents a kilowatt hour. That's where we are today. And so I 00:32:25.280 |
think that fusion does exist. I do think that this is an 00:32:29.320 |
incredible technical leap to replicate something that exists. 00:32:33.880 |
And I think that's where the intellectual dishonesty is, it 00:32:37.040 |
does exist. It has been captured, it can be harnessed. 00:32:41.400 |
And there is a positive energy equation, just in a different 00:32:45.040 |
modality that doesn't speak to these technically minded 00:32:47.840 |
individuals, a couple of other points about what I saw. I think 00:32:50.920 |
it's incredible what happened, okay. But just to make sure 00:32:53.800 |
we're clear, this is 192 lasers, the size of three football 00:33:00.120 |
fields, that consumed 322 megajoules of energy, which then 00:33:05.720 |
ultimately delivered two megajoules to a target, which 00:33:08.640 |
then released three. So this is why I'm saying we had positive 00:33:11.200 |
what's called ignition energy, we did not have positive 00:33:14.000 |
electrical energy captured. So yeah, could we figure this out? 00:33:17.000 |
Absolutely. Can we then shrink the three football fields down 00:33:21.320 |
to something that looks the size of a laptop? We possibly could, 00:33:25.400 |
will it take 20 or 30 years? Possibly. But in the meanwhile, 00:33:29.240 |
if the goal is unlimited, costless energy, you're on that 00:33:34.600 |
Yeah, but why can't it be both? So you said I was being 00:33:37.040 |
intellectually dishonest. What was I dishonest? Yeah, you're 00:33:41.120 |
comparing what you're saying is right. Yes, you can get you 00:33:44.080 |
would you seem to be an agreement? Like, yeah, I just 00:33:47.440 |
think that I think that you're trying to say that this is an 00:33:49.640 |
entirely new thing. No, it's a different approach to a thing 00:33:57.120 |
Let me bring what I would argue to my and I think this is 00:34:00.040 |
important be the net energy you can capture on, say, a football 00:34:04.000 |
field sized facility from solar is, you know, a tiny fraction of 00:34:08.440 |
the energy you could generate from a football field sized 00:34:11.320 |
fusion reactor. And that's why the argument would be like, 00:34:14.240 |
hey, you know, when we were developing computers, hey, we 00:34:16.920 |
have abacus is we shouldn't be developing computers. And I 00:34:19.360 |
think that's, that's the analogy I would use here. 00:34:21.200 |
This is why the cost per kilowatt hour is what the the 00:34:25.520 |
levelized cost of energy tries to do. It tries to, to normalize 00:34:29.880 |
that argument away, because everybody would say that, hey, 00:34:33.160 |
hold on a second, you're going to need plainfuls of this or 00:34:36.320 |
boatloads of that. And people said, No, what's the levelized 00:34:39.400 |
cost of energy? How what is the cost per kilowatt hour to 00:34:42.800 |
generate energy? And what I'm saying is, that is an absolute 00:34:46.960 |
scale, and free is zero, and we're at 1.5 cents. 00:34:50.880 |
Here's what I would say once check out one sec, the okay, 00:34:53.600 |
come off, the opportunity here is not necessarily about cost 00:34:57.400 |
reduction, it is about scalability. And if hydrogen is 00:35:01.000 |
abundant, which it is on this planet, it is nearly infinitely 00:35:04.280 |
abundant, we can take that hydrogen, and we can scale up 00:35:07.720 |
energy and electricity production in a way that is 00:35:10.120 |
unimaginable compared to solar. And I don't think that solar 00:35:13.440 |
should be excluded. Solar is key today and should be scaled 00:35:16.920 |
up. And I'm 100% agreement with you. But the scalability to go 00:35:19.960 |
100 x if we want to make 100 times more electricity, I think 00:35:25.960 |
So I think we have reached a good settlement here. Chamath 00:35:30.840 |
you're saying, Hey, listen, we're getting solar down so 00:35:33.520 |
cheap, we can solve this problem, all forms of energy. 00:35:35.920 |
Okay, great. We are solving that. So for our needs today, 00:35:39.000 |
and then what freeberg is saying, but what if you had 00:35:42.360 |
unlimited a thing that we can't even imagine? Beautiful. Now, 00:35:46.080 |
watch, as I get sacks involved in a science conversation. He 00:35:50.040 |
has zero interest in Mr. David sacks. If in fact, there was 100% 00:35:57.880 |
more free electricity available in this time frame, the next 00:36:00.960 |
100 to 200 x 100 x the available energy, in other words, supply 00:36:05.680 |
of energy just becomes flooded. And it's free, essentially, what 00:36:10.400 |
would be the geopolitical reaction on planet Earth in 00:36:14.560 |
terms of this incredible rivalry, rivalry we have with 00:36:18.840 |
China, and for humanity on a political basis? Such a good 00:36:23.360 |
question. J. Go ahead, sir. Good question. Here we go. Thank you. 00:36:26.400 |
World's Greatest moderator. Why don't we let freeberg answer? 00:36:28.840 |
No, no, you're the politics guy. Get in there. Take a second to 00:36:32.640 |
He has thought about this. Let's I want to hear his answer. I 00:36:36.280 |
actually want to hear your answer. Yes. You know, in a 00:36:38.520 |
world where you know, energy becomes more abundant, David 00:36:42.040 |
hasn't been paying attention, guys. This is what he's trying 00:36:47.160 |
Jake, I'll call your intellectual dishonesty and 00:36:55.080 |
I love you. I love you. You know, I do. I like to have a 00:37:01.120 |
great night on Monday night, a Sunday night. Great. Let's talk 00:37:04.200 |
later. I just want to be together last week. You call me 00:37:08.160 |
petty to I think we're well, here's what I want to be clear. 00:37:10.520 |
I think that I'm just glad you guys are fighting not mean sex. 00:37:14.080 |
Guys, please let me finish. Okay, I think that this 00:37:16.280 |
breakthrough is really valuable. I think it's interesting to see 00:37:19.640 |
that these kinds of scientific breakthroughs continue to happen 00:37:22.800 |
in government sponsored facilities and not private 00:37:25.320 |
companies. And I think that that's probably where a lot of 00:37:28.520 |
these innovations will continue to come from because look at the 00:37:31.160 |
scale of what had to be built. Three football fields and 322 00:37:35.160 |
megajoules of energy and 192 lasers. This is really 00:37:38.280 |
complicated, expensive stuff. I'm an enormous fan of these 00:37:41.320 |
kinds of scientific breakthroughs. I want to be 00:37:42.760 |
clear. I think that where I struggle is translating this 00:37:48.240 |
into actually an investable area. And I worry that this is 00:37:52.920 |
going to consume lots of money by folks that could otherwise 00:37:57.480 |
put money to work in things that will actually pull forward our 00:38:01.280 |
energy independence and energy abundance sooner and faster. So 00:38:04.360 |
for example, you know, there are all kinds of things that we 00:38:07.640 |
could do to secondary, ternary, third, fourth and fifth 00:38:11.800 |
generation batteries that aren't happening today. There are a 00:38:14.600 |
bunch of things that we could do to actually create an 00:38:16.760 |
infrastructure of green hydrogen. And the the 00:38:19.480 |
simplistic answer is we could do it all. But the reality is 00:38:22.680 |
money is finite, and we can't. And all I'm observing is I do 00:38:27.400 |
think that more practical things that do have geopolitical 00:38:30.120 |
ramifications sooner are not going to get funded because 00:38:34.280 |
people do get enraptured by this. And my skepticism is that 00:38:38.680 |
this is still in the realm of government sponsored research 00:38:42.600 |
and is not really an area that for profit private companies 00:38:47.080 |
can tackle. And so I would rather those for profit 00:38:49.320 |
companies, for example, why combinator just today, put out 00:38:52.800 |
something where they were, you know, call to action a request 00:38:55.240 |
for startups in climate. And when you look at that list, 00:38:58.280 |
those are really practical, investable areas. And I just 00:39:01.160 |
want to make sure that the capital allocators that listen 00:39:03.520 |
to this, weigh those equally. I am glad. I'm glad that the US 00:39:08.480 |
government did this. I hope they do more of this. But if you're 00:39:11.800 |
asking me quite honestly, I would rather the next $10 00:39:14.200 |
billion go into energy efficiency HVAC, then fusion 00:39:18.880 |
because a fusion exists and B, I think it's going to happen at an 00:39:23.080 |
innovative bench scale level by the government and not by a 00:39:28.440 |
Let me let me just respond to that real quick. Tomorrow. I 00:39:31.840 |
think that the idea of allocating our resources as a 00:39:35.200 |
society should be done on kind of, you know, on a portfolio 00:39:39.120 |
basis, 80% on the pragmatic near term, 15% on the kind of next 00:39:44.320 |
gen and 5% on the moonshot. And this maybe starts to shift from 00:39:48.680 |
the 5% to the 15%. Maybe it's still in the 5%. But I don't see 00:39:52.840 |
kind of overfunding happening. So I'll tell you guys there was 00:39:55.400 |
a survey done. There's 33 private companies in fusion that 00:39:58.680 |
are kind of fusion companies today. VC back eight new this 00:40:02.720 |
year. So the number is kind of increased by 33% this year. And 00:40:07.120 |
so far this year, those companies have raised around 00:40:11.640 |
is a fraction of what was done by 15 minute delivery companies 00:40:17.240 |
Exactly my point. And and by the way, the biggest funding is 00:40:20.000 |
happening in iter, which is the largest construction project in 00:40:23.320 |
Europe. And this is a $30 billion production scale fusion 00:40:27.640 |
demonstration system that should be online by the end of the 00:40:29.800 |
2020 government sponsored government sponsored. Yeah. And 00:40:32.480 |
so to my point, I'm a huge fan of government sponsored 00:40:35.320 |
research. We get finally, the first science corner, everybody 00:40:39.720 |
brace themselves. It's the first science corner, where David Sacks 00:40:43.160 |
has the cherry sacks. Here we go. Come on, you can do it. You 00:40:47.520 |
What was your question? Your question was a little bit was 00:40:52.280 |
let me reframe. Okay, I'm coming in. What's the geopolitical 00:40:55.360 |
impact if this does happen 100x energy? It's fantastic for the 00:40:58.720 |
United States if it actually happens. And the reason is, if 00:41:01.440 |
you look across the world, there's this thing in politics, 00:41:05.120 |
known as resource curse, where the worst governments, the most 00:41:09.240 |
despotic governments tend to be in the countries that have the 00:41:13.560 |
biggest natural resources, ironically, so the countries 00:41:17.680 |
that have huge amounts of petroleum or other kinds of 00:41:20.600 |
minerals, they've tend to have pretty corrupt governments. And 00:41:25.480 |
the reason for that is that if you're sitting on a giant oil 00:41:29.480 |
reserve, you don't need to make anything else work, you just 00:41:31.960 |
fight over who gets to control that oil reserve. And that's 00:41:34.760 |
what politics ends up being, you don't need to create policies 00:41:38.480 |
that foster innovation, or attract knowledge workers, 00:41:42.120 |
right, you just basically mine that oil. So if all of a sudden, 00:41:46.200 |
you're talking about turning energy into a software problem, 00:41:50.560 |
or an innovation problem, that looks a lot more like the 00:41:53.480 |
software industry. That's an area where the United States has 00:41:56.400 |
a huge advantage. And yeah, I think it would pull the rug out 00:41:59.480 |
from under many countries all over the world in favor of the 00:42:03.400 |
United States. I mean, it's a big if because where I agree 00:42:06.840 |
with Chamath is this stuff still seems pretty far off. And it's 00:42:10.400 |
still pretty unproven from a commercial standpoint. But I 00:42:14.080 |
agree with freeberg, why not try investing in and cultivating it 00:42:18.620 |
Okay, fantastic. This was a fantastic science corner, where 00:42:22.500 |
we actually engage David Sachs, which country 00:42:28.740 |
All right, take it easy. Yes, I just want to let you know, I 00:42:32.140 |
think you like that answer, right? Well, I love all of your 00:42:36.020 |
intellectual enough for you. Or was I love your intellectually? 00:42:40.700 |
knows I was a steel person. felt silver silver silver manning. 00:42:46.300 |
But I love what your intellectually silver person. 00:42:54.500 |
I think that's your you're in platinum with some diamond dust. 00:42:57.380 |
I just can I just say spent a couple nights together this past 00:43:01.180 |
week has really really done the mood of the show. I mean, when 00:43:03.580 |
you guys go out and drink together and you guys are 00:43:06.820 |
left our asses off Sunday night. Can I just say, we Saxon I had 00:43:11.260 |
the best 48 hours together in a decade. This is Sunday night. 00:43:15.460 |
It turns out Chris rock and Chappelle are playing at the 00:43:17.900 |
Chase Center, where Chamath used to own a piece of the Warriors, 00:43:21.540 |
right. And this incredible arena has this incredible show. And you 00:43:27.180 |
know, me and Saxon, some friends will leave it at that go to the 00:43:29.820 |
show. Our bestie dream on is at the show. So I text Dre and I'm 00:43:34.340 |
like, Hey, you're going to see Chris rock by chance tonight 00:43:36.620 |
with Chappelle? Say yes, sir. I said, Hey, we're gonna go with a 00:43:40.180 |
couple of friends. Maybe we roll together hang out after the 00:43:42.620 |
show. We go. And after the game after the show, which was 00:43:48.340 |
incredible. We go backstage and I'm sorry to the to the practice 00:43:53.340 |
court. And we're hanging out with dream on in the practice 00:43:56.300 |
court with Dave Chappelle, Dave Chappelle, and I start shooting 00:44:00.460 |
hoops. David Sacks is talking to Chris rock about free speech. 00:44:04.260 |
Steph Curry comes out and starts giving Dave Chappelle and Jay cow 00:44:10.020 |
a shooting lessons. Where Chappelle and I are breaking 00:44:14.580 |
like old men, you know, you know, on a concrete court, all 00:44:19.140 |
of a sudden, Steph says, Hey, Jay, how you got it? And by the 00:44:22.300 |
way, he's fan of the show. He says, you're short every time. 00:44:25.420 |
And just, you know, hit the backboard, you got to go long, 00:44:28.260 |
then he tells Chappelle, you got to change this. All of a sudden, 00:44:31.100 |
we start hitting shots like, you know, we're on the rain, 00:44:33.460 |
injuries, you're raining, freeze, rain. It was literally 00:44:36.260 |
like cut into here were these mid range jumpers or threes. I 00:44:39.220 |
was a free throw line extended free throw line extended cut 00:44:42.900 |
into here rain man and rain dance from a long came Polly 00:44:57.780 |
I was hitting brick after brick, rest in peace, Philip Seymour 00:45:02.020 |
Hoffman. So then we're chilling. And Saxon are talking to Dave 00:45:08.940 |
Joe Lakeham owner of the Warriors are there the majority 00:45:11.740 |
owner, you know, as opposed to you being a minority owner. 00:45:17.900 |
The only person who drops more names is Phil Hellmuth. I mean, 00:45:28.020 |
Saxon, am I leaving? I want to say it's so bad, but I'm not 00:45:31.540 |
gonna say not dancing. So brutal. There'll be zero docks. 00:45:36.540 |
So we I kid you not. Chappelle comes over and says, Jacob, 00:45:41.660 |
fact, you guys want to go to after to do it. Go go see me do 00:45:47.100 |
a show at like 1am at this like local comedy club with 70 seats. 00:45:50.140 |
I said word. Yes, we go at 1am. Chappelle sits on stage, smoking 00:46:01.340 |
doing 90 second pauses, and then having a beer and 00:46:06.420 |
interacting with the audience. And does a two hour set after 00:46:09.820 |
doing this set with Chris rock at the Chase Center. Me and Saxon 00:46:13.460 |
Draymond hilariously laughing. The stuff Chappelle is a genius. 00:46:19.980 |
And when you see his show and Chris rock, by the way, he puts 00:46:23.660 |
a tight set together. I mean, Chappelle's got this 00:46:26.380 |
storytelling thing where he kind of meanders a little bit and 00:46:29.100 |
then he hits you with it. But Chris rock is just bang, bang, 00:46:31.980 |
bang, bang, bang, bang, bang. Extraordinary, just two 00:46:35.260 |
incredible minds at the top of their game artists, artists at 00:46:39.220 |
the top of their game, doing what society needs. But more 00:46:43.700 |
importantly, doing what David Saxon I needed, which was to 00:46:47.420 |
laugh our asses off together and remember our friendship. So it 00:46:51.140 |
was a great night out. I want to say to bestie Draymond Dave 00:46:54.060 |
Chappelle and Chris rock. Thank you. The David Sacks J. Cal 00:46:58.220 |
bestie friendship has never been stronger. I don't know about 00:47:01.220 |
freeberg and Chamath that seems to be on the rocks. 00:47:03.420 |
Yeah, it's weird. We'll be in. We'll be we'll be vacationing 00:47:07.420 |
together next week. So I love I love freeberg. Yeah. Well, well, 00:47:11.740 |
we are going to be whatever, whatever's going on. We'll take 00:47:15.180 |
Well, I just want to say the alliances amongst the besties 00:47:17.900 |
people. I'll be honest, I got mad at freeberg when he edited 00:47:21.740 |
that Google bit to say the exact opposite of what he actually 00:47:24.460 |
said that we don't bring that up. Yeah, just beep that. But 00:47:27.980 |
yes, that is the thing that bothered me. I have to be 00:47:30.500 |
honest. That's you playing to the crowd versus you being 00:47:33.340 |
honest and telling what you think freeberg that freeberg 00:47:36.260 |
can let me put that in the form of a question freeberg has your 00:47:39.260 |
fame as the sultan of science. Because listen, you nobody knew 00:47:43.260 |
who you were outside of Silicon Valley before this. Has it 00:47:47.460 |
No, no, no, not my family. Look, I'll be I'll be honest. And I'll 00:47:52.740 |
speak openly about this. I had said that there could probably 00:47:55.140 |
be a significant headcount reduction of like 75% at Google 00:47:59.740 |
and the business could keep operating. And I took it out. 00:48:02.940 |
And I took it out because I have a lot of friends that work at 00:48:06.100 |
Google. Google is a close partner of mine. They're an 00:48:08.980 |
investor of mine. And frankly, I just want to be careful about 00:48:12.860 |
that. It's not something I commonly do. You know, as you 00:48:15.260 |
guys know, I usually speak my mind pretty clearly, but I was 00:48:17.140 |
just trying to be respectful. And that's the reason I did it. 00:48:19.340 |
You know, so that was fine. What I'm saying is not that it's 00:48:23.260 |
just that the part that you edited in actually made it seem 00:48:27.820 |
like you were not saying that at all. But the opposite, I think 00:48:30.300 |
if you had cut the whole thing, it would have been more honest. 00:48:33.140 |
So to keep that other thing in actually led the perception of 00:48:36.420 |
the opposite. So I think that I think no, no, I don't want to 00:48:40.500 |
trigger it. I'm saying, but I think you should have a 00:48:41.900 |
principle. I don't hold on very often tomorrow. Like, yeah, I 00:48:44.980 |
know. I think that we should just have a principle to not 00:48:48.660 |
play to what the perception of what we say should be, 00:48:51.180 |
especially if it means we could be saying the opposite of what 00:48:55.460 |
intellectual honesty is a best tenant. It's a best tenant. I 00:48:59.780 |
think so. Absolutely. Bestie 10. Always and the other bestie 00:49:02.540 |
tenant besties always come back together. If we have a fight, we 00:49:06.580 |
Can I'm still mad at you? Just kidding. Okay, but just talking 00:49:12.340 |
about hypocrisy. I mean, so how great was Sunday night? How 00:49:15.660 |
great was that gets up there and he gets like right out of the 00:49:17.980 |
gate. He's attacking woke right out of the gate. 00:49:26.700 |
shooksmith. He took down Will Smith. I mean, the Will Smith 00:49:31.460 |
takedown, which you will see in this special is so complete. It 00:49:36.660 |
is just chef's kiss. But how great was his set? Let's say 00:49:46.500 |
He, he did. But I thought the more important part of the set 00:49:49.420 |
was the he came right out like calling out all this, you know, 00:49:53.020 |
you know, holier than thou woke stuff. Yeah. And there was that 00:50:00.140 |
And also he said, Listen, words can't hurt you. Unless you write 00:50:04.740 |
them on a piece of paper and time to a brick. We these a 00:50:09.660 |
bigger point right now and this does tie into our first story is 00:50:12.740 |
I think comedians look at Twitter as a place to get 00:50:15.940 |
canceled, not a place to be part of the discourse. And that's a 00:50:18.940 |
huge loss that's indicative of our society being broken. And 00:50:23.300 |
it's incredibly important that these comedians be allowed to 00:50:28.860 |
mock and to speak and to step over the line and challenge us 00:50:34.860 |
as citizens in a free society. And we should cherish them. And 00:50:39.220 |
we should not even try to cancel them, let them cross the line, 00:50:42.740 |
let them say things that make us uncomfortable, so that we can 00:50:45.780 |
understand ourselves and our society better. And I just want 00:50:49.020 |
to say, Why can't you include lives of tick tock in that? Oh, 00:50:52.380 |
do we want to have a discussion about it? I can't go. I'm okay 00:50:57.500 |
with mocking. I just don't start you guys. You did so well. Come 00:51:00.500 |
on. Here we go. No, J. Cal cut it out. All right. Well, we have 00:51:04.500 |
two more science corners to get. All right. You know what I mean? 00:51:09.780 |
I just know what's wrong about the Coupa deal. And sex. This is 00:51:14.940 |
touched into it? To be honest? Oh, really? The thing you guys 00:51:18.060 |
asked is like, you know, what signal will Elon's moves at 00:51:22.940 |
Twitter be for the rest of the tech industry? I think the 00:51:27.820 |
biggest wake up call is to actually PE companies. So if you 00:51:32.420 |
play this out, and you think that Coupa is, you know, 00:51:36.620 |
explain what is, please. Coupa is a is a software as a service 00:51:41.140 |
company that does revenue management, I guess, or expense 00:51:45.660 |
forecasting or some something in the financial realm. I don't 00:51:49.260 |
particularly know to be honest. But anyways, this is a company 00:51:52.420 |
that you know, was off 70 or 80% from the high like a lot of SAS 00:51:56.340 |
companies were when rates started to go up. And they got 00:52:00.420 |
this offer from from Toma Bravo. But here's what's so interesting 00:52:03.700 |
about this deal. If you think that, you know, these guys 00:52:07.140 |
bought a company, I'm just going to make up a number at 20 times 00:52:10.980 |
EBITDA. Right. And, and you see Elon at Twitter, and you think, 00:52:16.980 |
well, wait, maybe we can't cut 75%. But maybe we can cut 50% 00:52:20.420 |
headcount and the company can still do well. And, you know, 00:52:23.940 |
you take half of the expenses out of the business. All of a 00:52:27.620 |
sudden, you know, if you're even two doubles, you're actually 00:52:29.580 |
buying it at 10 times. So I think the thing that is the that 00:52:34.820 |
is the real insight here is twofold private equity can still 00:52:40.340 |
put out a lot of private credit to fund these deals. And SAS 00:52:43.420 |
companies are perfect because they have huge free cash flow, 00:52:45.580 |
right. So instead of funding it based on earnings, they can fund 00:52:48.460 |
it based on ACV and ARR. So private equity will be super 00:52:52.100 |
active. And to all these rifts basically show what the 00:52:56.820 |
efficient frontier is for the number of employees you need to 00:52:59.020 |
run a company. And if you can cut 50% of the headcount, 00:53:02.340 |
private equity folks will do that. And so I think Cooper is 00:53:06.620 |
like the canary in the coal mine, it is the beginning of 00:53:10.380 |
what I suspect is a tidal wave of PE sponsored deals in tech 00:53:16.820 |
companies, largely SAS, but may go into other realms, recurring 00:53:21.300 |
revenue, that can go to these two things, tap the private 00:53:24.660 |
credit markets and finance it based on ARR, and then fire 50% 00:53:31.620 |
So on on Cooper, I thought the most interesting thing was just 00:53:35.500 |
the we got a public comp, well, we got a comp on what private 00:53:39.620 |
equity is paying for public companies right now. So the deal 00:53:43.620 |
happened at an $8 billion valuation, that was a 31% 00:53:47.620 |
premium to the public price. It was 8.4 times next 12 months 00:53:52.260 |
revenue. And on a trailing basis, it was about 10.4 times 00:53:58.140 |
the last 12 months revenue. And by the way, all the comments 00:54:01.220 |
were around how what a rich price Toma Bravo was paying 00:54:05.060 |
people generally thought they're paying a premium to the 00:54:07.580 |
valuation. So it was 77% premium before the rumors came out that 00:54:13.740 |
this was happening. So the premium, yeah, and there was and 00:54:16.900 |
there was a there was a bidding war with Vista. And so it was a 00:54:21.300 |
it was a really rich kind of deal that got done here. 00:54:24.020 |
Right. So my point is that people thought this was a really 00:54:27.700 |
rich deal. And yet the valuation multiples are so much lower than 00:54:31.820 |
what private company founders expect. So remember, last year, 00:54:36.020 |
at, you know, the peak, founders were thinking 100 times AR was 00:54:40.820 |
normal 100 times, and, you know, you could roughly say, you know, 00:54:44.580 |
AR is roughly equivalent to next 12 months revenue. It's not 00:54:47.820 |
perfect, but it's roughly the case. So these founders were 00:54:51.420 |
expecting evaluation, multiple 10 times what the public markets 00:54:56.980 |
are paying. And the public more and actually, the public markets 00:55:00.300 |
are half of where Toma Bravo was in this particular deal. So the 00:55:04.740 |
public markets right now are valuing the median SaaS company 00:55:09.780 |
at about five and a half times, and a high growth with that be 00:55:12.780 |
for like a 20% year over year growing company. And they're 00:55:15.460 |
valuing the high growth companies that may be eight 00:55:18.060 |
times, you know, and Toma Bravo did this at 10 times. So that 00:55:22.020 |
gives you a sense of what the ballpark is. And these are 00:55:24.300 |
companies that are already public, they're at scale, 00:55:26.980 |
they're doing roughly a billion dollars of AR, they have already 00:55:30.940 |
kind of won their category, to some degree, whereas private 00:55:34.820 |
companies are subscale. They're, you know, typically, you're 00:55:37.900 |
talking about companies with one, five, 10, usually under 00:55:41.180 |
$20 million of AR, they are, they're not de-risk, there's 00:55:44.700 |
still a ton of risk. We've seen many, many SaaS companies 00:55:48.060 |
fizzle out and plateau at 20 million of AR, never get to 100 00:55:52.340 |
million, nevermind a billion. And yet these founders think 00:55:55.300 |
that they're entitled to, you know, even in this market, 30 00:55:59.100 |
to 40 times AR, no way. I mean, like, it's getting to the point 00:56:02.700 |
now where, you know, maybe it should be 10 times 20 times like 00:56:07.620 |
max, for that to be for a company that's growing two and a 00:56:11.140 |
half, three x year over year. So I still think that like, so I 00:56:14.220 |
think basically, what we're seeing here is even a good 00:56:16.740 |
scenario, like a coupe acquisition that was done at a 00:56:19.900 |
premium, like it's still a wake up call to the private markets 00:56:23.580 |
that the valuations are still completely and utterly out of 00:56:26.380 |
Yeah, let me ask you a question, Zack. So this company was 00:56:29.220 |
growing 45% last year, they're growing 35% this year. And they 00:56:33.580 |
got this multiple. Why is it not worth a significantly higher 00:56:38.180 |
multiple if a company is growing two and a half to three x, which 00:56:41.020 |
is 250% 300%. And these guys are only growing 35%. 00:56:45.340 |
Sure. I mean, it is. And that's what you're paying a premium 00:56:48.940 |
for. But so the so the here's the theory of it is that if you 00:56:53.300 |
can invest in a private company that say tripling year over year, 00:56:55.860 |
and they can do that for another five years or whatever, then 00:56:59.100 |
you're paying for that you're paying for that outcome in a 00:57:01.060 |
couple years. Basically, well, think about it discount to the 00:57:04.220 |
outcome in a couple years. Well, if you're paying 30 times today, 00:57:07.060 |
and it triples next year, you're only paying 10 times next year. 00:57:09.580 |
And if you're only paying three times, so if that keeps going, 00:57:12.940 |
that's where your arbitrage is. But here's the thing you have to 00:57:15.980 |
weigh against that is that these early stage private companies, 00:57:19.460 |
many things go wrong. And they hit a plateau, they fizzle out 00:57:23.180 |
or their growth rate starts to the bigger they get, the harder 00:57:27.300 |
they should be priced at a discount, not a premium, because 00:57:30.580 |
there's risk, there's more risk, they're growing faster, but 00:57:33.740 |
there's more risk. But also, it's very hard once you get to a 00:57:37.060 |
bigger number of ARR 50 100 million of ARR, it's extremely 00:57:41.420 |
difficult to be doubling or tripling year over year. Let me 00:57:43.580 |
just point one thing out. So I looked at the numbers on Coupa. 00:57:46.540 |
I think they had about 170 million of stock based comp 00:57:49.460 |
expense in the last nine months. So those are employees that are 00:57:54.900 |
getting $170 million in compensation in the form of 00:57:59.340 |
shares. So they get those shares, they can then sell those 00:58:01.820 |
shares and get cash for them on the market, and then on the 00:58:05.460 |
public markets and pay their bills. So when a company like 00:58:08.420 |
this goes private, for those employees to just remain at 00:58:12.500 |
their baseline comp, that stock based comp needs to be replaced 00:58:16.860 |
with something else, or else they're seeing their salaries 00:58:19.500 |
reduced. So, you know, there's this balancing game when these 00:58:23.820 |
companies go private, in terms of how do you give them the comp 00:58:28.020 |
that they're earning to keep them engaged in the business, 00:58:30.700 |
the answer is don't versus no, but you let them you let them 00:58:33.900 |
quit because you want to do a riff anyway. Right. So I mean, 00:58:36.940 |
do you but for the people that stay, right, so there's a 00:58:40.260 |
balance because it's not just hey, cut the opex, you have to 00:58:43.460 |
cut the opex, including stock based comp. And this company 00:58:46.220 |
generated about $100 million, sorry, $210 million of free 00:58:52.660 |
cash flow or operating cash flow in the last 12 months. So if you 00:58:56.660 |
if you take out the stock based comp, these guys are actually 00:59:00.020 |
break even or losing money roughly. Yeah, so yeah, and 00:59:03.780 |
break even roughly. So there's a real question mark on this 00:59:06.980 |
business and businesses like this that go private, where if 00:59:09.700 |
you actually cut the opex and you cut the salaries and you cut 00:59:12.180 |
the headcount, but you have to find new ways to pay people 00:59:15.860 |
because you've been paying them with stock in the past, how do 00:59:18.100 |
you kind of bridge that gap? And I think that's probably a little 00:59:19.980 |
bit of the balance and the art of what these guys do well, 00:59:22.460 |
Jamal, if I may, can you explain to the audience what a private 00:59:26.660 |
equity firms expectation is in terms of return when they buy a 00:59:30.620 |
company like this, and then sacks, I saw your tweet that you 00:59:34.260 |
Well, I think it's changed over time. And this is what's so 00:59:38.540 |
powerful about the private equity industry. Look, you have 00:59:42.340 |
to think about what their incentive is, because it kind of 00:59:45.420 |
guides the yes. Early on, they were very much like venture 00:59:49.700 |
capitalists, they were out in the, you know, edges of risk 00:59:53.580 |
taking, doing all kinds of very difficult, gnarly deals. So if 00:59:58.660 |
you look back in the history of private equity, you know, these 01:00:01.300 |
huge, crazy deals like RGR Nabisco, or TWA Airlines were 01:00:05.220 |
the first of the industry, and they weeped enormous returns, 01:00:10.780 |
but there was a lot of risk, and it required very heavy handed 01:00:14.060 |
management. Oftentimes, what that meant was firing a lot of 01:00:20.820 |
institutionalized, and they don't generally feature 01:00:27.140 |
themselves as a place to get the best necessarily returns, but 01:00:30.540 |
they are places where you can put enormous amounts of money, 01:00:34.220 |
where the likelihood of loss is extremely zero, and you generate 01:00:40.260 |
very good rates of return. Now, again, this depends on whether 01:00:44.820 |
you want to look at IRR, or DPI, right? So a lot of people will 01:00:49.460 |
market IRR, which, you know, I think is kind of like a 01:00:53.860 |
gameable metric. But you know, those IRRs can be 20-25%. If you 01:00:58.980 |
look at DPI, which is really how much cash you get back, you 01:01:02.220 |
know, private equity firms can generate one and a half to 2x of 01:01:05.460 |
the money you give them. But they do it consistently, and 01:01:08.860 |
they very rarely lose money. So all of that is important into 01:01:13.220 |
understanding what's going to happen in this cycle. These 01:01:16.260 |
folks are going to buy a ton of these private software 01:01:19.900 |
companies. I think that they are going to fire lots of people. I 01:01:25.180 |
think they are going to make these companies run hyper 01:01:27.620 |
efficiently, and they will make sure that they generate that 1.2 01:01:32.900 |
to 1.7x that has been historical. Very rarely will 01:01:36.780 |
they lose money in these things. By the way, that's going to mean 01:01:40.580 |
that a lot of these other companies will have to reset 01:01:42.580 |
valuation. So you saw yesterday, checkout.com went from a $40 01:01:46.620 |
billion valuation down to 11. You're seeing some companies 01:01:51.140 |
only go down 10 or 15%. But it's a process, isn't it, Shamath? 01:01:55.620 |
Isn't this just like what happens in real estate where 01:01:57.980 |
beginning of this process? Yes, because in real estate, my 01:02:00.780 |
understanding having lived through these boomer cycles is 01:02:03.060 |
the person living in the home still believes their home is 01:02:06.100 |
worth, you know, this incredibly valuation, and then the people 01:02:08.860 |
who want to buy it are like, that doesn't match reality. And 01:02:12.060 |
then the real estate brokers go back and forth trying to get 01:02:14.820 |
people to, you know, go through this messy middle and come to 01:02:18.100 |
true price discovery a private company, it's hard to get true 01:02:20.860 |
price discovery until they're on the brink of insolvency. We 01:02:24.420 |
We just got some data on that actually, can we bring this 01:02:27.340 |
Cooley data? Yeah, so Cooley looked at a law firm in Silicon 01:02:33.020 |
Valley. Yeah, they're a prominent Silicon Valley law 01:02:35.260 |
firm, they looked at 1000 deals over the last three quarters of 01:02:37.900 |
this year. And what they saw is that you're the later the stage, 01:02:42.660 |
the bigger the valuation correction. So series D rounds 01:02:45.220 |
went from three and a half billion to 527 million. That's 01:02:48.380 |
an one 7% Oh, yeah, that's an 85% drop. Series C went from 01:02:53.260 |
502 million 230 million to 74% drop. Series B went from 164 to 01:02:59.500 |
90. That's a 45% drop. And then series A went from 58 to 45. 01:03:04.220 |
That's only a 22% drop, there's just less room to compress 01:03:07.980 |
there. But the point is that series B roughly a 50% drop 01:03:12.420 |
series C roughly a three quarters drop, and series D, 01:03:17.180 |
roughly a 85%. Yeah, one seventh drop. So I think founders right 01:03:23.580 |
now are they're just like a little bit delusional about 01:03:26.540 |
this money they raised last year, they're still way too 01:03:29.620 |
anchored on last year's valuation. And if only they 01:03:32.780 |
would think in terms of this capital they raised last year, 01:03:36.060 |
in terms of, of its real dilution in terms of what the 01:03:40.860 |
company is worth now, I think they'd be treating it more more 01:03:43.780 |
precious. So for example, sex, for example, hold on, if you 01:03:47.420 |
like, they won the lottery, and they don't want to they don't 01:03:49.900 |
realize they won the lottery. I had this conversation with the 01:03:51.860 |
founder, this is the only money they're ever going to see is the 01:03:54.700 |
bottom line. And they're spending like they're going to 01:03:56.460 |
win the lottery every year. So for example, let's say you take 01:03:58.980 |
a company, yeah, let's say you take a company that raised 200 01:04:02.900 |
million last year at 2 billion. So it was 10% dilution. So in 01:04:06.300 |
their heads, they're thinking, Oh, well, this isn't that 01:04:08.140 |
expensive, like 10% dilutions around the era, but really, 01:04:11.900 |
probably the company is worth maybe 400 million now, right? 01:04:15.740 |
Because it's gone down 80%. Yes, 200 million of your 400 million 01:04:19.860 |
is half the value of the company. Yes. And you're 01:04:22.140 |
squandering it, you're squandering it at a rate of 100 01:04:24.660 |
million a year. So you're basically burning up 25% of the 01:04:28.060 |
value of your company this year, and the next year. And then by 01:04:30.620 |
the way, you're going to be in crisis after that, because 01:04:33.140 |
you're probably like a lottery winner buying like a giant 01:04:36.220 |
I had an observation that a lot of the investors that sit on the 01:04:39.380 |
boards of these companies. They have an incentive to not see 01:04:45.060 |
those valuations come down too quickly, do they not. And so 01:04:48.900 |
there is this sort of like, interest in, hey, I don't want 01:04:52.820 |
you to have to go reprice the company or do a down round 01:04:55.180 |
because then my portfolio gets written down. And then I'm in 01:04:57.900 |
the middle, everyone's always in the middle of a fundraising 01:04:59.780 |
cycle with LPS. And then I'm going to have a tough 01:05:02.060 |
conversation with my LP is about my, my value. So do you not see 01:05:05.860 |
VCs and investors playing an active role in trying to keep 01:05:10.620 |
the valuations propped up to some extent, particularly where 01:05:13.340 |
they have big markups 100% by extending bridge rounds or doing 01:05:19.300 |
look, nobody, nobody likes to go through a down round. And that 01:05:23.460 |
includes founders and existing investors in the company. That 01:05:26.340 |
being said, we're not talking here about new financing 01:05:29.420 |
conversations we're talking about is advice that is 01:05:33.420 |
happening in board meetings. And, you know, maybe other VCs 01:05:36.940 |
aren't pushing as hard as we are. But you the advice I'm 01:05:39.500 |
giving in board meetings is what I'm telling you publicly today, 01:05:42.740 |
which is, this is the last money you may be able to raise on 01:05:45.780 |
attractive terms, if at all, you need to treat it much more 01:05:48.900 |
preciously, the world has fundamentally changed. And by 01:05:51.780 |
the way, we haven't even gotten into what's coming the demand 01:05:57.700 |
Explain what demand contract contraction is for the audience, 01:06:00.620 |
Okay, look, there's going to be three major sources of slowdown 01:06:04.260 |
for software companies next year. Number one, new business 01:06:08.020 |
is going to dry up companies are just going to be spending a lot 01:06:11.420 |
less money next year, because they're all cutting costs. So 01:06:14.500 |
you should expect your new business to be roughly 50% of 01:06:18.060 |
what it was. Next year, it'll be 50% of what it was last year. 01:06:22.060 |
That's my rule of thumb. For most companies, new business 01:06:24.900 |
down 50%. Number two churn is going to be higher. We haven't 01:06:28.260 |
seen that much logo churn yet. But next year, a lot of 01:06:31.620 |
companies are going to start going out of business. And it's 01:06:34.260 |
going to happen over the next two years. So you're simply 01:06:36.580 |
going to see logo churn rates, say among small businesses go 01:06:40.220 |
from like a historical norm of 15% to maybe 25 or 30. 01:06:47.980 |
customer simply doesn't exist. Yes, that's what I'm going to go 01:06:51.380 |
means. Yes, the actual entity. Yes, logo churn means the 01:06:54.820 |
entity doesn't exist, then you've got seat contraction, 01:06:58.300 |
which is these companies are not hiring as fast. In fact, they're 01:07:01.900 |
doing layoffs. So they're simply not going to buy as many seats 01:07:05.260 |
of your software as you need to in the past. For the last 01:07:07.940 |
decade, we've had a tailwind, an enormous tailwind for software 01:07:11.100 |
companies of seed expansion, which is every year your 01:07:14.180 |
existing customers would buy more seats of your product for 01:07:17.060 |
their new employees. Now they're actually going to have fewer 01:07:20.180 |
employees or maybe headcount freezes. So they're actually 01:07:24.180 |
By the way, if you if you take all those three things, the deal 01:07:27.500 |
of the century was figma selling to Adobe for 20. Because if you 01:07:32.460 |
take those three things, I mean, oh my god, they just absolutely 01:07:37.020 |
top tick before any of this stuff was no. So today, Adobe 01:07:41.020 |
could probably buy this thing for like 7 billion, 20 billion. 01:07:45.580 |
So does that mean they try to do a breakup thing get out of the 01:07:48.300 |
deal? I don't know. But if I was if I was figma, I try to close 01:07:52.340 |
this thing ASAP and get that money. Yeah, yeah. 01:07:55.100 |
Yeah, you're right about that. And by the way, what I'm seeing 01:07:57.780 |
from founders is that they still want to grow 100% plus over the 01:08:02.940 |
next year. The problem is that the headwinds are going to be 01:08:06.060 |
intense. So if you're flying a plane, and the headwinds are 01:08:10.180 |
extremely intense, and you try to maintain your speed, you're 01:08:13.020 |
going to burn an enormous amount of fuel, you're going to be 01:08:15.140 |
incredibly inefficient. It's better to basically just 01:08:18.900 |
moderate your speed. Let the headwinds basically pass, we're 01:08:22.980 |
gonna have major economic headwinds for the next four to 01:08:25.540 |
six quarters, call it year and a half. It's okay to have a slower 01:08:29.460 |
growth rate, preserve your cash, don't burn up your fuel bunker 01:08:32.820 |
down. So what we're trying to do is we're trying to give 01:08:36.220 |
permission to our founders to grow at a slower rate because 01:08:39.620 |
they feel this enormous pressure from their VCs to grow at at at 01:08:43.700 |
insane rates. Can I build on this? I think freeberg said it 01:08:46.260 |
very well, the scan in venture capital is demonstrated in the 01:08:49.460 |
following chart. This is using Cambridge and our friend Brad 01:08:52.820 |
Gerstner helped put this together. So what is this? This 01:08:55.980 |
goes back all the way to 1997. And the gray bar is what venture 01:09:02.740 |
capitalists share with their limited partners as to how well 01:09:08.340 |
they are doing what is called quartile of venture capital. And 01:09:11.340 |
this is the top 25%. Okay, so this is this is a venture 01:09:14.620 |
capitalist. And you know, our returns have been consistently 01:09:17.820 |
top quartile. So instead of cherry picking anybody else, 01:09:20.540 |
I'll just use us but it could be Sequoia, benchmark, you name it, 01:09:23.860 |
we will go back we're in their launch, you will go back to 01:09:26.520 |
folks craft, we'll go back to folks and say, Hey, guys, the 01:09:29.580 |
total value of our portfolio is three times your money in 1997 01:09:34.140 |
is vintage. Okay, it was four times your money in the 2010 01:09:39.300 |
vintage feels really good. But again, the job of the venture 01:09:42.780 |
capitalist is to convert the gray bar into the purple bar. 01:09:49.940 |
And historically, there's been a decay. So for every dollar of 01:09:55.260 |
gray bar that you show, you typically only get 73 cents 01:10:00.340 |
The valuations that you get when you sell your company or goes 01:10:06.300 |
public, end up being 73% of what you marked at the peak what you 01:10:11.740 |
Exactly right. And the actual value of this purple bar going 01:10:17.300 |
back, you know, 30 years is 1.7 x. So just to put numerical 01:10:23.340 |
numbers on this, if you were a venture capitalist, you would 01:10:26.740 |
raise $100 fund at the peak, you would actually show that that 01:10:32.220 |
$100 became 200 and about $28. But when when push came to shove, 01:10:38.540 |
and when it was all said and done, you would return $170 01:10:42.260 |
back to your investors. That's the rough equation. So what's 01:10:46.340 |
the problem? Well, the problem as you can see in this chart is 01:10:49.860 |
right around 2015 which is all of a sudden, you know, what we've 01:10:55.780 |
started to see are these continually elevated gray bars. 01:11:01.540 |
Yes, this stuff is worth seven times six times five times. But 01:11:06.780 |
we have not seen the purple bars catch up. Now some people will 01:11:10.380 |
say, Well, yeah, but you have to give it time. And, you know, 01:11:13.060 |
this is it has to make other reasons look at. And all you 01:11:16.820 |
need to do is do what's called a regression. And you need to 01:11:20.420 |
regress these things to the mean and make the following 01:11:23.100 |
assumption. Assume for a second that this time is not different. 01:11:28.740 |
Assume that these historical averages 2.2 x 1.7 x holds. 01:11:33.860 |
Well, that's what the the black line here shows you can 01:11:37.820 |
calculate the area above the curve as the value at risk, 01:11:42.940 |
right, the amount of money we will destroy, because of all 01:11:46.460 |
these shenanigans that free bird just talked about propping up 01:11:49.340 |
marks, not willing to look at actual market clearing prices. 01:11:52.940 |
Well, if you do the math, the sum of the area above this black 01:11:57.500 |
line is almost a trillion dollars around the world. And it 01:12:01.180 |
is about $600 billion for us venture capitalists. This is the 01:12:07.020 |
dynamic that the private equity industry is going to prey on. So 01:12:10.340 |
if you saw Toma Bravo just closed a $32 billion round, you 01:12:13.660 |
know, Vista is raising a $20 billion round, everybody's 01:12:17.180 |
stepping into tech, they are going to destroy those gray bars. 01:12:20.540 |
You describe that as bottom feeding? No, no, they are. They 01:12:24.500 |
are the rational actor. Yeah. Okay. Who is finding the true 01:12:27.580 |
market? I think the private equity industry is unbelievably 01:12:32.660 |
precise, and talented in being dispassionate and telling us 01:12:40.380 |
They're just not shooting for the private equity industry is 01:12:43.540 |
going to be created by profligate founders. And look, 01:12:46.900 |
you could blame VCs for the high marks last year as well. They 01:12:50.260 |
were profligate too. But look, if you're a founder, if you 01:12:54.100 |
don't start acting in a more capital efficient way and 01:12:56.820 |
preserve your cash, your company is ultimately gonna be owned by 01:12:59.820 |
a private equity firm, and they're gonna make all the 01:13:01.580 |
money. Well, here's because because because when you sell to 01:13:04.900 |
them at a low price, all you're gonna end up doing is paying 01:13:07.740 |
back the liquidation preference. And then that private equity 01:13:11.140 |
firm that was willing to do or less, but that private equity 01:13:14.580 |
firm will be willing to do what you were not willing to do, 01:13:17.540 |
which was simply act, but cut your burn, cut your costs and 01:13:21.180 |
act in a more capital efficient way. And they will end up making 01:13:23.660 |
all the upside for your decade of hard work. Because you got a 01:13:27.380 |
basically addicted to venture capital and the high valuations 01:13:34.220 |
I'll give you an alternative, I'll give you an alternative. 01:13:37.060 |
The alternative is that the majority of acquisitions made by 01:13:40.300 |
private equity firms are not actually pure acquisitions, 01:13:43.420 |
they're bolt on acquisitions, meaning that these are companies 01:13:46.300 |
that are added to existing platforms that they own. So this 01:13:49.260 |
acquisition they're doing of Coupa, I think it's very likely 01:13:52.100 |
over the next couple of years, you will see like the playbook 01:13:54.660 |
and private equity includes not just cost cutting, but also 01:13:58.220 |
synergy building. And they typically do bolt ons and add 01:14:01.340 |
ons. And this happens across all private equity platform deals of 01:14:04.580 |
new products and services that can be sold through the existing 01:14:07.340 |
sales channel, the existing customer base, and as an add on 01:14:11.180 |
to the existing service or product that's already offered. 01:14:13.580 |
So one of the feet, one of the things that I think you may see 01:14:16.460 |
in Silicon Valley, over the next couple of years is a 01:14:19.020 |
rationalization away from funding feature companies, and 01:14:22.740 |
thinking much more carefully about what can be true standalone 01:14:25.220 |
product companies. Sure. And many of these companies that 01:14:27.420 |
have raised a ton of capital, and have gotten crazy valuations, 01:14:31.500 |
at the end of the day, they're more likely better equipped to 01:14:34.460 |
be a feature of another platform than they are to be a standalone 01:14:37.780 |
platform company of their own. And that's where the majority of 01:14:40.420 |
these acquisitions will likely end up going. Yeah, in the 01:14:44.060 |
private equity landscape, and they will be vacuumed up and 01:14:46.980 |
attached to existing platforms that these private equity guys 01:14:49.540 |
are building out. And by the way, just look as an example of 01:14:52.140 |
what Oracle did over the years, what Salesforce did over the 01:14:54.660 |
years, what Google did so many of these companies bolt on 01:14:57.460 |
acquisitions by bolt on acquisitions by building a 01:15:00.100 |
channel, building a platform, and then adding on top of that. 01:15:03.780 |
And I think that a lot of these guys are going to try and 01:15:05.540 |
mimic two critical points. Number one, what about the 01:15:08.300 |
bottom 75% of VCs? Oh, if you show that chart, just for one 01:15:12.180 |
more second. I just want to remind everybody, that is the 01:15:15.700 |
absolute cream of the crop VCs. The best. These are folks. I 01:15:20.220 |
mean, again, I'll just say us Sequoia benchmark, we've 01:15:23.140 |
consistently been crap. Thank you launch craft. These are 01:15:25.860 |
these are top chefs. Return Street. Thank the Lord. What 01:15:28.660 |
about the bottom 75%? Not gonna be able to raise funds, man. 01:15:31.260 |
It's over a lot of these people who raise first time funds in 01:15:33.860 |
the last three or four years. So it's also the company's 01:15:36.220 |
own. Because like, like, it's like the today is the now is the 01:15:41.180 |
moment for the sober founder and the sober venture capitalist to 01:15:45.060 |
sit and say, what is the real valuation? What do we need to 01:15:48.500 |
do to make sure that this company has a chance because 01:15:50.380 |
what SAC said is so true. Otherwise, all these profit 01:15:56.660 |
In order to win today, you're going to have to grind, you're 01:16:00.100 |
going to have to work 5060 hours a week, you're going to have to 01:16:03.180 |
be absolutely embrace the age of austerity. And you're gonna have 01:16:06.780 |
to focus on your customer, your product and your bottom line. 01:16:10.140 |
The age of excess is over. If you're not working 5060 70 hours 01:16:14.500 |
a week, you're not going to cut it in Silicon Valley. Also key 01:16:17.500 |
second point, profligate, extravagant or wasteful in the 01:16:21.300 |
use of resources just so we get the word of the day from David 01:16:24.020 |
Sachs. That's David Sachs is word of the day after a very 01:16:26.980 |
powerful bull. This is this is that one crazy. You see the 01:16:30.740 |
trauma boy for winter, we will, we will went viral. This is I 01:16:37.020 |
think, Elon's biggest, non obvious impact in this moment, 01:16:40.900 |
JTOW, here's your one answer to your question about what 01:16:43.860 |
happens to the the the bottom 75% of venture firms, it's 01:16:48.860 |
equivalent to what happens with the, you know, kind of, this is 01:16:52.380 |
the bottom of the top, the slide that I just shared, it's the one 01:16:55.060 |
we looked at a few weeks ago. And I keep referring to it 01:16:57.740 |
because it's just such a staggering like demonstration of 01:17:01.660 |
what people call the power law, which is how you know, kind of 01:17:04.660 |
excess returns accumulate to minority of investments. So just 01:17:08.340 |
a few investments make up the bulk of value that the you know, 01:17:12.500 |
market cap of 43% of companies that have gone public since 01:17:18.700 |
2020 is $750 billion. The market cap of three, the other 300 is 01:17:25.380 |
only $26 billion. And the cash that went in to the $750 billion 01:17:30.500 |
is 136. And the cash that went into the 26 is 107. And so the 01:17:36.380 |
cash that went in to generate that 26 billion 107, that's your 01:17:40.460 |
bottom 50%. And the top 50% put in 136 to make 750. And I think 01:17:46.460 |
it gets even narrower as you move further up to that top 01:17:48.780 |
quartile. So you know, it's just I can tell you what LPS are 01:17:55.020 |
This is the companies that went public. So this is also of the 01:17:57.900 |
top company of the top funds and the top companies that were 01:18:01.140 |
actually able to IPO. And so it highlights how much of a power 01:18:04.980 |
law actually plays through. And so the majority of these 01:18:07.740 |
companies as in Chamath, even in your chart, you show the top 01:18:10.940 |
quartile, the bottom 75%, or the bottom 50%. I've looked at this 01:18:15.740 |
data as well, of those various vintages are below 1.0, they 01:18:19.420 |
lose money for their LPS. Oh, consistently. And it's a cycle. 01:18:23.500 |
And so what ends up happening is the next generation comes 01:18:25.300 |
through, and LPS, they make a portfolio of bets. And they hope 01:18:29.140 |
that they make enough bets in the right VCs that their 01:18:32.180 |
portfolio generates greater than you know, market returns greater 01:18:34.940 |
than call it 15 20% target 15% target. But they're going to 01:18:39.300 |
expect that you're not I have an LP report. I'm out there raising 01:18:43.500 |
launch fund for right now. And I moved from like the accredited 01:18:46.060 |
the individual investors say that Oh, yeah, because you're 01:18:48.100 |
yeah, I'm on 506. Yeah, so I'm publicly raising it. And I've 01:18:51.820 |
moved on from individual investors $45 million in commits 01:18:55.620 |
after five webinars. Amazing. Now I'm talking to no, it was 01:18:59.500 |
amazing. It's just 506. He is going to change the entire 01:19:02.100 |
industry letting the you know, the masses have some access to 01:19:06.180 |
this capital. And this opportunity accredited and QB is 01:19:10.540 |
Do you have to do deal with everyone? One of them is easy to 01:19:14.500 |
It's incredibly complex, because you have a large number of 01:19:17.700 |
people and they all want to talk to me. So I did webinars, five 01:19:20.180 |
webinars, and it resulted in 100 commits, hundreds of commits 01:19:24.380 |
You'll be able to get all those capital commitments drawn down 01:19:27.980 |
when you need to, like you have to go ping a couple 100 people 01:19:31.580 |
We need to have more operations people and we only do for we 01:19:35.260 |
let them one thing. One thing you may want to do is like for 01:19:38.340 |
these smaller slugs is you can pre wire you can set up an escrow 01:19:41.940 |
account where you pre wire 100% of the capital. Yes. And then 01:19:44.660 |
you also don't have to you take it down when you're going to 01:19:46.900 |
deploy it. So you keep your IRR correct. So we're actually 01:19:49.380 |
looking into those solutions. I'll talk to you offline. But I 01:19:51.860 |
just did my first two meetings with endowments, etc, fund to 01:19:54.620 |
funds. The entire discussions right now are around what is 01:19:59.700 |
your secondary strategy? How are you getting in earlier, not 01:20:03.820 |
later? And how are you building a larger position? It is an even 01:20:08.300 |
like some of the QP is where sophisticated in our you know, 01:20:10.700 |
are in over 10 venture funds, the entire discussion governance 01:20:14.540 |
of these companies. Are you taking board seats or not? How 01:20:17.940 |
early are you getting in and building a larger position over 01:20:20.420 |
10%? And what is your secondary strategy? When are you going to 01:20:24.500 |
start taking some chips off the table? So the and I got to say, 01:20:27.580 |
if you're an LP, who didn't sell into the upmarket at all, and 01:20:31.700 |
you're on your first fund, you know, and you had all these 01:20:34.060 |
great marks, and they're getting the coming crashing down, 01:20:36.340 |
they're not going to deal with you. They just have too many 01:20:39.340 |
I don't think I don't think they've they've started to come 01:20:42.300 |
down yet. I don't think we know what the top quartiles really 01:20:45.740 |
going to look like over these last few years. I think that's 01:20:48.220 |
going to take four or five years to really sort out. Of course. 01:20:52.380 |
explain why Chamath just so people understand. Yeah, I 01:20:54.860 |
Well, I think I think that there are lots of valuations that have 01:20:59.900 |
supported huge TV PIs. These, you know, paper gains that have 01:21:06.900 |
allowed venture funds to raise enormous amounts of incremental 01:21:11.020 |
capital and new funds. And so they are going to try to wait as 01:21:16.500 |
long as possible before they're held accountable for that. And 01:21:19.740 |
the best way to do that is to not change the valuation. And so 01:21:23.300 |
it will happen slowly, it'll be a trickle of these things. And I 01:21:28.620 |
think that takes probably four or five years for it to really 01:21:31.580 |
sort itself out. But in the meantime, companies will still 01:21:34.660 |
need to get financed, companies will still need to get built. 01:21:38.220 |
That's why I think like the public markets, I think what 01:21:41.060 |
SAC says is true. Giving us a signal of what these true market 01:21:44.980 |
clearing prices are, will eventually slip into these, you 01:21:48.380 |
know, series D or E companies, because a venture capitalist who 01:21:51.500 |
has now taken some big write downs in one part of their 01:21:55.060 |
portfolio, I suspect will now be very open to selling to private 01:21:58.660 |
equity for another part of their portfolio so that they can 01:22:01.820 |
Totally. Totally agree. Yeah, it's gonna be rough out there. 01:22:07.700 |
Yeah, I just started season one. I'm third episode in. Okay, 01:22:12.180 |
what a treat. We won't say anything. But how great was 01:22:15.620 |
Oh, the rap was awesome. It's just incredible. 01:22:21.180 |
Just finished watching all of Handmaid's Tale, which I will 01:22:26.180 |
tell you is, that is a fucking stressful show. It's like, it's 01:22:31.020 |
like you're putting in work when you're done those episodes. 01:22:33.500 |
You're like, oh, emotional labor. You know, when they said 01:22:36.420 |
this emotional labor, watching that show is like, it could 01:22:39.740 |
could, it could not be more sadistic and insane. Oh, my God, 01:22:43.220 |
it is brutal. But you can't look away. Incredibly well done. 01:22:47.460 |
Alright, listen, this has been an amazing episode. And this is 01:22:52.740 |
news for the other besties. Friedberg and I have been 01:22:56.140 |
secretly collaborating. Now we have to a plan that we're 01:23:01.780 |
working on a joint plan for all in summit 2023. Because we are 01:23:09.540 |
I love it. The tip. I know. That's it. We don't need to 01:23:14.900 |
sacks. I'm a permanent No. That's all that's fine. We know 01:23:18.220 |
I love that I have sacks as my anchor on this one. I can always 01:23:28.460 |
Totally in this case, but power and influence is something that 01:23:33.220 |
you that's free bird celebrity. Friedberg had so much of a good 01:23:37.380 |
time at all in summit 2022. That his hatred of my producer fee is 01:23:42.780 |
less than his joy from the event. And we are collaborating 01:23:46.980 |
on super super gut. I have made up for my producer's fee by 01:23:52.140 |
using super gut and becoming a big proponent. 01:23:55.780 |
Announcing the debate. I've used the promo code. 01:23:57.620 |
Paid you for that. That's the quip. Oh, yes. Yes. You know, 01:24:01.700 |
this is no conflict. No interest to go along with doing the no 01:24:04.820 |
conflict. No interest. That's what's going on. bars. Amazing. 01:24:10.220 |
The only the only person that you haven't taken money from is 01:24:15.740 |
by the way, can I point out on the most loathsome person in 01:24:24.460 |
Do not name the company. Do not name the podcast. 01:24:30.820 |
and you win. No, I thought you won. No, he lost. He was 01:24:38.060 |
Jesse get on your chest. Not I only want the bracket. Do not 01:24:43.780 |
mention the podcast. We're not giving them any. Just black out 01:24:46.780 |
that in post. I want you to black out easy. I don't want to 01:24:50.900 |
give these guys any credit. So here we go. The worst person in 01:24:54.300 |
Martha and sax we would say it's a B podcast. That's run by 01:25:00.180 |
literal socialist. Well, look, Chamath got a very tough draw. 01:25:08.980 |
group went up against the Warriors with KD. It is no way 01:25:14.180 |
what about sex? Easy draw sacks versus in the most hated person 01:25:24.300 |
Andy Jassy is a complete gentleman. Andy Jassy is 01:25:34.220 |
That is I want to recount. No, I wanted to win. Saks. I want to 01:25:38.340 |
recap. This is election interference. Union busting. 01:25:41.140 |
This is election interference. I get something you're a 01:25:43.660 |
specialist in, I guess is worse to be a union busting Amazon CEO 01:25:50.380 |
This is ridiculous. I just want to point out that the biggest 01:25:56.620 |
travesty here is that I did not make the list. There are and you 01:26:03.340 |
Guys, shout out to producer Nick who just retreated. 01:26:07.300 |
With basically you basically pick the what is it the 30 most 01:26:12.820 |
relatively well known people in tech. That's what tilts Jake 01:26:15.300 |
Caldwell. This is terrible. Worst person in tech. I don't 01:26:19.740 |
this year constantly kowtowing to the media. You're you're 01:26:25.820 |
Stop with that. I need to be a worse human like you Saks. I'm 01:26:29.420 |
going to try my best this year to work against humanity and 01:26:32.820 |
society and be more loathsome than you. I'm really going to 01:26:35.940 |
redouble my efforts. Obviously, I can't catch up with you buy 01:26:39.060 |
into all their phony. I'm too kind. I got a big heart. I care. 01:26:42.220 |
I have empathy into all their phony. I know my empathy. But 01:26:45.100 |
here's the problem. These guys left me off on purpose. If you 01:26:48.740 |
want to pull up the replies between Andreessen and Bill 01:26:51.340 |
Gates. Oh, Andreessen. That's a lock. That's Andreessen. Of 01:26:55.820 |
course. Andreessen. A 16 co founder and man of terrible. 01:27:00.620 |
Is a world class shit poster. Bill Gates is hiding somewhere 01:27:06.460 |
nobody Bill Gates is doesn't tweet. Mark Andreessen blocks 01:27:10.540 |
unblocks he shitposts with the best of them. He's up there. I 01:27:14.780 |
mean, that guy's a dark meme Lord. Any other I mean, I 01:27:19.660 |
really I really sympathize with each month that you got your ass 01:27:22.380 |
handed to you there. That's just that's like going up against 01:27:25.060 |
the dream team. Hold on, hold on. Slow down, bro. You're not 01:27:27.620 |
even letting us read these things. All right. Give me a 01:27:29.260 |
joke. Twitter former idiot CEO versus the Airbnb CEO making 01:27:34.820 |
Oh my god, that's so well written. Brian. Hold on. This 01:27:43.540 |
weekend. Guy, guy who really tried to make us believe web 01:27:47.260 |
three was going to happen versus world coin and open of course, 01:27:50.540 |
critics and wins much more loathsome than Sam Altman. All 01:27:54.140 |
right, listen, free bird, you didn't even come. I want to 01:27:58.060 |
just congratulate free bird on an amazing the best science 01:28:03.340 |
corner ever. An amazing product is super gut that has helped me 01:28:06.940 |
lose weight. I feel great. And for you know, recovering from 01:28:12.140 |
whatever illness you had. All right, everybody. I love you 01:28:14.260 |
besties. Shout out to David Sachs. Love you guys. And we'll 01:28:19.020 |
see you all next time on the all in podcast. Love you besties. 01:28:29.380 |
we open sources to the fans and they've just gone crazy with it. 01:28:49.660 |
should all just get a room and just have one big huge orgy 01:28:58.740 |
because they're all just like this like sexual tension that