E117: Did Stripe miss its window? Plus: VC market update, AI comes for SaaS, Trump's savvy move - Transcripts

February 24, 2023

  • Favorite
  • Share
(0:00) Bestie intro: Jason's Japan trip! (1:04) Stripe's precarious situation: Did it miss its window? Breaking down its $4B tax bill, slowing growth curve, enterprise vs SMB customers, scalability issues, and more (23:07) Lessons for founders: How...


What time is it in over there? Well, we started at 8 a.m., so now it's 8.28. It's 8.28, I'm gonna be out on the slopes at 11. So I'll be out there skiing, I mean, Niseko in Japan, and take a quick flight to Sapporo, and then you drive two hours into the mountains. Yesterday, catskied.

There's an abandoned ski mountain in Japan. By the way, in honor of you, I grabbed a Sapporo from the fridge today.

Oh, very nice. Oh, very nice. This week's episode brought to you by. So they drive the catski up, and then you ski down, and it's all fresh track. So it's literally an abandoned ski resort, you know, during the financial crisis here.

I just asked you what time it was.

That's all I asked you. Cold, small talk, it's called banter. I thought you might be interested in your bestie's life,

but apparently not.

Let's get to the show. Everybody wants to hear the show. A lot of news going on in, you know, in our industry, there's been a big discussion about RSUs and stock options, both the cost of these things, and then there's another issue of people staying private for too long. If you remember, for folks listening, Airbnb, Uber famously took over 10 years to go public. People like Bill Gurley wrote about this, hey, you should get public. When the window is open, obviously the window is closed right now, or largely closed. Stripe, now people are speculating they missed their window. They have a $4 billion tax bill due to cover expiring employee RSUs. Those are restricted stock units, and at the same time Foursquare, a company from the Web 2.0 era, this is, you know, 10, 15 years ago, when they were very popular check-in software mobile location app, they are going to let their previous employees' stock option grants expire, according to the information. They issued these options in 2016, seven year window before expiration, more than 100 former employees will be impacted, and some of them are the very early team members, and this employee stock option problem is becoming acute because,

hey, people waited to go public. Basically what happens is you grant an RSU, which is effectively W2 income when it's realized with an expiration date, but that expiration date forces you to be public so that that RSU can be exchanged for value, then that's like a 10 year window. So then these guys have to go in and modify that date and push it out by another four, five, six years or whatever. That is a deemed event by the IRS that then creates withholding tax issues. So you then have to withhold tax on behalf of the employees, and so that collective number is the four billion

that Stripe is trying to raise. According to a leaked pitch deck, Stripe implied they needed 2.3 billion in capital by the end of Q1, 2023. They're working with Goldman Sachs to raise a few billion at a $55 billion valuation, that's down 42% from the peak of 95 billion in 2021. One wonders if they had gone public

what their valuation would be right now. Can we just say real quick why this matters?

Jake it like, Jake it like, yes. So anyway, why does it matter, Chamath? Yeah, why does this all matter? Like why do we care?

Thank you, that's where we're getting to, yes. I posted a link, this is a 2013 interview that Zuck did with Michael Arrington of TechCrunch. And if you go all the way back, the apprehension to go public was one thing that we really anchored to a lot at Facebook in the early days. And at the time, I don't know if you guys remember, but there was these arcane laws around the number of shareholders that you could have. And I think the issue specifically was that after 500 shareholders, you have to publicly release your financials. And so we did all kinds of things to make sure we never hit the 500 cap. And we tried to push the IPO date as far out as possible because we thought that it would keep people more focused. And then in 2010 or 11, I told this story a couple of times, one of the things that I was advocating for pretty aggressively was trying to launch a mobile operating system to compete with iOS and Android. And we had put together all this work and brought in Intel and AT&T and all these people. And it came down to the fact that we needed a couple billion dollars to float this thing. And we didn't have that money. So the only solution to that would have been to go public, but it wasn't the right moment in time and Zuck was uncomfortable with it.

A year after going public, one of the things that he said publicly in this TechCrunch thing was, wow, I should have just gone public sooner. It wasn't nearly the bad thing that I thought it was gonna be. And when you look subsequently at how much money they've spent in AR and VR, spending half a quarters of that cash could have given them the chance to disrupt Android and iOS in 2010 and 11, which in hindsight is obviously a no brainer bet, right? So even though I think we at Facebook were the ones to really put this in the water table about not going public, I think a lot of startups should have gone back to first principles to really question whether waiting as long as possible actually makes sense. So I was curious about the Stripe situation. So I asked my team to do a little bit of work on how would you value this thing if it were going public? And the interesting thing about Stripe is that it operates in a really transparent middleman business. So what's interesting about Stripe is that so many of the people in the ecosystem are public. And so what that means is you can build a pretty accurate mosaic of how well or not well that business is doing by interpolating all the other data from all of these other companies that are public and are forced to report. And so there's like a couple of really interesting things that jump off this page. And so the first thing that we did was we looked at what is the future profitability look like, acts of growth. And what's interesting is that you look at companies like Visa and MasterCard that are doing quite well and have done really well for a long time, but you look at this outlier and Ad yen And Adyen is probably the most obvious competitor to Stripe.

And the thing that is demonstrated here is how incredibly profitable this business is and how much operating leverage they have, which means that their OpEx is relatively constrained.

Because in terms of- When you explain the X and Y axis here, just so people who are listening can understand the chart.

Sure, so if you take the market cap on the X axis and divide it by their sales estimate, you get a multiple of the enterprise value to their sales. And if you look at the 2024 estimated EBITDA margin that they're forecasting, X of their long-term sales kegger, what you start to get a sense of is the operating leverage that this business has. And so all of this basically nets out to three interesting takeaways. When Stripe got underwritten at $96 billion, it's this data point right here where you see your Stripe previous round.

Five X enterprise value to divide it by 2024.

Divided over their long-term EBITDA exactly by their sales estimate. And then if you look at the $55 billion valuation, it's down. So what it looks like it's happening is appropriately so. People are doing the right thing, which is they're re-rating the stock, right? By approximately 50, 60%. But what's interesting is not where they are in terms of where they used to be, but the interesting thing is where they are relative to their most obvious competitor, Adyen. So Nick, please bring up the next one. So this is where things get really interesting because we looked at what was Adyen and what was Stripe's GMV per employee a couple of years ago before all hell broke loose in the private funding markets. And what you see is they were pretty equivalent businesses and they had roughly the same amount of employees. But this crazy thing happened, which is that if you look at the gray bar, this is the number of employees that Stripe has, it went crazy from a little over 2,000 to almost 8,000. So a four X of employees in two years.

Wow, four X in 24 months. They added 6,000 people. Just pause for a second on that. 6,000 people in 24 months and 700 days or so, right?

Three people and people a day. And if you do the same calculation for Adyen, it shows that they a little bit less than grew by about 75%. And then if you look at the growth of GMV and you impute how productive is each employee, basically this is the story of what's happened to Stripe and Adyen, which is that Adyen has found operating leverage, right? So they've found and maintained incredible profitability and Stripe has added an enormous number of employees. Now the question is why, right? So it turns out that these guys at the top line are growing roughly the same, except Adyen actually takes meaningfully less on a per transaction basis than Stripe does. And the reason is that Adyen services these large head customers. Think big, bulky folks that have huge amounts of transactions and so as a result have pricing power. And Stripe has some of those customers as well. In fact, they just announced that they're gonna process a large portion of Amazon's payment volume. But what's happened at the same time is that those kinds of deals aren't necessarily that profitable, and so you have to hire a lot more people to build a lot more features so that you can generate revenue from the long tail of customers, all of these SMBs. And this is the tail of these two companies which is that Stripe has some head customers, but many, many, many tail customers, Adyen has mostly head customers, fewer tail customers, and so the leverage in the business is that Adyen has most of these employees in Europe where the cost of these folks is much, much cheaper they have less than half the number.

And so as both of these companies continue to grow, you have one that has maintained, and frankly, raised their long term profit projections because they see it in the business, even at lower transaction costs, and Stripe, which is having a little bit more trouble. So I thought it was a really interesting expose. The takeaway for me is that if you were sitting inside the company, and obviously hindsight is 2020, the most profitable thing they could have done from an enterprise value perspective would probably have been to go public in 2018-2019, because they could have raised max value at max valuation, cleaned out all these options issues and have a huge balance sheet of cash with which to do stuff, whether it's acquisitions or other things. Because the thing that I struggle with is, is there going to be long term profitability in all of these tail products? Because if you look in the SaaS ecosystem, and SaaS, I'll hand the ball to you, there's companies building all this other stuff. And these point

products are probably pretty good too. SaaS, what do you think about edge-in going after the fat part of the long tail and then Stripe going after the long tail having many more customers?

Well, I think they're both viable strategies. And I mean, I've actually written about this, I wrote a blog some time ago called Enterprises versus SMBs, who's the better customer for B2B SaaS companies. And I think the sort of old school traditional view is that enterprises were always the best customers because they have the biggest budgets that translates into the biggest annual contract values, or ACVs. This provides the highest ROI on sales efforts. So now you can make a sales driven distribution strategy pencil in the first place, the prospects are easy to identify, you know, after all, if you're going after the Fortune 500, you can just make a list of the 500 companies. So I think the traditional gold standard was sort of the head, like you're saying, Jason, the enterprises. However, I think in recent years, it's become more popular to pursue the Stripe

strategy of the sort of more SMB. Why is that more popular?

Well, because first of all, the SMBs are more early adopters. So when you're a startup, and it's way easier to satisfy their standards, to satisfy their needs, their needs are less complicated, you don't have to have SOC 2 compliance and everything else.

If you saw- They're more risk taking, right?

Yeah, if you solve an immediate point for them, they'll just buy it. Okay. Whereas I think enterprises are more late adopters, they tend to be more skeptical of new software categories. Yeah, I think in addition to that, the SMB sales cycle is really quick. I mean, I'd say typically one to two months, you can close a deal, the sale itself is simpler. Like I said, the product requirements are simpler, and the low end of the market tends to be the most underserved part. So it's great to play where the incumbents are not. That's a traditional strategy as you go after the low end of the market, that's been kind of overlooked or ignored. And that's kind of what Stripe has done here too, is no one was really surveying these developers. So I tend to think it's a good strategy too. And the truth is, it's not one or the other. I think you just have to pick which of your battles that you want to fight.

And some starts to go after enterprises and some will go after SMBs. And it really comes down, I think, to founder market fit. I think founders who are better at sales, probably skew more towards an enterprise strategy. Whereas if you're more of a product founder,

you go after SMBs. Brilliant summary. Over time, Saks, for a company to thrive over long periods of time, do you have to serve as both? Or do you think you can stay in one of

those things and grow indefinitely? Well, what I've seen is that if you start the low end of the market with SMBs, over time, you can move up market. Because what happens is that as your product gets more and more sophisticated, and your company and your ability to execute and deliver gets more sophisticated, you can start satisfying the needs of bigger and bigger companies. So you start SMB, then you go mid market, then you eventually get to enterprises. I think if you start with enterprises, it's very hard to go down market because it's a lot easier to add requirements to your product than to actually strip complexity out of a product that's actually surprisingly difficult to do. So I think either strategy can work. Either you start the low end and move up market. That's the classic Clay Christensen innovators to the limit type thing. Or you just

start at the top and you stay at the top. It makes sense. It's just I mean, adding 10 people a day

over two years, that's a large number of people to add to a company. Well, in fairness to Stripe, they were very honest about this. And they were like, we overestimate it got confident and we overhired. And they found that all the coordination costs to Sax's point became too high. That's exactly what the colosseum said in their memo. So I think that they're trying to course correct and get back to this. I think the point that I'm making, unemotionally, I don't own Stripe nor Adyen. I don't have a horse in this race is more that in this market, specifically in these middlemen, highly transparent middlemen markets, it's very difficult to hide the cheese, meaning the ability to get to an extremely precise valuation model is pretty easy. You know, this was half a day's work that we did. And the point is all this data is out there. And so it means that if you're going to go public as a company like this, you have to be quite thoughtful about how outside in folks will value you because the terminal buyer is very, very sophisticated

and pretty smart about how to think about spaces like this. Freiburg, when you look at this, it kind of dovetails with the get fit Brad Gerstner, you on Twitter doing more with less employees. Zuckerberg again says he is getting rid of managers. He's asking managers to Sax's discussion about, you know, the layers of management that got added and added, where high performers would be would have five people put under them, 10 people put under them. Is it going to be? Are you impressed with how quickly the industry is responding to this new environment? Or are they not responding fast enough in terms of headcount revenue? Because now we're looking at revenue per employee. This is really never looked at that. It's been a decade since we

looked at that. There's a little bit of a different situation where it's about the scalability of a business. Like when I look at like the value that a business has created, you start first with like, can you make a product? Can you sell the product? Do people want to buy the product? And then, you know, can you make money selling it? And then there's this metric that a lot of people use, which is LTV to CAC, which is the lifetime value of acquiring a new customer divided by the cost to acquire that customer. But I think you can generalize that ratio to talk about business performance more broadly, which is capital deployed, which is typically what CAC is used in terms of growth on the denominator, and then capital returned over time, which can be the numerator. And so you can kind of think about that LTV to CAC ratio being something more broadly defined as something like ROIC or what have you. The question for the scalability of any business is, does that ratio, whether it's LTV to CAC or ROIC, return on invested capital, does it get bigger? Or smaller? Does it increase or decrease?

Does that ratio increase or decrease as you get bigger, as you spend more money, as you deploy more money? If it's getting smaller, then mathematically, you can resolve pretty quickly to the asymptotic valuation that that business will achieve for the asymptotic revenue that that business will achieve. And that's a very scary kind of circumstance, when a business that's tracking that metric starts to see that metric shrink. If that metric is growing, then you have a hyperbolic kind of moment and you can build platforms and add products and invest very heavily and take lots of risk and take lots of bets. When it's going the wrong way, you have two options. Number one is you have to make a change or pivot in the business to get it to go the other way. Or number two is you have to take advantage of that moment before the market finds out about that moment. Because as soon as the market realizes that that ratio is going the wrong way, your valuation multiple, what you're worth as a multiple of revenue or profit, shrinks dramatically. Because then the market can also see that asymptote and outcome. I think it's very often the case that one should, as a board member, as an investor, urge entrepreneurs, CEOs, founders, managers to think really clearly about that metric. What's the right way to define the denominator and define the numerator in our business and define that ratio over time? As soon as it starts tracking the wrong way, you have a moment.

You can either fix it or you got to go sell the business or go public and raise capital before the market catches on and your valuation shrinks. So when I see what Chamath's showing in this data and talking about this shrinking valuation issue for Stripe, it really, I think, highlights this important point, this broad point, which is did they miss the window? Did they miss the moment where suddenly the shrinkage is causing an asymptotic outcome for this business that makes investors a little bit like, well, I'm not as excited about that because there's no longer as much upside. And it might be time to devalue the company. And did they miss the moment to go public, raise a bunch of capital to go and try new things and hopefully pivot into a way? So I don't know enough about the business, but that's my broad assessment of this moment.

The interesting thing about that space, we talked to one of our friends at our poker game who who runs a large consumer facing business. And I don't know if you were there for that conversation, not Freberg, but I was you were there. Yeah. And one of the interesting things he said is we are at a level of scale where we just bid these guys against each other. And these things tend to now be loss leaders for them, which is to say effectively that cost structure becomes really important. So your CAC becomes very important because your LTVs are capped, right? And the LTVs are capped because these companies have enough negotiating leverage to say, well, if you want my business, here's the cost of doing this business, which makes a ton of sense if you're any large purveyor of services that require payment processing infrastructure. So one of the interesting dynamics I think we're learning in this market is how it's really not a market, right? There are segments and there's embedded profitability in each segment. So to your point, Freberg, this is the sum of at least three or four different LTV to CAC ratios, right? The tail looks very different, which is why you have to build a ton of features. And the head just wants pure play.

And it's all about cost first. The tail, because all of these guys want to pick up every nickel and dime that's on the floor, because for them, on billions of transactions is meaningful to them. It's an EPS miss or beat, right? For them, which has huge implications to their stock. This is a market that I think is going to be really

fascinating to uncover and peel back the layers of over the next few years. By the way, by the way, we haven't even talked about what Stripe does as a business. I know we have a diverse audience that

doesn't all come from tech. Yeah, so yeah, so Stripe is, will process your transactions, but they were the first people to make it as simple as putting a snippet of code into your app to process a payment, they compete with Visa, MasterCard, and those other places, they charge you a percentage of each transaction. So to Chamath's point, these larger answer devs, developers, five, 10 years ago, love this, because they get instantly at payments, right? It's sort of abstracted the whole thing just the same way cloud computing does, right? Storage at S3, etc. So you can kind of think about it that way. But a large whale in the system, Chamath, which you said Adyen has a lot of whales, not a lot of long tail stripe because it's developer friendly, and a snippet of code. They have this huge long tail anybody can do Stripe. In fact, people who are using things like sub stack or Patreon, I believe, they can just drop in their Stripe account. So people now, businesses have one have a Stripe account, they just drop it in there. So for me, that seems like huge potential in the future, because some of those could become

the whales in the system. And the long tail gives Stripe a lot of pricing power because there's no way for any one of those entities to have enough leverage to tell Stripe, hey, I don't want to pay 2.9% plus 20 or 30 cents a transaction. Whereas if you go to the head, I think Adyen is charging like 1.3 or 4%. So it's a wholly different market. And the pricing as a result

is totally different. Yeah. It's interesting to me, Saks, that we now are getting down to, you know, brass tax here, we're analyzing these money printing businesses and saying, what is the ultimate value of this? 1020 years from now, Chamath and I got a front row seat to that because there's a natural audience to every single service. For AOL, it was 30 million paid subs a month at I think the peak was 30 bucks a month people were paying Chamath. So at the top 2499. Yeah. So you know, you start looking at those numbers, you know, a billion dollars a month almost in just and it was a fixed cost business. But then boom, we just hit a ceiling and competition emerged emerged in the in the case of broadband. And then that business just slowly deprecated over time. So Saks, what does this moment tell you for founders, a lot of the listeners here and capital allocators, in terms of assessing businesses for the last and this will pivot into our next story. The last couple years, you know, if you were a first time fund manager, you were investing in 19 2019 to 2021, high valuations, those those funds, are they ever going to be able to throw a profit and then people were investing in those based on momentum, logo chasing.

This is now back to, you know, sharpening your pencils, Bill Gurley style

investing. Yeah, yeah, we've talked about it before, there's nothing new here. When you're in a boom, the only three things that matter are growth, growth and growth. And when you're in a downturn, the three things that matter are growth, burn and margins. It's not that growth stops mattering, it's just that people also care about burn and margins. And you know, the companies that fair the worst are the ones that have inefficient growth that basically burn a lot of money to grow. They have, you know, low or negative growth margins, they are burning way too much money. The burn multiple doesn't make sense, basically the ratio of money burnt to net new ARR that they're adding. Those companies get called out when all of a sudden you have a regime change like we're seeing now.

CAC is one of the early signs of this. Chamath, you and I saw that. Remember AOL was sending DVDs everywhere and CAC became $200 or $300 for every AOL subscriber. And then they were playing this funny accounting game. I don't remember this Chamath, where they were saying, hey, the LTV is like five years for an AOL. They were looking back at that number, not for good broadband coming. And so like we can totally spend $300 on TV ads to get a dial up customer 24 a month. And boy, did that whips on them. So listening to everybody talk here, I'm just like, wow, keep your eye on the CAC folks, the customer acquisition cost, how much you get you spend to get a new AOL, Netflix or SaaS product or a Stripe customer is critically important.

We look really closely at CAC payback, you know, how many months does it take to to pay back the cost of acquiring a customer. We don't look at that exclusively though, because you know, what expenses go into CAC is highly dependent on your accounting.

Unpack that for a second because there's the money you spend on a Facebook ad or a LinkedIn ad or any other great platform for driving, you know, customers to sign up for it. Yeah, you spend

Yeah, the new one. Yeah, you spend money on an ad or you spend money on a salesperson. Obviously that goes into CAC but then what about sales operation headcount? Does that go in? Is that opposite counter? Is that sales headcount? Is that customer acquisition or something else? So there's a lot of like subtle accounting decisions that kind of a big impact on with numbers. That number. Well, this is why this is why I've always recommended just looking at burn multiple. What I really want to know is, is how much money is this startup burning in relation to how much revenue is adding? Just like the ratio of those two

things, because there's no one with numbers, that number two things, because there's no one to high burned 100. Yeah, so this month we spent $300,000. And we burned $100,000. And then we added $100,000 in new customers ARR. So that's one x so that you have on your chart here burn multiple of one to 1.5 or under one is amazing or great. So yeah, but if you burn

200,000, 100,000, 100,000, I warn founders going into this year do not have a burn multiple greater than two because there's just so many headwinds right now that what happens is if you end up missing your revenue forecast, your burn multiples going to look terrible, we could shoot up to 345 and up. So it's better to have some cushion by going into the year being super

efficient on the converse side, Friedberg, if your lifetime value of a customer is incorrect, which we're seeing now with people canceling SaaS products or reducing the number of seats. Or in cloud computing, people are now saying, Hey, maybe I should take myself out of the cloud and host my own servers or some of my own servers and reducing their cloud bill. Cloud growth is slowing at Azure, Azure, cross the board, Amazon Web Services, etc. The it's still growing, but it's slowing the growth. So that LTV if you get that wrong, that can whipsaw you as well. Yeah.

Yeah, I mean, LTV, which is like, what do you make over time from a customer or however you want to assess it? A market deployment, it should be on kind of net cash, meaning like how much profit do I pull back into my bank accounts at the end of the day, after paying third parties and internal people. And where a lot of people, I think in models, I've seen on, you know, what's the lifetime value of a customer, they kind of take either revenue or just the simplified gross profit number. But the reality is, if you're scaling the number of engineers you need, because you have many more customers, and you got customer service calls, and you know, you've got to do custom deployments with your customers, all of that kind of adds up to additional cost. And some of these businesses, you see that the SaaS companies, for example, that all have gotten their multiples hammered. It's because the kind of microscope has come out at this point, just to some degree set aside general macroeconomic factors that are driving some of the multiple compression, but as the microscope has come out, it turns out that the efficiency of the business is not what everyone hoped and dreamed a SaaS business might be, that the efficiency of the business maybe looks a little bit more like either a services business or there's a big kind of scaling hardware component that the margin that you actually make for every dollar of revenue generate fundamentally is smaller than you know what you think it is. You have to add people to support and ops and new servers and all the stuff you're highlighting and a lot of that's excluded and then it doesn't take, you don't realize all that when you're small or when you're medium and growing you realize that when you're bigger and And when you're bigger, you're like, oh, wow, how do we get these costs out? Well, if we cut these costs, customer quality would decline, customers would churn, all this bad stuff would happen. So yeah, that LTV number is generally not right. And that's why I say it's much more about kind of a true ROIC calculation, which is how much capital am I deploying? And it's not just being deployed in marketing dollars, it's being deployed in other ways. And then how much capital am I making back net profit over time?

And I think that's the right way to always analyze a business generally. But like, particularly in businesses where it's easy to obfuscate either of those numbers, and they could seem like it's an extraordinary business, you can get hurt when you get bigger or when you're scaling. And in a market like this, where you're trying to go public, it's like, whoa, that really hurt.

You know, so I think that's a lot of what we're seeing. Let's talk about the other side of the table, Chamath. We've been living through a zero interest rate hallucination. Basically people were growth, growth, growth, logo, logo, logo, whatever. When they're making these bets, capital allocators now, we're back to brass tacks. Okay, what's the margin? What's the lifetime value? And is this actually real? Is there a real business here? Or is this just a grand hallucination? That hallucination exists not only on the founder side, but on the capital allocator side. This week, we had a interesting semi viral thread on Twitter, somebody named Tyler Tringas.

He's an early stage investor, don't know who that is. But he did a thread predicting A16Z, just to pick out one firm, was a zero interest rate phenomenon and an incredible machine to accumulate AUM as it's under management. And so what were your thoughts just writ large on the capital allocator side of this grand hallucination of zero interest rates?

I mean, I think it's a little unfair. I think this is written more just to try to generate views and clicks because you have to see the underlying return data to really have a sense of knowing is it I think it's fair to say a couple of things that there was probably two and a half or three years of capital raised in the industry that's going to get really put under pressure. And the reason is that there is not a lot of time diversity in that money, meaning people got it. And they put it into the ground right away. And one of the principles of having a more predictable return set of returns over time is that you leverage time, right? So if you had $100 and you want it to have a diversified stream of returns, you're much better off spending a dollar a month for 100 months versus $10 a month for 10 months. So just that thing will cause a lot of impact and headwinds for a lot of the capital in 2021 and 2022. Then the other thing you have to keep in mind is that over many cycles where we've had high rates and low rates and medium rates, our industry typically returns $1.60 for every dollar it raises, and that's over many cycles. And so if you believe that we're going to revert to the mean, out of the trillion dollars we've raised, maybe we'll return 1.6 trillion. Now, that sounds good, except the problem is that 1.6 trillion is marked at 5.5 trillion. So you're going to have to give back- So there's a lot of pain. You're going to have to give back a lot of paper profits in order to get back to that 1.6 and be okay with it.

And the question is, what has happened in decision making in the meantime, meaning how many people did you hire? How many deals did you do that you regret? And then how does it change your psychology and how you treat the next investment that comes over the desk? Can you separate yourself from these bad losses and not be on tilt and make a good decision?

So you had a terrible two day session, like Phil Hellmuth did last week, losing $350,000. Can you play the next week and not be on tilt and start to build back your stack and make 30,000 a night for 10 nights or 10 of the next 20, 15 sessions or whatever it is. It's actually had a rebuttal or something you wanted to add to this?

No, no, not really a rebuttal. I mean, look, I think if you're gonna be intellectually honest about it, I think that 2021 is gonna be is gonna likely be not a great vintage for VC. Why? Because the valuations were just yeah, the valuations were just really high. They've come down by what at least 50% on average,

maybe more more, maybe more. Why? Because I mean, 50% now, but you still have more medicine to take. I think when you look at

some of the businesses, you know, a lot of these companies are growing into their valuation. Look, I think for any given set of companies for any portfolio, the most important thing is what's in the portfolio. So if in 2021, you had the founding of the next Google or whatever that effect is going to swamp the effect of price levels in that year because of the power law. Again, the number one most important thing is just what's in that portfolio, what's in that basket. The second most important thing is the entry prices. And obviously, if the entry prices are twice as high in a given year than they are, and every other year, and twice as high as what the exit multiples are going to be in 10 years, when that portfolio becomes liquid, that's gonna hurt the returns. But we won't know which of these effects predominates until five years from now when we see, you know,

I mean, when I saw that tweet thread, I thought, maybe this is an issue for some venture firms, but we're not going to see even the inklings of it for another five or seven years. Takes a while. That's a problem that may manifest itself in year 10. And between now and then, any firm that has a good track record of returning capital, or frankly, has a good brand and good marks, will still raise an inordinate amount of money because this is an asset class that I still think on the margins is a more of a must-have asset allocation than on the margins, I'd just rather ignore it. Because it is the future of how GDP will get created.

And so everybody kind of has to pay attention. Imagine if in 2021, the next great mega outcomes in AI were created, right? Because those founders were just slightly ahead of the curve. They were like a couple years ahead of the curve. If those create the next whatever trillion-dollar companies, Google, Apple, Google, whatever, even the fact that price levels were 2x what they should have been, won't matter. What'll really matter is the distribution. There'll be a bunch of bad portfolios, there'll be some really incredible ones.

And that's the way it always is with venture. The thing to keep in mind is in 21 and 22, rates were still effectively too low. And I think we did this analysis, Nick, you can throw up that thing, but it's not correlated with big outcomes. Those vintage years, 2023 is the first vintage year where we're actually starting to see high enough rates that have historically generated that kind of return. And so I do agree with you, David. I just think it's shifted out by a couple of years, 23, 24, 25. Those can be some real power law years, I think, because we're going to have, just based on what the Fed is saying, 5.5% interest rates for the foreseeable future, which is a huge number.

That's a huge, huge, huge, huge free rate. I'll tell you what that is. You know what it is though, Chamath? I think, to build on your point. And Freiburg, I'll bring you in after this. It creates an environment in which discipline on all sides of the table, boards, management teams, investors, rank and file, everybody has to be focused. Everybody has to have sharpened swords. And that little bit of headwind is, and the ability to raise capital being harder, is building more reserve and more resilience and grit in this set of founders. It's kind of like parenting in a way, like if you are too permissive, you give too many options, kids aren't disciplined. And now this group of entrepreneurs I'm seeing who haven't given up, my Lord, are they becoming animals in terms of like pure samurai, in terms of how they're running these businesses. Anything that's not efficient, projects that were the third or fourth most important project, cut, cut, cut. Now it's taking them 18 months, Freiburg, to maybe get discipline.

But maybe you could speak to the next three years and the opportunity for investing in this cohort. Because man, that last cohort is going to be really, really challenged. And they'll probably do 6% returns, just like your money market account can do right now, five or six, or what bonds can do. But this next group, man, we're seeing dogged entrepreneurs who are focused on reality, and there is no hallucination now, that this is going to be easy. There is no grand delusion here.

What are you seeing in the market with founders?

That is in the market with founders. If the market average return in venture in early stage investing is going to be 6%. Remember, it's not evenly distributed. So 80% of funds could end up having net negative, real returns, and 20% make money. And then there'll be a very few that will make real money. And that's the nature of having a very kind of low average return on the industry is, there may be a lot of wipeouts on the investor class. Folks that have only had one or two funds and then just got blown up in the cycle. I think that there's two groups of companies out there. One is companies that obviously have been funded and are doing stuff and are active businesses, and they've raised money in the past. And that's where there's going to be really ugly times. I've mentioned this in the past, but I do think that there's a significant number of these companies that if they were to be truly valued on first principles in private markets today, they'll get valued as at a value that's less than their preferred equity. which means that there's a difficult restructuring needed in the company, and not everyone's gonna be willing to embrace that.

So that's what's gonna trigger a lot of the wipeouts in the market. It's not like the businesses are valueless. It's that the capital structure makes it difficult to refund them, to fund them and continue their operations. Now, for all the new businesses, as you highlight, man, there's so much extraordinary leverage out there. Left and right, I think we talked about this maybe a year ago, that there was a big bubble coming in AI. But I mean, left and right, nearly every market, every segment, you won't see a pitch deck that doesn't have those two letters in it. Right. I mean, I'm sure you guys find new pitch deck.

It does feel, I'm telling you, it does feel, it is hard not to feel like you're a little bit of a lemming if you buy into the AI stuff, but I will say that the use cases we're seeing are really pretty incredible. Totally. I didn't feel this way with the last couple of waves, like the whole web three thing never totally made sense, and crypto always felt a little bit speculative, like kind of unsure, but the AI thing seems like it's gonna deliver real value. And I'm seeing like already three major enterprise use cases. Number one is just auto summaries, like being able to summarize very quickly a thousand articles or a meeting, spitting out a summary of what just happened in a meeting, and it could break it down between a recap and action items.

It just does all the work for you,

pretty incredible. Second thing is like in-app customer service, kind of like a co-pilot, but there's no reason ve contact customer sport anymore, because you can just ask the AI inside the app, and why would they have a contact customer?

And the AI will get it right, and they'll be faster, right? The AI's like a power user.

They'll get it right. It's like a power user who's sitting next to you, it's your co-pilot and is making you much more effective in the app. And then the third thing we're already seeing is auto complete for everything. I mean, it is like bonkers how, You know how you get like little type head suggestions in email, but it's like two or three words. The AI is gonna be able to do type ahead for any content type, paragraphs of time,

to-do lists, tables, it's so paragraphs of time. It's bonkers. You see it in Google, you see it in Google Sheets now. Like if you type, you know, equal sum, it's like, oh, here's what the seven most likely things to happen next are. In which case, it's kind of like, you use the chess.com app. I don't know if you've used it with like the heads up display, where it's showing you the different moves and this is a book move versus this is not a book move.

Go ahead,io. Let me make a prediction. All of the things that you guys said, I think our incredible consumer surplus business opportunities, which means that the ultimate winner is us. And we're gonna become as you said,

incredibly, incredibly productive.

As you said, the consumer and more leveraged in how we spend our time, which will allow us to do all kinds of other interesting things with all the time that we save. That I think is almost now a certainty. The problem with consumer surplus businesses is oftentimes there is no money made in the funding of them. And really where the money is made isn't enabling it. So for example, so far what I would say is there's very little money that has been made in AI. There's been an enormous amount of money that's been made by Nvidia. And the reason is because they are the pick and shovel provider into the industry. And so that's an example. AMD I think can also benefit. So the silicon players seem pretty obvious here. Maybe some of the cloud players, the problem is the cloud players are trapped inside of other big companies with many other business models. But I just want to put out there that I think David, you're right that the consumer 100% wins.

But economically, it's not clear to me that there is a winner that is venture fundable. Well, hold on a second. Yeah. The Levi Strauss's of the world, right? In the gold rush, the people that made the picks and shovels in the jeans are sure to make money. Yeah. And the people that panned for gold is much more speculative and harder to see right now. Yeah. I disagree with that.

Well, yeah.

Yeah. Yeah. I disagree with that. So I think you have a point that, so I mentioned three use cases, I think are killer use cases that we're already seeing demos of today. And when you look at them, you're like, okay, this has real applicability. I mean, the AI is going to be, it's going to powerfully change our work lives. I'm just focused on enterprise. So now I don't know who benefits economically from that. That functionality that I mentioned, I think is likely to be pretty commoditized pretty soon, but it's going to be incorporated into lots of different apps

in ways that are hard to predict right now.

Well, I think that this AI revolution is going to do for SaaS what mobile did for a lot of the web 1.0 companies, where like for a lot of these web one companies, they were either disrupted by mobile or they're turbocharged by mobile. So you think about Facebook, it successfully made the transition and mobile made its business so much better because people were just using it a lot more on their mobile devices. There were a lot of other businesses that just kind of fell by the wayside because they just couldn't make the adaptation from desktop to mobile computing. I think AI is going to be like that for SaaS, where there's going to be a lot of SaaS products that are just absolute turbocharged.

Like you're 100% right.

Like you're 100%. If you can incorporate the AI into your SaaS product, put in a co-pilot, put in autocomplete and all sorts of other forms of value that we're just scratching the surface of, you're going to be able to deliver so much more business value. But if you're not able to do that

and somebody else can, then you're going to get disrupted. Look at some of these enterprise spaces, like take something like APM, right? Like application performance management. That's an entire ecosystem of enterprise companies. It's probably 10, 15, $20 billion of collective market cap. And I'm just going to say something, not to defend anybody, but like that can mostly be automated by AI. Those are simple heuristics that can be embedded in a way that's completely novel, where this code library just gets dropped in and all of this stuff happens relatively automagically now. So there are all kinds of other sectors, to your point, that get crushed. Then the question is, who provides that layer now for free in their existing SaaS toolkit or their product that now all of a sudden captures more value as a result? And they can sell it for pennies because it's incremental to them

in terms of their margin in revenue. I think you're right. Hardware wins. I think cloud wins big because if you keep adding to these models and one's 10, 20% better, people are going to be willing to pay for that. But then when you think about consumers, whether they're enterprise or actual consumers, I believe, Chamath, this stuff is going to provide so much value that people are going to take their wallets out and be more than willing to spend for it. It's more valuable than Netflix subscription. I disagree. Okay, I'm going to take this side, but imagine you take your videos of you learning to ski and you put it into an AI coach and it's like, here's how to, and it just draws on it. Here's how to be a better steer. This is going to blow people's minds and you'll be more than willing to spend 25 bucks a month

on that.

And it just draws on it. I disagree with that. And the reason is because we've spent now two decades and that's a lot of muscle memory to unwind of people that have been consistently given more for less. And I think that we shouldn't underestimate the expectations we've all collectively created by building software tools that have that inherent deflationary aspect to them. And so I just think that it's a very high, high bar, I still think there are subscription services to be built, I don't disagree with you there Jason. I just think that in general though, the de facto business model that we've created in tech is more for less. And we've used technology to give us operating leverage to create margin structures that other companies couldn't copy.

And I think that AI accelerates that not changes it. I think it's going to be the opposite. If you look at Netflix, if you look at Disney, they've been raising prices, providing more value. I think that this is going to provide so much value that the incremental 10 bucks a month, five bucks a month per employee is going to pay off so much that this could be if slack or like some presentation software, there are a lot of people who are making PowerPoint, AI PowerPoints, where it makes you a new deck, or a figma with AI, these things are going to be so powerful. People are like, it's totally worth an extra 100 bucks a month, because I can get rid of another employee. This one employee can now do the work of three. Fuck it, man. I'll give you $1,000

Some inflation really good a model thousand, some inflation really good a model, if you just added the LTV of that software company is going to make more money. I'm just

saying it's deflationary, that's deflationary company It's gonna make more money. I'm just saying, okay, it's deflationary on the entire economy. But that software company that figures out how you can fire to accountants and keep one and make them as good as you know, three, yeah,

you're able to charge higher self where Yeah, you're able to charge the software, right? You're selling consumer

surplus. Okay, I think we're in agreement. Freeberg sold into silence. You want to chime in on this? You still with us? Something sounds

technology drives prices down.

Well, technology is about doing more with less, right? It's about doing more with less and the AI helps you do so much more with the same amount of time or less time.

I think your whole point about Disney and Netflix, etc, is because they aren't, you know, innovating on either sides. And so in order to drive earnings growth, they're having to raise prices. But that doesn't know to the benefit of technology.

They're innovating massively, they're adding massive features to their products and massive new shows. I mean, I think there's pricing power in this AI. That's just my belief. I

could be thought about leverage. Yeah. I mean, look, I think I think your point like so my general rule of thumb, thumb on technology is the technology creator, the technology company should generally be capturing about one third of the value that they deliver to the customer. Unpack that why? Where do you come up with that? So I mean, it just kind of where I'm giving example. Yeah. Yeah. So like let's say that you as a food delivery company, you have to pay a human 10 bucks to deliver food from you. Now let's say I run a robot. My amortized cost of running that robot is two bucks. So it's eight bucks cheaper, or call it $1.

So it's

it's $9 cheaper. I should charge you four bucks and give an

example, you know, because four bucks is super competitive with the existing market. And it'll keep me competitive against the other automation companies that are going to start to emerge. It's just kind of how market dynamics end up working out. If you charge too much, you're going to invite people to come in and compete with you. If your technology commoditizes, remember, all technology commoditizes over time. And if you don't charge enough, you're not going to make enough money to be able to reinvest in scaling your business and doing more kind of interesting things as a platform. So, you know, generally, AI provides more leverage to Sax's point, if I can build an application, I don't know if you guys have seen these incredible UI apps that are built in AI now, where I can say, with a prompt, hey, make me talked about it two weeks ago. Yeah, right. Make me a dog walking app interface. And it builds like the three steps of the dog walking app, and gives you a bunch of options. And you can pick the one you want, I would typically have to pay a design firm $50,000 to do that work for me. So if it be AI is doing it automatically, you know, I should be paying, let's say $15,000 for that product for that capability, the margin on that is 100% try 50 margins, right, whatever it is very low.

And the margin on that's 100%. Whereas the margin on paying people to do design work as a design firm is very, you know,

not not a great margin, you're having to pay stuff. Yeah, we talked about it two weeks ago. Yeah, I figured it out. You know why we're having that we're working it out in our heads right now. One group of us is talking about comparing AI software and AI services to the existing software stack. And on the other side of the discussion, we're comparing it to the humans who are currently doing that work. Imagine the 6% that two brokers get, you know, doing a the sale of a million dollar home in that 60,000, and AI could negotiate that and find you a better home and sell your home for the optimal price for less than that 60,000 What would you be willing to pay for that right? And the same thing with the designer of the logo.

I don't think that's how it's gonna play out exactly JCal because to completely eliminate a job function, you have to do you know, 100% of it. And you have to do it, you know, 100% of the job function as, but as well as or better than the human. Whereas, I think as opposed to a model where you still have the human in the loop, but they're much more

productive because they're working with an AI. They're augmented. They're augmented human performance. It's more the Iron Man model. So I think that's more effective.

Yeah. So I think if there's a job reduction, it would be more the case where they've got a team of five accountants and they go to two or three because now they're just much more productive. I don't think they go to zero. That's my sense anyway.

I look at outsourcing as a possible corollary to this. You remember when you moved the accountants to Manila where they're knowledge workers there and it knocked out half the price two

thirds of the price, whatever it was. This just feels like that on steroids to me. If you have a business model like Infosys or Tata or one of these things that's levered to utilization rate, this is the most obvious way to basically add many potentially percentage points, if not tens of percentage points of utilization to your business. That's all money, free money for you, right? Because now you'll have fewer people, they'll be more utilized and they'll have more leverage because they'll be using a bot or some AI agent to help them write code, write unit tests, all that typical stuff that right now you outsource. And even if you pay a marginal cost, you add the labor arbitrage to technology arbitrage. Now all of a sudden, these businesses look really, really interesting.

Yeah. It's inspiring, man.

I think customer support definitely gets revolutionized, right? Because the initial- That's a no-brainer. You know, the first line of defense is going to be the AI using text to voice, and it can choose what language it wants to output to, what accent, so you'll never know that you'll think you're talking to someone locally.

That's a no-brainer. Literally, you'll be in 50 languages with the right answer, and you don't need to build up that entire group. I mean, I think we're underestimating in some ways what's going

to happen here, what's going to happen here. But my point is, I think that a lot of the customer support inquiries just go away, because the help, the assistant gets built into the tool directly. Yeah, it's coaching you as you go. Yeah. If you can

just ask it, everyone gets the point of the inquiry. It's coaching you as you go. People do that right now on YouTube. If you just type the question into YouTube and you find the video, that takes five minutes, but you're saying this is going to take 15 seconds, Saxe, because it's going to be right there.

I think what Saxe said before is hugely important when you think about how AI touches non-technology businesses. What he said is the boundary condition, which I think is right, I think he nailed this, which is the boundary condition for AI to replace a human is where the threshold error rate of that AI is the same or less than the human. If you look at very complicated markets, where does regulatory capture rear its ugly head? It's in allowing humans to be error-prone and you can't do anything about it. Take healthcare. If you go into a hospital, there's a certain error rate in every surgery, right? There's a certain error rate in the things that happen, but there's probably a whole bunch of ways in which that entire infrastructure can be made much, much better with AI, right? A robot that does laser guided precision surgery, characterizing tumors with 100% accuracy, so you always get 100% of the cancer out when you go and get surgeries done. All these things are possible now. And all of a sudden, you take these error rates that can be as high as 20% or 30%. So for example, breast cancer surgeries, the dirty secret of our healthcare industry is that has a 30% error rate. That can and should go to zero.

And now all of a sudden, so these highly regulated markets, I think, can become much, much more efficient and leveraged and pass that consumer surplus on to people. In that case, it's healthfulness, which I think is a big deal.

This is so interesting. I did my Pranufo scan. Incredible. I mean, I got all the videos, I got all the loops, I went to the one down on El Camino Real, it was like going to a spa in and out. No big deal. But I got the results. And it's like, oh, here, here's a tiny of little things that are not worth cutting your body open to look at. But just so you know, your knee your shoulder, your kidney, there's a little pollop here, there's a little polyp, or whatever, there's a little growth here. But let's see in two or three years, just monitor it. And I'm like, like, Oh my God, I'm so grateful. If this thing gets down to like 500 bucks, which it obviously will or 1000 bucks, and everybody's doing it, and then all that data is in there. And then the AI is looking at it like you're saying, I mean, the

AI detected, I did my new vote scan. Yeah, how do you go

credible? I mean, the early detection was the AI able to tell

the doctor how full of shit you were? No, you know, you're not supposed to eat for four hours. So they they didn't get an accurate reading on behalf of BSI. There's your cold oven, everybody. Yeah, I here's a really important clip for founders play the Steve Jobs clip. This is super important

when looking at web three versus AI to Saks point, you've got to start with the customer experience and work backwards to the technology, you can't start with the technology and try to figure out where you're going to try to sell it. And I've made this mistake probably more than anybody else in this room. And I've got the scar tissue to prove it. And I know that it's the case. And as we have tried to come up with a strategy and a vision for Apple, it started with what incredible benefits can we give to the customer? Where can we take the customer? Not not starting with, let's sit down with the engineers and and figure out what awesome technology we have. And then how are we going to market that? And I think that's the right path to take.

Can I ask you guys a question? I sometimes I go down these rabbit holes, I'll watch hours and hours of Steve Jobs clips. What do you think makes him so calm? Doesn't he just strike you as incredibly just like calm and like comfortable with himself

and just aware I know what it is. What is it was so much better and aesthetically building product than anybody else? He when you think of that PC era of no taste beige boxes, and everybody having no style, and just no swagger. He was studying, you know, German design, Buddhism, tripping on acid, and like just understanding the universe at a level that Gates and the other contemporaries weren't, they just weren't as transcendent in understanding product design as he was. So it was like, when you were saying you were playing poker with a bunch of four year olds or something. That's the analogy. He's just on such a different level that he's watching people make, you know, as 400 and, you know, IBM, PS, whatever, like, just garbage computers, garbage operating

systems. And it's just like, the thing is, like, if you look at any era, just the way that he communicates, there's just a level of calm. I don't know how to describe it. You understand what I'm trying to say, like, he he just seems like he just sees through all the noise, like, he's seen through the matrix, like he's unplugged himself.

Saks is unimpressed. Okay, there you have it. No, I'm very impressed with Steve Jobs. I think he understood product

development better than anybody else. Yeah, clearly, that's it.

I mean, my favorite Steve Jobs passage is the one where he describes the John Scully disease. Do you guys remember

this? Yeah, no. Oh, here it is. You know, one of the things that really hurt Apple was after I left, John Scully got a very serious disease. It's the disease of thinking that a really great idea is 90% of the work. And if you just tell all these other people, here's this great idea, then of course, you can go off and make it happen. And the problem with that is that there's just a tremendous amount of craftsmanship in between a great idea and a great product. Yeah. So true.

Yeah, I mean, I tell people it's like a rugby scrum. You go, you know, you got to get a whole team to get the ball down the field. It's not like one person put the ball down the field. And you know, they kind of maybe suggested a play. But once you're on the field, everything changes. And everyone's involved in getting it down the field.

That quotes where the name for craft ventures comes from. Oh, really? Oh, little known fact. Yeah. I didn't know that. Yeah. Section 230. We talked about last week, the Gonzalez versus Google case, the justices hurt oral arguments and plaintiffs seem to fair poorly quote from Scotus blog justice. Elena Kagan suggests that even if section 230 is not well suited to address the current needs of today's internet, such as such a task was best left as we predicted. Last week, I think sex you did. That's left to Congress rather than the Supreme Court, quote, these are not like the nine greatest experts on the internet. Kagan observes actual thoughts.

Yeah, I mean, this is just, I think really a quick update to what we talked about last week, the justices heard oral arguments, they seem to be very skeptical of the plaintiffs arguments, even Justice Thomas, who has written the most skeptically in recent years about the broad immunity that tech companies enjoy under Section 230 seemed surprisingly sympathetic to the theory that the Ninth Circuit Court ruled on, which is that Section 230 protects recommendations, as long as the providers algorithm treats content on its website similarly. So even the justice who I think was most likely to rein into 30 seemed to be more comfortable with what the defendant, which was Google was saying. So it looks to me like Google and big tech are gonna win this one.

Any thoughts, Jamal?

No, not really. I think I want to know what you guys think about Trump showing up with Big Macs and water in East Palestine. I mean, he is a genius. He beat Buttigieg to

East Palestine. I mean, he is. That was unbelievably pull up my tweet. I think this is the power we because Trump has been out of the public discourse. He's a media. He is a media savant. Literally, Biden is in Ukraine, saber rattling over air sirens that may or may not be true. They were fake. Who

cares? Who cares? Anyway, discourse for two years. He's

a media. They were fake. It doesn't matter. No, it doesn't

matter. No, it doesn't matter. No, we don't matter. No, we do. So we do actually. You're there. Okay. Because... Hold on a second. I don't need to be there. Because Jake Sullivan participate in a press conference. And he was asked by a CBS News reporter if the U.S.

gave the Russians any kind of heads up the president was going to be in Kyiv. And what Sullivan said, and I quote is, we did notify the Russians that President Biden will be traveling to Kiev. We did so some hours before his departure for de-confliction purposes. You know what de-confliction is? It's when the U.S. tries to avoid an accidental conflict. And, you know, Putin's not crazy enough to try and assassinate Biden. So the Russians were not attacking Kiev that day. In fact, they haven't attacked Kiev, as far as I know, for weeks. So these air raid sirens were basically just pure theater. But the amazing thing is that

if you know, we don't know. So we do. So you don't know if you don't know that Biden orchestrated is my point. People on your side. Come on, Jason. But that doesn't mean Biden pressed the button. So don't also take it to the other extreme. It was theater. Who

knows why the siren went off, but put it aside, also take it to the other extreme. This was a joint event between the Biden administration and the Zelensky team. They organized it. The whole thing was choreographed. How did that red carpet get there, Jason? Was that

an accident too? Okay, let's put that aside. You're acting like this is accidental. I mean, how long are you? Let me give you your GOP win. Donald Trump is a savant. And he went to America, to the place that we were reporting on the under reported story. People in East Palestine are being ignored. And he comes there to help the people of America. I give you all credit. Your guy Sacks did the most amazing media move in history. He went to middle America where people are suffering as opposed to a war that nobody wants to be in and spend all

that money on. We won't spend money, but we will go spend billions in. Let's put that aside and take the win.

You're acting like this is accidental. I mean, how long do you want to be in it?

Ukraine, go. All right. You want to know what this reminded me of? And you may think this is a weird connection, but it reminded me of the ending to the movie, Boys in the Hood. Do you remember what happens at the end of that movie?

No, I haven't seen it in years. Go.

Okay. It was 30 years old, but Ice Cube plays this character, Doughboy, and his brother gets killed. And at the very end of the movie, he gives this speech to Cuba Gooding Jr., where he says, you know, I turned on the TV and there was all this shit about violence in a foreign land. And there was nothing on my brother getting killed. All this stuff about what's happening in foreign countries, nothing about what's happening here. And then I think the most memorable line was, either they don't know, don't show, or they don't care what's going on in the hood. Right. So what's going on here is the people of East Palestine, Ohio, are being engulfed in a plume of carcinogens and toxins, and Biden is off pursuing this crusade in eastern Ukraine, and it's not just him. I'll dish out to Mitch McConnell as well. Mitch McConnell was on


Oh my God, those are the neo-cons of our neo-cons, yeah. McConnell was on TV saying that the number one priority of the United States right now is defeating Russia in Ukraine. It's not helping the people of Ohio. It is not securing the border. It is not solving crime in our cities. It is not making our schools better. It's running off and basically supporting this war in Ukraine. So both these octogenarians, Biden and McConnell both, they either don't know, don't show, or they don't care what has happened to the United States of America. He's a genius. But it's not even genius. I mean, it's so obvious that you go there. It is so obvious.

Nobody wants to be a forever one. And Putin just didn't go there, and Biden didn't go there. It's not genius, it's obvious. DeSantis didn't go there. Where's DeSantis? He should've gone. He hasn't declared yet. He hasn't declared yet.

It is so obvious. Nobody wants to be in a forever war. DeSantis didn't go there. Where's DeSantis? He should have gone. He hasn't declared yet. He hasn't declared yet.

No, he's still. Make a trip. I think the most senior Democratic person that went over there was Josh Shapiro, who's the governor of Pennsylvania.

He got there before Buttigieg. What is going on? It's a never-ending war. Nobody wants to fight a never-ending war. This is what got Bush in trouble.

This was the big critique. We're spending all this money over in the Middle East on these conflicts. You're talking about Bush senior. Bush senior. I think actually it's a good analogy. With Bush senior, Bush actually, this is in 1991, he won the Iraq war. That was actually a stunning foreign policy success because he actually didn't go too far. He didn't go all the way on the road to Baghdad the way that his son George W. Bush would, creating an epic disaster.

You're talking about Bush senior.

Bush 41 delivered a victory there, and he still lost election. Why? Because he seemed out of touch. He wasn't focused on domestic problems. The American people want an American president to focus on American problems. And even if Biden delivers some sort of victory in Ukraine, if he ignores these festering problems at home, then he is, I think, vulnerable for this re-election. But I think the truth of the matter is that this war is going to turn out much worse than the Iraq war did in 1991 because in 1991, we showed restraint and we knew what our vital interest was, and we kept our objectives limited, and we kept the timetable very short. What is Biden doing here? Biden won't tell us what the objective is. It's just whatever the Ukrainians want. He won't tell us what the timetable is. It's basically for as long as it takes.

And then meanwhile, this week, you had Kamala Harris go to the Munich summit declaring that the Russians are guilty of crimes against humanity, which that's something that we could have assessed after the war. Think about the incentives you're now giving the Russian leadership before we said that we just wanted them to leave. When you accuse them of war crimes, it implies that we're going to go chasing them all the way to Moscow. They're not going to want to end this war. They can be put on trial at the Hague. I mean, this is highly inflammatory. So, you know, this thing is not going in the right direction.

Yeah. And that was the thing I didn't like about Biden's speech over there is just he's escalating, escalating, escalating. Hey, we have to stop Putin. I mean, which you do. He did invade another country. He didn't cause three or four hundred thousand Russians have died, according to reports. Over a hundred thousand Ukrainians have died, according to votes. Neither side is given the accurate number because they don't want to demoralize their constituents. But the amount of suffering going on here is extraordinary. And I think it should be the West who is going to send Macron, send somebody from Germany, send some, you know, group of people to then go to Ukraine and work this out. But you don't need to go on your saber rattling. It was too much saber rattling for me.

It is de-escalation. We need de-escalation in these situations, not saber rattling.

I agree with you. I agree with you. I agree with you. But Biden has really painted himself into a corner here because before the war, he refused to take NATO expansion off the table. He refused to recognize the Russian interest in Crimea. And we gave no support to the Miska courts, which would have given some limited autonomy to the Russian speakers in the Donbass area. If we had just done those three things, there would have been no war. Biden refused to do that. He refuses to take foreign expansion off the table even today. So he has nothing to compromise with. He is dug in. And the problem we have now is that it's a loose, loose scenario.

If the Ukrainians keep doing poorly, because right now it looks like they're on the back foot, what is the United States going to do? Are we going to let them lose this war, or are we going to keep giving them more aid and step in? It looks to me like Biden now has invested his whole presidency in this, and he can't just let them lose, which means more escalation from us. Huge mistake. And on the Russian side, if the Russians lose, then they have an incentive to use nuclear weapons to rescue the situation. So it seems to me that both scenarios here are really bad and we don't really have a good way out of this. We're looking for some sort of magical Goldilocks scenario where the Russians sort of lose, but not enough to use nukes. You know, the administration has not given us a clear picture of what victory looks like here that's actually reasonably achievable in a reasonable timeframe at a reasonable cost.

What do we think of Freiburg of Xi Jinping making overtures and hey, maybe we should work towards peace. If you follow the money, he wants cheap oil. He wants this thing to end, and he wants the West to be buying goods from China. The West wants to sell a bunch of armaments. The military industrial complex is absolutely in delight of replenishing all of these weapons. Perhaps a little cynical to follow the money concept, but what we should take on the chessboard of Xi Jinping is going to visit Putin before Biden does. And he wants to build bridges, and we want to saber-rattle. What are your thoughts?

Isn't he getting like, I mean, China buys energy from Russia today. They buy oil at a very cheap price. So if I'm China, I want this to last longer, don't I? Why would I want to end this and then have Russia's markets open up? Because if their markets open up, the markets normalize to market prices. Right now, they're getting a discount. So I think they certainly don't want things to escalate. The question is, how quickly do they want them to de-escalate?

So if I'm China, I'm kind of probably playing a little bit of a middle line here. On sale.

I obviously don't want to see a big hot war. China's got its own domestic problems right now that seem pretty significant and existential. And having access to cheap energy seems like a benefit. Obviously, if there was significant conflict and escalation of conflict, that would be very bad from an economic perspective for China. So they're probably somewhere in the middle, like a slow resolution, let's say. I don't know. I mean, this is pure speculation. It's just me.

But Europe, Saks or Chamath, Europe isn't gonna buy Putin's oil anytime soon, right? They're now going to buy...

No, but he's able to sell it to China and he's able to sell it to India and the rest of the world. There was actually an article in Today's New York Times about how the West may be unified about Ukraine, but the rest of the world is not. The article was saying something that Chris the War said for a while, which is, we actually don't have the whole world with us at all. The BRICS countries are not with us. The emerging world, the whole southern hemisphere is basically not with us. They would like the U.S. to play a more constructive role in finding a peace deal, not like you said, Jason, saber rattling or escalating. So the rest of the world is not happy with us, and this is why the Russian sanctions have not been effective. I think the Russian economies had like a three to four percent hit. It is not the collapse that was predicted because there are enough other countries willing

to do business with them. Would this have happened, Chamath, if Trump was president? And how would Trump have handled it, do you think? Just game theory here. I'm just curious. Because Trump almost won, right? I mean, if Trump had won, what would this look like? Would Putin have gone in there if Trump was president? And how would Trump have handled it? Because Trump seems to think, I would have just told him, don't do this, and they wouldn't

have done it. I mean, this is the most obvious compliment I can give him. I think that he is exceptionally pragmatic on being anti-war. And I think that that is one of the most positive characteristics that he showed. He was really the only president, I think, in modern history, right, Saxeepu, that hasn't gotten us embroiled in a new war. No new wars, yeah. It is the best part of him, yeah. He's been incredibly, incredibly consistent. So I suspect that there would have been some kind of a deal. I know that sounds so ridiculous to say, but there would have been a deal. A deal, I actually agree.

He's a deal maker. A deal, he's a deal maker.

He gave a statement this week. He gave a statement this week.

He went to North Korea and met with—

He'll shake hands with anybody. Exactly. fired all of the deep state blob that started to position anything towards a conflict. So I think he would have shut the door so ferociously on Ukraine and NATO, and anybody that crossed that line, he would have tarred and feathered publicly. And I think the end result would have been that Putin could have found an off-ramp well before he invaded. Probably.

Totally. Yes, I agree. I agree with that. Trump blamed Germany for all this, right? He called it.

Well, Trump very early asked the question, why are we spending all this money to defend Germany when Germany has this big pipeline deal with Russia? It doesn't seem like they need our protection. They should just pay for it themselves. But I think there's a separate point that Chamath just made that is a really good point, which is Trump's instinctual resistance to what the deep state wants. And he actually said it this week. He gave a two-minute televised statement that was all over Twitter, where he basically made the argument that, listen, the reason why we're in this war is because the military industrial complex and the foreign policy establishment, they basically courted this conflict. And they are working at odds with the interests of the American people. It's actually a fairly radical critique. I don't think a major presidential candidate has run against the military industrial complex the way that he is now positioning himself. And let me tell you this, you know, I've said it before, he's not my preferred candidate. But if this war spirals out of control, either, you know, it turns into a even bigger conflict that draws us in, or it turns into a big recession, because I don't think we've seen the last of the supply shocks from this war. If we get a recession that Trump can, I think lay at the feet of this war, he's positioning himself to take advantage that this could be a silver bullet for him.

I don't think he has any other way of winning, but you know, if this turns into a big mess, Trump is positioning himself.

You have your tin foil hat there. Put it on for a second. I want to talk to tin foil sacks, tin foil hat sacks. Let's put the tin foil hats on here. Do you think Putin is escalating this as a way to position Trump to where Putin says he could say this during the election? Like, listen, you know, I would love to talk to Trump. And what if Trump goes and talks to Putin,

or does a phone call with him? So your theory is that Putin is escalating this into potentially a nuclear war to get Trump reelected. That's your theory. And I'm the Trump is favorable

to him. I'm just tin foil hatting it. The reason why this has occurred. No, no. Now that this has occurred. Not that he did the, he did the invasions. You're the one in tin foil hat territory. Tin foil hat corner at the end. Putin, the reason why he started the war for it that he would end the war to give Trump a win. How's he going to end the war for Trump? What are you talking about? During the election, he does a call with Trump.

And he says, you know, I talked to Trump about this. And I'd love to do some negotiations with Trump. I've always had appreciation for his ability to help negotiate things I would love. I would feel better about negotiating with Trump, who hasn't saber rattled and told everybody in the world that I have to be,

that there isn't regime change. So I know how you come up with these conspiracy theories, and then attribute them to me and call me the tin foil hat guy. It's just a joke. Dr.

Silver bullet. It's just a silver bullet,

and the economy goes off the rails. Because of this war, Trump right now is positioning himself to take advantage of that fact. And I would say things about the war this week. So it's not just Trump. But look, the thing you have to understand about this war is it's existential for Putin. It's existential at this point. He cannot back off.

And it fails.

Yeah. If this war is off, too. Yeah. It plays right into his hand. It's a pacifist.

And it's extra curricular for us.

Yeah. Yeah.

And that's why it comes down to. And that's why Obama said, back in 2014, that the Russians have escalatory dominance. They will always climb the escalatory ladder all the way up to nukes if they have to.

And the sooner we recognize that fact, the better off we're going to be.

I think the good news is that we are speech that he did where he kind of highlighted

ideas, that we are, I didn't see the speech. Was it good? We we've just talked about it.

If it was doing it, it was two minutes. It was fabulous.

It's just like, just mentioned it sound a lot like what we were talking on this podcast, which is he talked about all these generals that retire. He called up Victoria Newland by name!

By name! By name! By name!

He really did put her eye on the audience. Because I didn't see this. I'm on a different time zone.

It must have broken when I was asleep. Well it's a two minute video in which he, like I said, he attacked the military industrial complex and the foreign policy establishment for creating this war. And he mentioned Victoria Newland by name. Let me tell you something. It's going to be a very popular message. Yes, It is very popular. Newland is the FALCY of this situation, okay. The same way that FALCY was supposed to be protecting us from viruses and then funding the gate of Florida. We did a function research. Now we got a label! Let me tell you something

Victoria Newland

We just got a label COVID-19 You just got a label! Victoria Newland Rogers

Victoria Newland was suppose to be our Chief Diplomat with respect to Russia and Eastern Europe and what did she do instead? She ginned up this conflict. How? Ginned up. We backed an insurrection in Ukraine in 2014. Jason, if you didn't like the insurrection of January 6th, let me tell you, you aren't going to like the insurrection that she staged in Ukraine, because they brought in these Ukrainian far-right nationalists as the muscle. And that is what created all these problems.

Did he bring Big Max? Did he bring Big Max with him? Did you say he brought Big Max to,

too? He's Palestine? He brought fast food to them? Yeah, he showed up, we're fast food to them.

Yeah, you're ignoring what Zach said, but no, no, I got it. I am not disagreeing with him. I think

if you want to the roots of this conflict, Jason, nobody wants to be in a forever war.

If you want to represent the roots of this conflict, Jason, no, yeah, why he mentioned Victoria newland, he mentioned her because she was the State Department official who was responsible for backing this insurrection of a democratically elected leader in Ukraine in 2014 named Yanukovych, okay? Yanukovych was trying to, was doing a balancing act between Ukrainian nationalists and Russia, and it was a very delicate balancing act, and we basically toppled him. And ever since then, the relations with the Russians over Ukraine have been headed south. If you're wondering why Putin seized Crimea, it was in direct retaliation for the coup that we backed in Ukraine in 2014. This is the origin of the conflict. And, you know, if you want to understand where this comes from, you have to go back to this. And the fact that

Trump's willing to talk about it is pretty incredible. I think that the good news for us is I think that heading into June and the debt fiasco that's looming, I think we're going to, and I think this will help a lot, get distracted with domestic issues in the sense that it'll take some heat off of escalating all of this foreign adventurism. You know, it's such a scene, like this is such a scene from Wag the Dog. Every time there's something inside the United States that we should really focus on, we have this Wag the Dog moment where we get distracted by some adventurism abroad, and we forget and we lose sight. So we have this East Palestine thing right now. In June, we're going to have to come back to terms with this debt ceiling issue, which is a huge one, how we're going to resolve it. It's not clear. Just this week, the Federal Reserve basically said, hey, folks, we're taking rates to five and a half plus, and they're going to stay there. That seems like no news. People just seem to digest it and move on. It's really incredible how we just find, we're like, what is it, Jason, the dog that chased

the bumper and caught the car or whatever? People just seem to digest it and move on. Yeah, you caught the bumper. We've got plenty of big problems here in the United States, plenty of big problems. And I don't know that Wag the Dog works anymore, because I think the American people want, like I said, they want an American president to focus first and foremost on American problems. And even remember, Bush Sr. in 91 won that war and still lost re-election. Still lost. So I don't think Wag the Dog works anymore. It works for some short period of time, especially while the media are portraying the air raid theater,

that's eventually the people smarten up. You're so right. So that issue, think about Bush. Bush came off of the Persian Gulf War with like a 91 or 2% approval rating. I mean, we've never seen anything like it, but he violated a simple tenet of his domestic policy, which is read my lips, no new taxes, boom, lost. And it was not even close in the end. So I think you're right. I

think people really care about the economy. Go Nikki Haley. And how much debt do we want to go into over foreign wars? The only thing I ever liked about Trump was his policy of not starting wars and not getting into them. And Americans want to focus on our balance sheet voter right now. I'm voting based on who is going to be fiscally responsible. Me and Freeburg are in the same vote

here. We've got to be real careful in how we handle China because you had Blinken on all the Sunday shows, basically denouncing them, expressing outrage that they might support the Russians, acting shocked, shocked that they could do that. We don't even have the ability anymore to understand that other countries do things in their own interest. And we can't accept that. And instead, we act as if foreign policy should be conducted according to this morality play that we've created. And if you don't do what we think is right, then we're going to express all this outrage and condemnation at you. And somehow that's going to get you to violate your own interests. That's not the way the world works. And what we're doing right now is pushing China and Russia together into a new axis block. This is very foolish, very foolish. Even during the Cold War, we work to keep Russia and China apart. And whenever you think of those regimes today, they were much worse back then.

Remember, the Soviets, you had a Stalinist regime, the Chinese had Mao. Those were the two of the three biggest mass murderers of the 20th century. And Nixon and Kissinger still went to China and shook Mao's hand and toasted him because it was important to keep China and the Soviet Union divided. And what are we doing today? We are basically pushing them together with all of this condemnation and outrage. It is not a smart

strategy. Now, I can't disagree. We need to be building bridges with India. That's a key, key relationship. And China, I don't know why we're not figuring out what we have in common.

This is poisoning our relationship with India. India is the biggest democracy in the world. And our relations with them have gone south since this war, because they have a friendship

with Russia that goes back around. I mean, I would rather see Biden go to India and start

building some bridges there. Yeah, I agree. I can't disagree.

Jacob, how's your fundraising going for Launch Fund 4? Thanks for asking. It's a great question. We're doing that public 506C, public fundraising thing. And so I did a bunch of webinars. And without doing a single in-person meeting, $51 million in requests came in just to a type form, basically a form online. And now we're going to be starting in the next month after I get back from Japan, actually meeting with the big LPs in the world. And I want to make a trip to the Middle East and just go all around the world and meet all the big funds. So thanks for asking. Yeah, I think it's gonna change everything. Good for you. Good for you.

I mean, can you imagine $52 million in commitments before actually doing the actual tour? That's awesome. Just out of the gate. And my last one was $44. And so I think this 506C, I can be public about the fact that we're raising a fund. And so it's just absolutely amazing.

Yeah, I think it's gonna change everything. Yeah. Good for you.

I need $4 million.

Well, congrats. And I have one question for you. Yes, go ahead. Can you be replaced with an AI?

The world's greatest moderator. I mean, it's not gonna make great jokes, not for now. And, oh, you know what, I had an interesting point about management fees and these funds, just to circle back. Did you know, this is what I heard, that benchmark during that worst vintage, you know, after, I think the great financial crisis, or maybe it was the dot com, it was either of those, they took their management fees, because that fund was so, you know, challenged, they deployed the management fees into primary investing, or I'm sorry, to follow on investing on their winners to regain the results. Can you imagine in this market, a VC who deploy capital in 2020 2021, saying, you know what, we've got these management fees, millions of dollars in the future, to pay for managing these instead of taking that money, that's not going to make I'm going to put that into your into the companies for my launch fund through each month. I had a couple of opportunities. And I was like, you know what, I'm going to take some of the management fees and invest in some of those existing companies to try to goose the returns for my LPs. And so we're at 104%, or 103% invested in the capital, just by just taking a couple 100 grand off of the management fees. And I'm like, well, this is a really interesting track of view. Like, why am I playing

for the management fees? Or am I playing for the Moic? I'm paying for the Moic, right? I mean,

you should be. Jason, by the way, it's not true that the AI can't tell jokes. Our friend, Oh, No. Will Lee tweeted how the AI told a joke in this, the style of Jerry Seinfeld, then he asked it's a tele jokin. This style of Dave Chappelle and it refused so the AI can tell

a joke if it wants to. It's racist. But no, only clean jokes. Oh, I see it. It doesn't work blue. I guess every, I don't think Dave Chappelle has to be blue. Yeah. I don't, but it would be

clean jokes, it doesn't work blue, I guess, I think, I don't think Dave Chappelle has to be blue. Yes. But it would not

tell him about Dave Chappelle, but it would not tell a joke

about Dave Chappelle. Wow. I mean, we gotta get Sam. He's an iconoclastic. Like he would be Sam would be in the are all in 52.

Well, by the way, actually, he's got a shot there. Well, by the way, actually, after our last episode, in which we were raising concerns about the AI bias they published a blog post saying that's the day after. if bias has occurred, it is a bug not a feature and they are trying to be even handed. So I'm glad they have that's more now that and that's their standard. And we're gonna hold them to that standard. But I'm glad well, they have to see be public about it. Like this

yeah, well, that should be public about like this. Yeah, yeah, I mean, I read the blog post, it seemed reasonable. It's great. They're addressing it. And I also think they're now doing embedded citations. So somebody tweeted at me after we had the whole discussion about credit. And when they were doing facts, they're now saying, and they haven't been announcement about this yet. But they were saying, according to this source, the following according to this source, so they're starting to source in the copy that's being written. So that's a big step. And then I was talking to Adam D'Angelo about Poe, which is an amazing app, you should try it. I think it's the best one out there right now of all the chats. Poe is an app based on the Cora data set.

And I asked a questions about the trip to Japan and a SECO and this and that. And it was extraordinary how well done the answer was with bullets. And then I asked him online, hey, what about citations back to the original Cora questions? And he said, yes, we're going to be adding that so and then I was thinking, wow, if you add to the Cora corpus, and then they link back to your answer. That's awesome for me as a person who's answered hundreds of questions on Cora to build my reputation. So I think Cora is, for me, I I think Quora could be the Google. I think Quora's got a better data set. And if they play that right, I think they could be better than chat GPT.

And they said you have to get permission. You're saying Po is based on the Quora data set? Data set. Po will answer questions like the best answers on Quora. Is that what you're saying? Yeah. That's kind of interesting.

It basically is using Quora as the primary data set. I'm sure it's using the rest of the web too and Wikipedia and everything. I think, I don't know why they're calling it Po. I think they should just do Quora chat bot or whatever. But just try it. It's called Po. Download it.

You can use it today. Yeah.

You want to know why I'm excited about that? Because you got a little tasty poo.

You got a little slice. Yeah.

I got a little slice of Quora. Oh, good for you. Well, I mean, Quora was always like, are they ever going to make money or are they just going to build this incredible data set and do nothing with it?

Yeah. What did I say? I said AI is going to be to basically SAS what mobile was to web 1.0.

You'll either get disrupted or you'll get turbocharged by it. It's going to be. I think Quora is the number one player in AI going forward. I know that sounds crazy, but the fact that and I think Reddit also has this same potential, if Reddit had a chat spot because think about how many times people do a search and YouTube is the other one. Where they say, what's the best sci fi movie of the year or which directors make the best screenplays or whatever. And then they put the word Reddit at the end or they put the word Quora at the end or they put the word YouTube at the end to just narrow down the Corpus of where to find the answer.

Go ahead Jema. I've worked with Jema. D'Angelo for 17 years now. Smart count. He was the CTO of Facebook when I worked there. The single smartest and best, single smartest person I worked with. And then separately, one of the most absolute,

genuinely best human beings in the world. Can we get him out? He doesn't, is he not a good public speaker or something? Because I never hear him talk. I'd like to get him out all in Summit.

D'Angelo is just so superb on every dimension. We should get him on actually, because I didn't know he was working in AI. He has a lot of interesting thoughts about,

you know, social networking platforms and- And he's on the board of OpenAI.

Oh, okay. That's interesting. Oh, really?

Oh, okay. That's interesting. Oh, really? Oh, get him on the pod, or maybe you own Summit, 2023.

All right, everybody.

He'll definitely make the anti-establishment list. Definitely anti-establishment, yeah. Okay, so for the sultan of sneaking out, he left and the dictator. And what do you want to be referred to now? Pacifist? The peacemaker. Pacifist? The peacemaker. Yeah, you are. The saxifist. I'm the world's undisputed greatest moderator

on the number one podcast in the world.

The peacemaker. The pacifist.

For now, until the AI replaces you. Yeah, I trained the AI to replace your sax. Ukraine, UK, New Ukraine, Biden, Biden, Biden. No, Nikki Hallie. No, stop making Nikki Hallie happen. The end. The dataset has been done. All right, everybody. See you next time. Love you, boys. Love you, Dusty. Bye-bye.

And David Sacks. They've just gone crazy with it. Love you, Bess. What? You're a bee. Bee? What?

What are you doing?