SPEAKER_00: I'm too old for this. And I'm too like cynical and like set in my ways. I just believe in

SPEAKER_01: capitalism. I don't believe in unions. I think that like there's union leaders who fight for folks to get the same amount of pay. I believe in meritocracy. Everybody should fight for their highest salary possible. But if you set this up like this, I'm going to retire.

SPEAKER_02: I think the anarcho-syndicalists would probably want guys like you to retire.

SPEAKER_03: They would. They would want to retire me, literally like a replicant.

SPEAKER_02: They're aiming to get rid of the- Capitalists. Yeah, the upper echelon of capitalists and instill workers as the leadership rather than

SPEAKER_08: owners. And they'll be called, what would you call a worker who becomes a leader? What was that? It's like a bee or something? Like a boss.

SPEAKER_11: This Week in Startups is brought to you by Coda. Coda empowers your startup by bringing words, tables, and teams together. Strategize, plan, and track goals effectively with all your valuable data in one place. Go to coda.io slash twist to get started for free and get six free months of the team plan. LinkedIn Jobs. A business is only as strong as its people, and every hire matters. Go to linkedin.com slash twist to post your first job for free. Terms and conditions apply. And Northwest Registered Agent. Starting your business should be simple. With Northwest Registered Agent, you can form your entire business identity in just 10 clicks and 10 minutes. From LLCs to trademarks, domains to custom websites, they've got you covered. Get more privacy, more options, and more done. Visit northwestregisteredagent.com slash twist today.

SPEAKER_01: Hey, everybody. Welcome back. This Week in Startups, 2,000 plus episodes. ThisWeekinStartups.com, TWIStartups.x. And you can find us on Instagram, Twitter. With me again, Lon Harris and Alex

SPEAKER_08: Wilhelm. We're here three days a week, Monday, Wednesday, Friday, right about noon, Texas time,

SPEAKER_01: 1 p.m. East Coast, 10 a.m. Left Coast. We go through the top big tech, little tech,

Chamath Palihapitiya: venture capital, and capital markets. And we'll take a detour now and again into a little politics, a little media. And of course, Founder Lessons is where we score our points. We put points on the board, unlike the Knicks in game three. But we'll make up for game four tonight. I was supposed to go

SPEAKER_00: to New York for game four, but I'm doing something at Austin University tomorrow night with Dave Rubens. Yes. Let's go Knicks tonight. I didn't go to game four, and I'm a little bummed about it, if I'm being honest. But it looks like the series might go five, six, seven, eight games, and seven is the max. So I might go to a later game. We got a big docket, because what we do,

SPEAKER_01: folks, is you can go to ThisWeekinStartups.com slash docket. We get all your stories. We look at how you're replying to us on X, where, you know, the intelligentsia are talking about tech. But just to kick us off, there was an interesting discussion going on, on Reddit. And this was, um, in my favorite group, which is called Anti-Work. This is a group of individuals who believe that you should be, uh, have maximum free time. There's no such thing as lazy. It's kind of a little bit communist and a little bit work-life balance. So you kind of get hardcore people who believe humans should not have jobs. Like, straight up, there should be no work. The only work you should do is, like, raising your family, hunting and foraging for food, apparently, et cetera. And they kind of resist

SPEAKER_08: modernity. And then there's also just people who are in jobs, not careers, hate work, and just want to vent about work. But this was an important one. Uh, so let's, let's cue this one up, Lon.

SPEAKER_18: Yeah. So, uh, it's a user called, uh, Crimson, like C3 Crimson. I'm calling them Crimson. I think

SPEAKER_02: notably here, they've tagged themselves, you know, there's Flare in Reddit. They've tagged themselves as an anarcho-syndicalist. So they, in the shortest description I can give, they're basically an anarchist who thinks that trade unions should be the vehicle that liberates

SPEAKER_21: American workers. So this would be, you know, just like you were describing a person who believes that, uh, we need to liberate people from work and that their, their labor is being abused,

SPEAKER_22: exploited, and stolen in general. Got it. And what is their, what is that category called again?

SPEAKER_02: Uh, they're an anarcho-syndicalist. That would be anarcho-syndicalisms. So it's like,

SPEAKER_23: you're a narco-anarchist. Anarchist.

SPEAKER_04: Syndoclism being... Being like, you, you believe in trade unions and that that's how workers should

SPEAKER_27: be uniting. So, Lon, just very limited government, and then worker-based unions as the quantum of

SPEAKER_21: government, essentially? Correct. Yes. Okay. That that's, that would be the primary political organizing tool, not a like overarching federal government. They're anarchists. They don't want that at all. Workers unite based on trade unions.

Chamath Palihapitiya: Yeah. Okay. So I thought this is interesting. I will say when I had a, a little micro flare-up

SPEAKER_00: of unionism at one point in one startup, I had a meeting with the leadership. And I said, I'm too old for this. And I'm too like cynical and like set in my ways. I just believe in capitalism.

SPEAKER_01: I don't believe in unions. I think that like there's union leaders who fight for folks to get the same amount of pay. I believe in meritocracy. Everybody should fight for their highest salary possible. But if you set this up like this, I'm going to retire.

SPEAKER_02: Well, I, I think the anarcho-syndicalists would probably want guys like you to retire.

SPEAKER_37: Yeah. I don't think you're- They would want to retire me.

SPEAKER_02: Literally like a replicant. Yes. They're, they're aiming to get rid of the- Capitalists. The, yeah, the upper echelon of capitalists and, and instill workers as the

SPEAKER_08: leadership rather than- Got it. And they'll be called, what would you call a worker who becomes a leader? What was that? It's like a fee or something? Like a boss. Well-

SPEAKER_02: The boss. Uh, it's a world without, it's a world without bosses, ideally. Um, wow. Okay. This particular post was not about anarcho-syndicalism. They're a remote employee or, and supervisor for a human services agency in Pennsylvania. They leave their work cell turned off after work and on weekends to preserve work-life balance. An admin recently chewed them out via email and said that their work phone must be turned on and with them at all times. Okay. Now the OP is saying this makes them feel that they are expected to be on call 24-7.

SPEAKER_21: They don't feel like they're being paid enough to meet that expectation. And they don't feel like that was initially sold to them as this is a 24-7 on call all the time job. They're saying this is a massive deal breaker to them. And so the original post was seeking for advice. How do I go back and say that I'm not willing to be 24-7? They did follow up by saying, uh, the management has now confirmed they are expected to be 24-7 on call, but that it can be a rotation. So not, not everybody has to be on call at all times. They could make a schedule where some of them are on call some days and some

Chamath Palihapitiya: of them have other days off. It's an important discussion, especially for entrepreneurs and setting expectations. The reason this person was a bit upset is because their expectations changed. And so I'm going to give you some language to use as a founder when you have to change expectations with the, with your team. And that would go something like, Alex, I know that when we hired you, we did not specifically address this, but starting in Q3, in seven weeks, we will be moving to an expectation that you're available to, you know, our caseworkers off hours. What this means is

SPEAKER_01: you need to have your work phone on. And if somebody calls, we expect you to respond within an hour or two, there can be mistakes that are made. And if you are absolutely not available because you're on a rafting trip and there's no cell phone service, we'll either get you a satellite phone or, uh, we will make sure somebody is covering your phone number. And so that's the

SPEAKER_08: communication part. So this would have been, if level set previously, less of an issue or a non-issue,

SPEAKER_45: but it really does depend, I think on your compensation level. And it depends on your line of work because in the line of work we're in, we do have founders who will text us and email us

SPEAKER_46: at three in the morning, having something between a panic attack, a manic episode, inspiration, or a chaotic moment in their life, right? Founders are a unique in group of individual. They own

SPEAKER_45: typically, you know, 40% of their company along with a co-founder, or if they're a solo founder,

SPEAKER_08: maybe 60 or 70%. So they're well compensated to have this level of stress, which is they do not, like CEOs, Tim Cook, something happens on tariffs, literally he's being woken up. Somebody, you know, his security team gets a phone call, like literally he's got 24-hour security. Somebody, somebody, Eddie Q calls, says, wake him up. Trump's in whatever. Here's the latest

SPEAKER_45: decision. Literally a security person comes and knocks on his horse and wakes him up, just like the president. Compensation equals that possibility. And so if you're making, and I'm going to pick a number, in the United States, the average salary is somewhere in the 50, 60 range. And if you took the average salary and you said you're below the average salary in the United States, you would be an hourly employee, most likely not management. In most places, this isn't legal advice, but you can look it up. You would be entitled to overtime. And then this falls into a gray category. Is it overtime? Because you're not working? So no. So what happens is you have to make the deal. Yes. And the deal to be made here is very simple. We're going to create an, if these people are paid under $60,000, $70,000, you say, hey, we're going to have an on-call schedule. There are 10 of you. Each person is going to be primary or secondary on two weekends. We're going to rotate it every five weekends. Primary gets the phone call. Then the phone, if you're using like open phone, we'll ring the secondary. If it's a second or third phone call, whatever, and the person's

SPEAKER_08: not available. As compensation for this, we're going to give each person who's primary four hours

SPEAKER_00: of paid time off, whether they get called or not. And if you do four hours, you're compensated. And if not, hey, you got three, four hours of paid time off. So if this happens 10 times a year or 20 times a

Chamath Palihapitiya: year that you have this weekend and you're on call, you could earn up to five to 10 extra vacation days, one or two weeks. This is the way to do it properly. Founders, I know you're swamped. You got

SPEAKER_52: a ton of stuff to do on your plate. You know, there's a lot of chores. You got to juggle your priorities. And that's why I want to tell you about Coda. That's what I use to keep everything under control. It's an all-in-one platform and it's going to consolidate documents, spreadsheets, and apps you have running into a single scalable workspace with an amazing UI. So Coda is going to help you set all your OKRs. You know, that's objectives and key results. And you're going to get access to hundreds of templates for tasks like setting priorities and mapping dependencies. At This Week in Startups, we use Coda every single day for important projects like the Twist 500 and Founder University. Why do we use it? Well, because we have all these portfolio companies who have tons of information and they submit updates. And the beauty of Coda is it's so flexible. We can use it for everything from tracking those companies to tracking their pitches to board updates. And it all can live in one place. And we don't get software overload. We don't have 20 different SaaS products that everybody needs to learn. They need to learn how to use one. So instead of juggling all these different tools, you've got one platform and the whole team is super organized. It's really intuitive. The workflows are seamless and getting more powerful by the day. So here's your call to action. Coda empowers your startup to strategize, plan and track goals effectively. Take advantage of this limited time offer just for startups. Go to coda.io slash twist today and get six months free of the team plan. That's coda.io slash twist to get started for free and get six months of the team plan. And then you could even say to

Chamath Palihapitiya: people, here's the schedule, pick which weekends you want to be on call. And there might be some people say, this is the greatest deal ever. I want two extra weeks vacation. Can I get as much

SPEAKER_01: on call time as possible? So you kind of flip it into an opportunity. Now, some other people might be like, you know what? I like to turn it off Friday, five o'clock. I got kids. I got meditation, whatever. I read a book. I don't know. Tim Ferriss told me something. Andrew Huberman, whatever. I got a cold plunge. Therefore, I need these 48 hours for my mental health, my preference and need to that. So these are the ways to handle it. But I will say, if you view your life as a career, you're a real

SPEAKER_08: estate broker or you are a venture capitalist, when you get that phone call, and I'll end on this

SPEAKER_45: point on the weekend, that is a timing moment of your career. And it is a competitive advantage to be available and to take it seriously. I get probably five of these a week, 10. They take typically three

SPEAKER_08: to seven minutes to dispatch. They typically come in the form of a text or an email or a Slack message,

SPEAKER_45: not a phone call. If it's a phone call, something's really screwed up. I'd say one out of 20 is a phone call. 19 out of 20, somebody emails me, term sheet fell through, CTO is quitting, we got a lawsuit.

SPEAKER_08: And I can say, great, we'll process it. I set up a time for us to talk and looking at it, here's my initial reaction. Boom. My job. That's the job of any venture capitalist. If you're a real estate broker and your client calls you and says, hey, I saw this home, I really want to see it. And they text you on a Saturday afternoon and you're with your kids at their soccer game, you look at your phone, you go, if I respond first, maybe I get this business. If I respond on Monday morning, 9 a.m., maybe I don't get this business in your best interest to do it. So I wrote that. Of course,

SPEAKER_44: somebody said, so your position is you got to stay 365, 20, whatever, be on call all the time. No, there's like, realistically, how often does this happen is the question. Now, if it was a call center and you're doing like, I don't know, sure, a microphone's call center and you're getting a call every hour, well, then you should be staffing it with full-time people or some other process. So I'll stop there and get your reaction, Alex, or questions.

SPEAKER_66: Yeah, just a couple of things. So first of all, I looked it up. Human services workers in

SPEAKER_67: Pennsylvania make a median salary of about 40. So this person could be making 50, say. So we're not talking about an incredibly well-compensated position to start. Nope. I think, Jason, you nailed it with the expectation setting, schedule splitting, and letting people have a little bit of flexibility. My spouse and her career has to do on-call weekends, like for example, the weekend we just had. And she gets paid extra for that. It's not

SPEAKER_66: voluntary, it's mandatory, but they definitely negotiate it in advance and get all that sorted

SPEAKER_14: out. Just a small fee, but it's not for every hour of the weekend. It's a fee that's negotiated for

SPEAKER_66: the weekend. It's a per child scene fee. It's the medical world, Jason, is a weird place.

SPEAKER_72: Okay. So if you have an engagement, you get paid for that engagement. You have to be around and

SPEAKER_67: okay, great. Perfect. She went to the hospital both days this weekend and I got to do solo childcare, which by the way, didn't get paid for world. Come on. Anyways, I think you raised a really good expectation setting. You're happy mother's day. The last thing I'll say is I read a lot of the developer forums out there and a lot of developers will occasionally work at a company that has a on-call system, like, you know, tracking the pager in case the servers go down. PagerDuty is a company

SPEAKER_76: that supports that. Indeed. You might even wonder where they got the name from because it's their

SPEAKER_67: duty to have the pager. I'll just say this. Those are also divided up based on who's on call to make sure it doesn't burn people out. And we're talking about higher paid careers in those two cases. So if I was earning 50, I would expect at least the same courtesy given my lower comp points, Jason. So I'm kind of on your side here, but I also think that this company that Crimson works for, Crimson, didn't handle it well and just kind of like poor internal management. You can't just tell you people suddenly, oh, you can never turn this phone off and then just drop the ball there.

SPEAKER_78: Absolutely. Yeah. Poor management. And Lon, comrade Lon, I'll go to comrade Lon.

SPEAKER_02: I think this is a real, it really comes down to incentives. If this person felt incentivized,

SPEAKER_21: for some reason, it could be money. It could be the opportunity to move up. It could be just love of the game. Like they're working on a project they really care about. If they felt incentivized, it wouldn't, I don't think be that big of an issue to keep their phone on them and to check in. Like I check in with you on the weekends. We were talking about that graphic design project and

SPEAKER_02: like some other stuff we want to do for social media. It happens every weekend because

SPEAKER_21: I'm compensated enough to work at this job. I'm motivated. I love the show. I want to make a better show. It's not an misaligned incentives when we powwow on a Sunday evening about the show. But if I was being paid a lot less, or if I felt like the expectations were out of control in other ways, I would feel less incentivized. And I would be like, oh, I got to check in on Sunday. It's a chore. So I think to me, that's what this feels like. And the employer needs to do a better job of however you incentivize it, motivating this employee to want to go the extra step and check in on a

SPEAKER_00: Saturday. And the chewing out was like an interesting part of it. So you didn't set the expectation, then you chewed out. It's like, that's kind of like, you know, you're setting up complete failure in the relationship there. Now, if you did set it up, and this was the third time the person like blew off the client, you could, as a manager, listen, there's two sides to every story. I don't

Chamath Palihapitiya: know what chewing out means. The person could have been incredibly appropriate and delightful, but it

SPEAKER_08: could have been felt as chewing out. Because the level setting wasn't there. If they said to them, listen, John, this is the third time you didn't pick up a client call. The last two times, we lost the client. They fired us. And then this time, we don't know. That's going to cost the company $100,000. And if this happens again, we're going to have to let you go. And you're going to have to really give some consideration to if you want to hear or not, because we can't afford to lose clients because the company is, you know, we lost money last quarter, and we're trying to get back in the black right now. So there's a way to handle this that isn't like chewing out. And some people like candid talk equals chewing out. If you have people, you know, in startup land, I'm going to bring it back to startups, who consider radical candor, shout out to Kim, who wrote the book at Google. They consider radical candor chewing out, which like a government employee, which is, I think, a government position. Like, they're already in a toxic relationship, I'll be honest, a unionized relationship. It becomes very toxic very quick. Management is against labor. Labor is against

SPEAKER_45: management. Everybody is like, you know, being petty. You really can't have that radical candor. And so you need to set a culture and say, hey, we're going to just speak candidly at all times. And

SPEAKER_08: you can speak candidly to me. You can say, as a boss, you should have told me about this. You know, I wouldn't have taken the job if you did. You should be able to have that, you know, conversation. You can say, you know, I think I maybe I need to look for another job if you expect me to be on call every weekend, because it's a little too stressful for me. I can't enjoy my life. So maybe it's time we part ways. It's a two-way street, that candidness. Okay, let's get back to

SPEAKER_27: the talk. Let's go ahead and talk about graduating from seed to series A. This is the thing, Jason,

SPEAKER_67: I've been tracking since my time at Crunchbase News. I call it just graduation rates. If you raise one round as a startup, how often and how long does it take you to reach the next? And one thing we talk a lot about here on the show, seed state startups, of course, Jason, because you're an investor in a bunch of them. And we care a lot about the pass-through rate to series A. And well, the numbers don't look particularly good. I have some data here from our friends over at Carta. Shout out to Peter Walker from their data team. And if you take a look at this chart, if you're watching the video, I'll quickly explain. Years are on the left and time passes as we go to the right. And you can see how many companies from a particular cohort done by quarter and year reached a series A round after raising their seed. As time goes along, Jason, you'll note that more companies raise their series A, unsurprisingly. Data goes up and down over time. The thing that stands out to me is a deterioration in the rate at which companies are graduating from seed to series A in the last couple of years. And the most shocking data point Carta actually highlighted, which is that if you look at Q1 2023 seed rounds, those that are now two years old, only 15.5% of companies in their database that raised a seed two years ago have raised an A, which is, Jason, far below the highest watermark ever set, which was 2002 Q3, when 40% of seed stage companies have raised an A in eight quarters. I was blown away by this. I'm curious, one, your reaction. And two, what was it? Yeah, 2020 when that happened,

SPEAKER_89: which was peak Zerp. All right. We all know if you're a founder or even if you're on a small business, you're thinking about your company 24 seven, 365 days a year. That's the life of a

SPEAKER_52: founder. This is not clock in, clock out, nine to five gig for you as the business owner. So when you're hiring, you want a partner that's as equally as committed as you are. And that's of course, LinkedIn jobs. LinkedIn jobs is like your co-founder. They're going to make it so simple for you to post your jobs for free on LinkedIn, where there are 1 billion members. You're going to be able to share what you're posting and actually keep all the promising candidates organized in one place. And also LinkedIn is going to help you quickly write a job and get it in front of the right people, whether you want to post for free or use some promotion to get it in front of even more qualified applicants. So do me a favor. Don't take my word for it. I mean, you should. I know what I'm talking about is where I find my great people. But just understand that 72% of small businesses using LinkedIn said that it helped them find the best candidates. So find out why more than 2.5 million small businesses already use LinkedIn for hiring. So here's your call to action. Post your job for free. Why wouldn't you do it? It's free. F-R-E-E. That's a good price. LinkedIn.com

SPEAKER_91: slash TWIST. Once again, that's LinkedIn.com slash TWIST to post your job for free. Terms and

SPEAKER_00: conditions do apply. So when you look at this chart, the red is they haven't, right? A very small number have, and then you go to pink and you go to this, you know, orange. Yeah, you can see that in 2002, it was at its worst. So that looks like it's eight, nine, 10, 11 quarters, right? For 2022 companies pulling through. Pull through is another way for graduation rate. Now we don't expect it to

SPEAKER_08: be more than 20, 30%, right? So if you invest in four, you put $100,000 in four different companies, and one of the four in the seed round makes it to series A, you will have paid $400,000 essentially

SPEAKER_45: for your equity in that company. That's a nice way to think about it. If you're pre-seed, like we are, this is even less, this would be half of the amount, right? So very few pull through. But if you're

SPEAKER_08: pre-seed, like you're running an accelerator, you are investing at but a $1.7 million valuation as Y

SPEAKER_45: Combinator, Techstars, or ourselves, the Launch Accelerator, or in our Founder University program, even $1 million. So we will expect 40% of our companies, we track this statistic, we expect 40, or we've seen 40 to 45% of people coming out of our programs on average make it to another round of funding that is at a higher valuation, and I think we say over $100,000 in total investment. So

SPEAKER_08: if somebody puts a $10,000 check in and doubles the valuation, maybe we discount that. So it might be $25,000,

SPEAKER_45: $50,000 or $100,000, I think we said as like a minimum benchmark for what we call pull through for graduation, not to A, but to at least a round that's a higher round because people do

SPEAKER_08: multiple notes, et cetera. The carded data is but one subset. So caveat, caveat, caveat. Other issue here at Play is that founders are getting to profitability quicker and don't need to get an A. They could do an, and they, there's a seed complex out there where you can do bridges and notes. And it isn't as much about hitting that perfect series A. You could have companies like Comm or FitBot in our portfolio that don't do a series A for many years. They might do one or two notes or none and just do a secondary because they're so profitable. In the case of FitBot and Comm, both consumer subscription applications, they were so profitable, they didn't need to. I've also seen this with companies like Tonebase, another one of our, that's a web-based subscription for learning how to play musical instruments and specifically classical. I've seen a lot of these companies where I'll offer them additional investment and they say no to me. Like all three of those were not interested in raising additional money or selling secondary because they were growing so nicely. So there's a little bit of that in there, what I call the alicorns. Unicorns, unicorns, or in my mind unicorns, flying over rounds of funding because they got Pegasus wings. Really nice trend. What this means also is the number of startups could have increased. So if you have a larger number of startups, but the same number of series A investors, it's hard to kind of get that series A. So you would do something creative like another seed round, another convertible note, and you just push it out, push it out and try to reach profitability. If we had more M&A, which over the weekend, I'm not going to say which company, we were informed one of our portfolio companies was getting bought. This is a great transaction, apparently. We'll see the devils in the details, but another great transaction for us if it goes through and it looks like it's going to. So the wrath of Lena Kahn ending and

SPEAKER_00: the Trump M&A bonanza begins, which, you know, I would just say is a Republican M&A bonanza and

SPEAKER_45: hopefully less regulation on M&A. We need to let M&A rip or else we will not have risk capital at work. Let it rip, President J.D. Vance. I mean, Trump.

SPEAKER_67: I don't think either one of those is in charge. Now, Jason, I want to ask you about series A benchmarks because when I was learning venture capital, it's pretty simple. You wanted to have a million in ARR and then you could raise a series A. Clearly, those times are behind us. I was curious if rising series A expectations are impacting seed pull through or graduation rates.

SPEAKER_103: Of course. Yeah. It's two or three million is going to be the number of revenue, amount of revenue.

SPEAKER_67: Yes. Yes, exactly. So I went around the internet and I found some stuff about this. So Valor VZ says two million plus ARR. Founder Institute was about 2.4 million ARR. And the thing that I'm trying to sort out is how fast you have to be growing at, let's say, two or three million ARR to raise a series A. Because I saw numbers from 200, 300%. I'm curious, like, what do you see out there

SPEAKER_76: amongst your winning companies that are making it to series A? Ideally doubling. Yeah, ideally doubling.

SPEAKER_01: So 100%. Yeah. If you're going from a million to two million, then a VC can look at it and say, wow, it's not an accident in all likelihood. Then look at the quality of the revenue. We've talked about that many times here. Are a lot of people churning? Or do they have an incredible sales team that can sell sand to somebody in the desert? And now that person's like, yeah, I don't need any more sand. You got me once, but I'm not going to renew. So you got to look at the quality of the revenue. But people are looking at the early stages on small numbers for triples. Ideally,

SPEAKER_08: a double would be fine. And what that means is they probably have a theory. We're going to give you five or 10 million. You're going to add these two or three critical executives. You're going to invest in these one, two or three areas. And then we can increase that slope, which is what venture capital is for, is to increase the growth rate. Startups are meant to grow. And they're meant to grow like rockets, not helicopters or airplanes. If you've got an airplane and it's flying and it's gaining altitude, Lon, that's fantastic. Great. But we all know it's going to get 40,000 feet. And then

SPEAKER_00: that's the tolerance of that vehicle. In venture capital, the tolerance is getting to orbit. You want to be able to get to space and you need that critical velocity to do it. The critical velocity is jet fuel. And that jet fuel, it's not for airplanes, it's for rocket ships.

SPEAKER_67: So the question then becomes, Jason, if we now understand the Series A benchmarks, and we say that the seed companies are struggling to graduate for a variety of reasons, are they simply just not meeting the Series A benchmarks? And is that why they're stuck?

SPEAKER_00: No, a lot of them are meeting it. I think there's just not enough VCs to go around. And the VCs only have 10 dance cards. So if you have a lot of these funds getting really big,

SPEAKER_45: they have to put bigger numbers to work. And then there's other people to come in. And fill that gap, specifically SPVs. So we talked about the company that makes human robots over and

SPEAKER_08: over again here. I don't mean- Figure. Figure, yeah. And listen, I wish the founder tremendous luck. And I was on his side with the BMW issue. So this isn't like a Palmer Luckey situation or Zuckerberg situation. Don't start clipping me and trying to get me in trouble here. But they're raising money according to what I've read in SPVs at $38, $39, $40 billion pre-revenue or pre-any significant revenue.

SPEAKER_45: Those would be the red flags that people would see in a market where optimism trumps reality.

SPEAKER_00: And that's fine. Investors are willing to create reality for this massive optimism. Founders are allowed to capitalize on that. But man, that can create a lot of distance and a big bubble

Chamath Palihapitiya: between things on the opposite. And so that means like a VC firm doesn't do that round. And the VC

SPEAKER_01: firm would be looking at it saying, well, what's the IPO going to be? Or who's going to buy this for how much? And what's their theory going to be? They're going to look at those two exit potentials, and then they'll work backwards. And they might give you credit for year three or four's revenue,

SPEAKER_44: but they might not give you credit for year 30's revenue because there's no room for a return for

SPEAKER_52: their LPs. Founders, if you're serious about raising money, you need to set up your business the right way. Tight is right. And it all starts with having a registered agent. Investors simply won't fund your business if it isn't structured correctly. Before a VC can wire you the first dollar, they're going to check. Is your company incorporated? Is it in good standing and compliant? Missing a filing or losing your status? I mean, it's just going to be a deal breaker for the VC. It's like you're not taking things seriously. And that's what happens during due diligence. That's when a VC makes sure they're not making a mistake by giving you investment dollars. And hey, angels do this as well. And that's where Northwest Registered Agent comes in. For just $39 plus state fees, they act as your registered agent. They handle all the paperwork, they keep you compliant, and they make sure investors see you as a serious business that's worth funding. In just 10 clicks and 10 minutes, your business is officially set up and investor ready. Northwest handles filings, they protect your privacy, and they ensure that you never miss a deadline. These are the chores that you don't want to have to deal with. You need a partner, and Northwest Registered Agent is that partner. Thousands of founders trust Northwest because they keep businesses in good standing. And with their expert corporate guides, you get real support, you don't get bots, you get real people on the phone. So here's your call to action. Very simple, easy peasy, lemon squeezy. Don't let bad paperwork cost you your next funding. Go to NorthwestRegisteredAgent.com slash twist and get your business investor ready today. For just $39 plus state fees, you can set up your company the right way, fast, private, and compliant. Go to NorthwestRegisteredAgent.com slash twist today. That's the thing that's

SPEAKER_08: troubling to me in this whole space right now. And that some entrepreneurs, you know, and they're

SPEAKER_118: allowed to do this, if there's willing buyers of securities at a fair market price.

SPEAKER_120: We are selling to willing buyers of the current fair market price.

SPEAKER_08: Is the figure fair market price $40 billion? Who's to say? The buyers are willing to pay it, therefore that's the fair market price. Literally, that's the definition of a fair market price. But if the VCs and what we'll call the Sharps, the people who are really good at placing bets, are opting out at that level, that tells you one of two things. The VCs are wrong, or the other market participants are wrong. Who do we think is wrong in this equation? Sequoia, Andreessen Horowitz, or a bunch of people who are high net worth individuals putting money into an SPV with a 5% or 6% load-in fee? Yeah. People have been doing it for 50 years? People who backed Apple, Google, et cetera? Or the people who, you know, YOLO'd a bunch of crypto and

SPEAKER_45: decided, hey, let's see if I can YOLO one more time. Sometimes the YOLO's are correct,

SPEAKER_08: as we've seen. Other times they're not. Overall, I feel like the industry is in the process of coming out of a brutal four-year struggle with COVID. Literally, if you look at what the industry venture has gone through, it's kind of like having some kind of COVID, and now you're coming out of it. M&A is happening again. Funding's happening again. We're seeing weird things. Weird things always happen. So when you see weird things, that's actually a sign of a healthy environment. Some people get a little too frisky. Some people are too conservative. That's actually, as long as transactions are occurring, that's good. But my Lord, I just hope we're not in the

SPEAKER_00: long COVID situation, where venture has long COVID. It's just constantly not able to hit the strides it hit previously. We need to get more numbers on the board, more IPOs, more M&A.

SPEAKER_02: Let's talk about copyright. So we have confirmed news that the president has fired Shira Perlmutter, the head of the U.S. Copyright Office. Her office put out a report recently casting doubt on the rights of AI companies to use copyright-protected information to train their models. So the report didn't say that what a lot of people who work in AI or invest in AI, they basically want the government to give carte blanche, there's no copyright protection for training your models. It's fair use. And the Copyright Office did not quite go that far. So they said that this report potentially harms the argument that tech companies can just hoover up all the published information in the world and use it for whatever purposes they like. The wording is specifically copying expressive works from pirate sources in order to generate unrestricted content that competes in the marketplace. That would mean Meta and other companies that use these big pirated archives like LibGen, or we talked about The Pile on another episode, these massive troves of already stolen material. They've been sucking that up and saying, well, we didn't go violate copyright. We just used this third-party collection of copyright of restricted material. This report is basically saying that's not okay. And it's out of step with

SPEAKER_31: copyright law. So that could be a big obstacle to training these models moving forward.

SPEAKER_130: I just wanted to say this report, pretty good. It's long. I didn't get to read every word this

SPEAKER_110: morning. So I think it was like a hundred pages or something. But if you're curious about how AI

SPEAKER_67: works, generative AI works, copyright history, different arguments, it's pretty even-handed. So I would really recommend everyone take a look at it. There's a link in the docket. We'll also put a link in the show notes that go up on YouTube and so forth. But definitely just take a peek. I had to really narrow down what I wanted to pull out for today, Jason, which is the quote we just went through. But I'm curious how much heat you think this puts Meta under it? Because famously, they were

SPEAKER_133: known to use LibGen for their AI model training. This seems to me to be an inside job. I think there

SPEAKER_00: are a lot of people, you know, the way our government system works here in the United States, for those of you outside of it, is you give donations, you get access, and you back. Money plays a big role in politics here in the US. You back your candidate, whether you're giving a hundred bucks to Bernie Sanders or you're giving 10 million to a super PAC for Biden or Trump or

SPEAKER_08: whoever, Clinton. So money buys access, and entrepreneurs and corporate interests have more

SPEAKER_45: money than individuals. This is the nature of our system. So it is part of political speech

SPEAKER_08: that, you know, well, depocketed sources can do this. Who's depocketed? Obviously, Meta's depocketed. Obviously, Google's depocketed individual venture capitalists, etc. And the law is the only

SPEAKER_00: backstop to this. So what needs to happen here is because special interests are fighting to break

SPEAKER_08: copyright, which I encourage them not to do because it will have so many second and third order impacts, that you cannot predict, like people will stop making content that would make your thing. It's much better for these companies to pay a licensing fee. If the opportunity is as great as it appears to be, then you should do what some firms have been doing. Microsoft and ChatGPT, OpenAI have licensed

SPEAKER_00: content from people and paid a fee to do so. You two gentlemen can go do your race right now in a in a GDP, an LLM to find the deals they've done. But we remember, you know, like some magazines had

SPEAKER_08: actually done licensing deals for them. Those deals are going to need to be done at scale. And those deals present an opportunity, the opportunity being when you open up your LLM to say, we have the rights to Reddit, and other people don't. So you could, you know, if you're a Grok, come out and say, hey, we're Grok, pay 30 bucks a month, and you get Reddit, Twitter, Disney, and Shimon & Schuster, and the Wall Street Journal. And then ChatGPT or Gemini could come out and say, hey, and you can use our services and you get X, Y, and Z. This could be a competitive advantage for each. And then you could explicitly say, tell me what Reddit users are doing. Tell me what Disney films, you know, make me a short film with every time Boba Fett has appeared in a Disney film. Those cool features tell me every New York Times article that mentions Bob Dylan and make me a timeline and give me the songs mentioned in it and make a playlist out of that with Spotify. Like these are really interesting queries I just came up with off the top of my head. Those should be used in the branding of LLMs in the

SPEAKER_00: future. And they will. And that is the brilliant way for this to emerge in the United States. And then you could buy Reddit. You could buy Quora. And you could have that forever,

SPEAKER_08: just like Elon bought Twitter. And it's now part of XAI. Nobody else can use Twitter. He will defend

SPEAKER_45: that. Sam Altman, I think Reddit did a deal. Am I correct that Reddit did a deal with Google?

SPEAKER_27: There is a Reddit OpenAI partnership from last year. And also there is the Google Reddit search deal that Reddit discussed in its earnings post IPO. Jason, you're dead on.

SPEAKER_67: Just to answer your question though. So OpenAI partnerships include Washington Post, Axios, Future, Hearst, Time, Reddit, Axel Springer, et cetera. So quite a number of them.

SPEAKER_135: And those are non-exclusive as my understanding.

SPEAKER_27: I don't know on a per deal basis, but I do believe most of them at least are not to buy my knowledge.

SPEAKER_67: Also keep in mind that Times is suing OpenAI over this issue. But what's interesting is that the copyright office came down, I would say a little bit generously on the side of if you're taking in information for training and then using it internally, 100% clear, or if you're going to take in data to train and don't allow it to be copied, as in the prompts can't spit out the original, probably okay. That's I think pretty fair, Jason. But if the political climate is so sharply

SPEAKER_27: opposed to what I would say is a very reasonable middle ground here, what should founders know and do? And then also, do you put any truck behind the argument that if we do have strong copyright

SPEAKER_137: protections here in the US, contra AI training companies, we're going to lose to China? Or is

SPEAKER_139: that just kind of fear-mongering? Absolutely not. We will win because we will be able to put pressure

SPEAKER_00: on China and block those models for copyright violation. If a Chinese company tried to operate in the West and sell Microsoft Word, we would block and sanction that company in Italy, Africa,

SPEAKER_08: India, and the United States, South America. And by the way, Chinese companies have tried to do that. And Microsoft, as but one example, wound up being able to sell their products in

SPEAKER_46: China and block Chinese pirates from doing this kind of stuff. And this will make our industry

SPEAKER_45: stronger ultimately. So it is ridiculous to say, because the Chinese are stealing intellectual

SPEAKER_08: property, therefore we should be able to. It is ridiculous to want to protect your intellectual property in China. And then at the same time, say that other IP holders in the United States shouldn't be because we won't win the AI race. We will actually win the AI race. We will slow down to speed up. If it turns out that licensing Axios and licensing the Washington Post and giving

Chamath Palihapitiya: Washington Post $10 million a year, or five people giving them $10 million a year, they can then hire

SPEAKER_08: more journalists. They can then do more projects. Those projects then will accrue to US

SPEAKER_00: companies directly in real time. So imagine $50 million, divide $250,000 in total compensation for the most extraordinary journalists in the world that would top, top pay for like 40 year, 30,

SPEAKER_08: 40 year vets with master's degrees, et cetera. You know, now you're talking about four for every million dollars, 50 million. You're talking about 200 of the top flight journalists filing, I don't know, every, let's say 220 pieces a year. They're doing one piece every two weeks. Now you've got 4,000 and that's from one publication. Extremely rich detailed analysis and research pieces in our archives in real

SPEAKER_148: time. Smarter search results. Better way to take on Google. Not a bad way to go about it. Go ahead,

SPEAKER_149: Lonnie Donnie. And to flip it, I mean, I think if we say there's no restriction, you can feed all the

SPEAKER_02: history of published work into AI and use it to create any new thing you want. I think that creates an existential crisis for media, writers, journalists, publications. In general, we still, I think everybody, or most reasonable people would agree, you want to have both. You want to have the ability to use AI and generate AI outputs. And you want to have people still doing the original writing, researching, and reporting that some people are doing now. You don't want to create, we can't have the AI be the only writers. I think that doesn't really make sense. And so we need to do something to protect the media that we already have. And I think we're watching it shrink down to a nearly, you know, like minimum size to function. And I think that there's a real danger of that. And so creating an AI ecosystem that still provides in some way for there to be income for journalists, and even if it is being licensed by AI companies, that's virtuous. We need something like that,

SPEAKER_21: because clearly the ad market alone is not enough to fund the journalism that we need to function as

SPEAKER_02: a democratic society.

SPEAKER_00: Yeah. And created by humans is a company we've invested in that is going to work on creating clearing houses. Here, you can see on your screen, the AI rights licensing platform for books, and they're starting with books. So you can put your book in there, and then you can get paid by each LLM. They do the clearing. You can pick the number of years, how much of the book they're allowed to put in there, how much they're allowed to quote. You know, you want books to be written. If people get paid 50K on average to write a book, you know, like real authors, you know, I mean, I got paid a million, but you know, like, because I have a big following and that's kind of how books are done right now, you know, based on your following and how many.

SPEAKER_154: Celebrity status. Yeah. Yeah.

SPEAKER_08: I think if you've, yeah, it's, it's, it's even more than celebrity. It's, do you have distribution? So celebrity without distribution equals nothing in this new world. It's celebrity like Kelly Wurstler. She's got like 2 million followers. You know, she writes a book about design. She can get paid a lot of money for that book because, hey, you know, she got some, you know, number of followers. So she's like super popular. Now imagine, you know, she writes her book about design, how much she could get from each of the language models for like, let's say it's a coffee table book. Well, if you have a beautiful coffee table book of beautiful Italian design and they took all those original pictures and there's 200 beautiful pictures on a hundred pages in the book with captions and it really makes a great point. Think about how great that could be to have in your LLM when people ask about modern design Italian and you can pull up the high-res photos

SPEAKER_45: with a link to the book to buy it. So that can all be in the terms. We could do very granular licenses, Alex, where you say you could have a full res image. You could have up to three images for

SPEAKER_00: free. If you want 10 images and a thousand uses, it's $5 per thousand. If you want all of it and you want a hundred thousand user queries returned, you could literally price it per query returned and quoted. If you use one of these language models, go ahead and do a search your line and say, what does the wire cutter think the three best coffee machines are, three best coffee grinders over the years? And do that into your favorite three LLMs and be ready to pull it up. Alex, while he does that, I will tell you a story. I paid for the New York Times for many years. The number one reason I did was because wire cutter. I'm a consumer of products. I trust wire cutters. I love the guy, Brian. I tried to buy a wire cutter. New York Times bought it behind a

SPEAKER_08: paywall now. Consumer reports behind a paywall. Also ask, Lon, and consumer reports. Three best coffee machines. Alex, you have a New York Times subscription, I assume? Pay? Maybe? Okay. If you would be so kind as to pull up those two pages. What you will find is Alex Wilhelm pays for New York Times. He's going to pull up best coffee grinder. He's going to pull up best coffee machine. He'll show those two pages in a moment for Mark. Lon, do you pay for Chad Chippie-T? I do. Okay. Do you pay for New York Times? No. Okay. Perfect. So we have a Chad Chippie-T person and we have a New York Times person and Chad Chippie-T person. So Lon, just pull up your pages and share them. What did it tell you were the number one, two, and three according to those two sources from

SPEAKER_169: Chad Chippie-T? Chad Chippie-T is going a little slow, but the number one coffee machine is the

SPEAKER_31: Breville Barista Express. Number two is... According to... This is Chad Chippie-T sighting wire cutter. Perfect. So he knows what it is. Yeah. Breville Barista Express is number one. Number two,

SPEAKER_169: Philips 1200 series. Number three, DeLonghi La Specialista Arte Evo. And then it goes on from there. I'm going to share the clawed ones in just a second here. Okay. And so Alex, what does it say

SPEAKER_76: on the actual page? Is it relatively correct? I'm going to need you to specify... They have like 16 different sections here. So best programmable drip, simple drip, budget drip, poor... I think the drip. Just do the drip.

SPEAKER_04: But I would also add, Chad Chippie-T also says, I don't have access to wire cutters articles due to

SPEAKER_169: restrictions. So I'm sharing some highly regarded coffee machines and grinders from the past few years recommended by experts and enthusiasts. Okay. So it knows now. Now I have that one,

SPEAKER_01: the Technovorum. I have bought both of those. Those are great machines. I had the top one at

SPEAKER_93: the office. I had the second one in, you know, whatever, uh, at home and in the ski house. Beautiful, beautiful, expensive, but great.

SPEAKER_169: Claude notably did go to actual wire cutter and consumer reports and pulled because it can search the web now. So Claude actually lists, here are wire cutters, top coffee maker recommendations and consumer reports. And then the same thing for the grinders. So it says wire cutters, top three are

SPEAKER_02: the OXO brew nine cup coffee maker. That's a great one. I had that for a while. Yeah. The Bonavita enthusiast eight cup drip coffee brewer and they Breville Bambino plus.

Chamath Palihapitiya: Perfect. So what you'll find from this, if you do it yourself, is it generally gets it right. And it generally means you don't need a New York Times subscription anymore.

SPEAKER_169: Uh, Grok also was able to search and went and actually pulled the real one.

Chamath Palihapitiya: So if it searches and it links to the page, that would feel more like a Google search,

SPEAKER_00: which would be okay. Um, and a Google search might actually put it there in the, in the snippet up top. The point being, you now can just cancel your subscriptions. The consumer reports that we'd have to do a better study here, but this is at the key to the lawsuit from the New York Times to the, the open AI is that, you know, they're end running and they're confusing consumers and then consumers don't need to subscribe. I literally am exhibit a for this. I did not renew my New York

SPEAKER_08: Times because I was like, I can get this information for my chat GPT subscription. And, um, and there you have it folks. I think this will keep happening. And then people have to put their stuff behind a paywall because of this unfairness. And, um, you know, to Sam Altman's credit, OpenAI is respecting

SPEAKER_45: and fixing things. Now the entire industry must unite as one and set the terms with each LLM. They need to write a legal letter to Google. They need to write a legal letter to Claude,

SPEAKER_08: Meta, Grok, et cetera. They need to inform them. We are a consortium of people who have content that

SPEAKER_45: is valuable. Here's how we want it to present it. Here's how we want you to do robotics. And content people are so dumb and they're so disorganized and they're so me that they don't know what a slam

SPEAKER_08: dunk case this is. This is, and this is why the New York Times is not falling for it this time around. The New York Times has been screwed in the past, but now they have a subscription business. When you have a subscription business, you need to look at what the music industry and Disney has done. These two organizations will sue you into oblivion. Music industry, a hundred times more than Disney, Disney a hundred times more than New York Times, but New York Times, now that they're a subscription

SPEAKER_45: business, they know, you know, there is no fair use. In their mind, the music industry believes

SPEAKER_08: there's no fair use. That's ridiculous and absurd. And they have taken that position with YouTube. So when you watch the reaction videos I've played here before, they will try to stop reaction videos. They will try to keep them from getting monetized. YouTube came up with a great system called content matching, content ID. If you're dire straits or, and somebody, the Daily Doug reviews dire straits,

SPEAKER_00: and he talks about this, how sometimes even though he's doing criticism and they will try to stop him

Chamath Palihapitiya: from doing valid criticism on a song, uh, because, and he's interrupting the song five times. So it's not like you're getting the CD for free. You know, you're, he's interrupting it constantly. I mean, I suppose you can edit it down, but you know, they will sometimes take his, uh, advertising revenue,

SPEAKER_01: I guess, or if he has no advertising on it, but you have to fight for that. And sometimes it doesn't fight because he's already got one strike against them and three strikes are out. Yeah. I mean,

SPEAKER_02: that's the problem, like content ID. I agree. It's a, it's a, it's a brilliant solution, big picture to

SPEAKER_21: this problem. The, the issue is that on a, you know, cause it's, it's, it's, it's IDing so much content per day. There's a lot of false positives. It tends to really lean on the content owner's side. So, you know, for like honest trailers, we couldn't even include like a five second snippet from a movie where they're singing a song or we'll lose our entire video, even though, you know, we're making fun of it. We're not just playing the song and listening to it. And they can take your advertising revenue. The whole video would then belong to the owners of that copyrighted song. And we played, you know, a three second. And what if you had three,

SPEAKER_176: you played five different songs from five different movies. Yeah. Any, any one of them could claim it

SPEAKER_02: and steal the money. And even to an even more granular, we did, uh, years ago, we did a parody

SPEAKER_21: of Mary Poppins and we rewrote all of the Mary Poppins songs, but because one of our parody songs sounded too close to the original chimney sweeper song, they dinged us. And we had to come up with a new parody chimney sweeper song that sounded less like the original Chim Chimney.

SPEAKER_08: So anyway, get together. This is an amazing moment for content companies to get paid. It's an amazing moment for the language models to create a sustainable, fair system. The only way you'll get technologists to do the right thing in my experience, when it comes to content is to do what the music

SPEAKER_72: industry did is, which is hold the line. And you just have to be relentless, never drop the suit, never compromise, go after every single infringement. Now that sounds crazy because I have

SPEAKER_00: complained about the music industry and their approach before, but I think that's how you get their attention. They clearly got Sam Altman's attention and now he's like doing these things. And if you ask for Disney characters or to do stuff with Disney characters on chat GPT, it's stopping. So if you were to ask it to make you a Star Wars, you know, short video of, you know, Darth Vader fighting Darth Maul, I think Sora is not going to let you do that.

SPEAKER_02: It's really fascinating how specific it is now. Like last week for All In, there were some parody

SPEAKER_21: videos of like, you know, a member of the All In team is the Pope or whatever, and that's allowed, but you can't make something that chat GPT thinks is making fun of the person. So like dressing them as the Pope is okay. But if it was like, make them the Pope and there's ketchup dribbling down

SPEAKER_02: the front of their Pope outfit and on their face, it'd be like, no, no, no, can't do that. That's

SPEAKER_27: that's over the line. That's too mean spirited. Okay. But Jason wants the media business to aggregate, to kind of like maybe collectively bargain perhaps as some sort of like union against the major AI companies. Not one of those. No, it's got to be lawsuits. It's got to be capitalistic.

SPEAKER_00: I'm just poking. No, but I mean, I get your point, but it is the difference is the content companies love unions. Let me say that one more time. Content companies love the unionization of

SPEAKER_93: content employees. You know what happened? One of the major content, I want to say farms, but you know, major online publisher said to me when, you know, all these, I guess, Vox and, uh, business insider and, uh, Buzzfeed, they all went unionized, you know, all those unions. I don't know. Were you ever

SPEAKER_203: dragged into that, Alex, to be a union? Uh, no, but I was, uh, I'm in favor of media unions. So I've raised my hand to take part in one before. Great. Biggest mistake ever for both of you.

SPEAKER_00: You know why? I'm in the writer's guild. No, but that's fine. That's like for the, uh, for movies and TV, but I'm talking about for these content ones, the owners of those businesses were laughing

SPEAKER_08: at you, Alex. You know why somebody like you is should get paid more than the average. And what they said was these idiots, this is literally what they said to me, like, you know, late night bar kind of situation. These idiots, like they don't, they have no idea how much money they're saving us. And when an all-star like yourself says, Hey, I deserve a raise. Listen, I'm bringing all this traffic or Molly Wood or, you know, uh, Kara Swisher, pick somebody, Walt Mossberg. They would get paid five times, 10 times what a rank followed. They would say, Oh, you know,

SPEAKER_163: we wish we could pay you more, but you know, it's against the union guidelines. We wish we could give you, Oh, you want more days off? We, we would do that, but it's the union. They loved it. Loved it.

SPEAKER_217: The reason why I'm, I'm only a 90% capitalist, maybe 95% is that I, I take a ding to support my, uh, my friends, but let's not get into Alex's politics.

Chamath Palihapitiya: You would take out, you would take a pay cut and be like, let's say you were not an entrepreneur now. Cause you went an entrepreneur. So you voted with your dollar. You're telling me like a younger Alex would be like, pay me 10% less so that like other people who don't drive the results I have.

SPEAKER_76: When I was managing crunch based news and I was really advocating to get a couple of people, some reasonable comp increases. I offered to take a pay cut to make it up. My Lord. I mean, I take care of, I take care of my people. I don't know, man. My God. I mean, but there's the corporation who's, but when you're running a team at a corporation,

SPEAKER_27: you often can't take care of your steps underneath you as you think you should. And so you have to really go at it with every tool you have to grind the money out. But again, I don't know. Now I'm

SPEAKER_217: like, this has sent me into a total tailspin. I know. I know. I am a capitalist, but I'm also a person who has been broke before. All right. Okay. Um, we're going to office hours. We're going to talk

SPEAKER_67: about something that I'm very excited about Jason, which is what if you're a live streamer as we all hear our live streaming, but you really want to be a frog or perhaps Darth Maul. Well, then you need

SPEAKER_217: some cool AR tooling. There was something called snap camera that got taken off, but don't worry streamers of the world. There's a startup called stream fog part of launch accelerator 34 that is making all this possible and more. So please welcome to the program. It's Kevin Bonzio. Kevin. Hey,

SPEAKER_230: Hey, how's it going? Hello, sir. How are you? Where are you calling in from today? Austin, Texas,

David Friedberg: Austin, Texas. I've been there. Local. Great. Are you coming tomorrow to hang out at the office with us? We're having a little founder day. Did you know about that? I know. And I will be there. Of

SPEAKER_232: course. Oh, I will see you there. We'll be having some Stubbs barbecue perhaps. Love that. Alon

SPEAKER_234: and I scouted Stubbs on Friday. We did. It was good. Man, that pork rib was no joke. Yeah. That pork, the pork ribs are very good from Stubbs. Yeah. And also that fried okra. I was,

SPEAKER_235: but fried okra does not travel. I had to throw it away when I got home. No. Yeah. You gotta,

SPEAKER_00: you gotta have that fresh. So Kev, maybe you could show us what you're working on. Um, you know, pictures with a thousand words and then tell us what's working. What are your wins? What are your fails? What's working? What are the blockers in your startup? Um, yeah. So, um, my

SPEAKER_239: name is Kev. Um, so in content creation, Jason, and you know it as basically being our ideal customer

SPEAKER_241: product placements and ads play a crucial role in like monetization. Um, but the implementation can sometimes be disruptive. If you have like a full screen banner or even a clip, uh, playing, some viewers will just hover the YouTube timeline to skip it or be on their phone. Right. And, um, there's, and we basically think that their stream folk uses augmented reality to place those ads organically as part of the content in a natural way. That's kind of our idea. And you already said it so I can like showcase it to you basically. So as you can see, I can keep talking. I have an organic way of just interacting with you and the viewer just still sees an ad placement. And that's kind of how we think ads and video streaming should work in today's world.

SPEAKER_14: For those of you who didn't see an Uber Eats just flew by Kev's face and, um, it was distracting,

SPEAKER_08: but not debilitating for the stream. Lon, Alex, if we did this during this week in startups, would you be appalled or think it's fun if like an Athena assistant ran in and, you know, took some notes?

SPEAKER_02: Would you be like, I mean, I think these things always come down to context. You know, if it's a serious show for doing a very serious segment and then it'll goopy animated character shows up and dances around behind me, not appropriate, but I do think depending on the vibe of the show and the hosts and the kinds of things you're talking about, something like this could be really fun and a great way to do an ad. It's not, uh, it's not only not like distracting, like takes over the whole show and then the content stops, but it's not, um, you know, like it feels organic. It feels like it's

SPEAKER_21: part of the show and not like this thing that's interrupting the show that you have to get through

SPEAKER_249: in order to get on the other side of it, which I feel like is the vibe of so many ads.

SPEAKER_27: Especially if it replaces an ad that I'd have to skip. Otherwise, like if Kev said, Alex, listen, you can see this ad live while I'm talking to you, or I can stop and put a full screen pizza hut ad that I definitely take what he just showed. That said, Uber Eats Kev sends me so many push notifications randomly trying to sell me booze that, uh, I never want to see that particular

SPEAKER_251: animation again. Thank you. That's a great note. There should be a setting for people who are sober

SPEAKER_253: to not get pitched on booze. That's such a good idea. But isn't it strange too? I feel like for so

SPEAKER_09: many years, we were told the algorithms are so smart. They're so advanced. They know everything

SPEAKER_21: about you. Isn't it weird that the algorithms don't yet know after years of presuming that Alex

Chamath Palihapitiya: is sober? Well, an algorithm might actually know he's sober, have figured out that sober people will binge and getting them back in the fold. That's horrible. I hope it's not that.

SPEAKER_261: I think it is actually. Remember, the algorithms are indifferent to outcomes. They're indifferent to

Chamath Palihapitiya: anything but an outcome. The algorithm is indifferent to anything but success. Success is defined as increased orders. Yes. If you want to increase orders, getting a person who is sober to fall off the wagon could be a daily 12 pack. It might be that it learned that. In fact, we were just talking on Megyn Kelly about how the algorithm at Meta was taking deleted selfies or selfies that weren't published. Yes. Yes. And then for young girls or women, then feeding them beauty products based on the selfies they either removed or didn't publish. In other words, I published a

SPEAKER_00: selfie and it's like, oh my, I have bags under my eyes. Great. Here's like some cream that theoretically

SPEAKER_02: removes the bags under your eyes. It was targeting girls who deleted selfies with beauty. I presume

SPEAKER_27: Kev is not trying to knock me off the wagon here. So I'm really curious about the split in your business because when I was just learning about StreamFlog, I saw you guys offer the service for

SPEAKER_67: free and then you can buy certain assets. So I thought, okay, cool. Kind of a simple freemium business model. But now the focus that I learned more really does seem to be about creating this new ad

SPEAKER_30: format. So can you just tell us how much progress you made on having this be a product that is in the market versus a cool demo that you just showed us? Yeah. So maybe let me start by saying we have

SPEAKER_272: like a creator tool for mostly Twitch streamers. They can use AR effects to just have a more interactive content and that is and will stay for free. So they can just have fun. We love them being engaged. But we have like kind of the other side of that marketplace or the B2B side where we allow marketing agencies or brands to run these campaigns on Twitch. And we recently started doing that on YouTube as well. So we have these two sides where the creator side is for free. But the interesting part is we can offer these creators now sponsorships. So we kind of act like a marketplace where we build both sides of that marketplace simultaneously. And now every creator cannot only have these fun, goofy effects, but say, hey, actually, I want to sell my camera space as a billboard and earn more revenue in an organic, non-disruptive way. So that's kind of our idea. We monetize directly the kind of the business enterprise side. Creators can use it for free and even earn money.

SPEAKER_274: Do you have the like Uber Eats coming to you and saying, here is an affiliate code, any new

SPEAKER_00: accounts open, you get 100 bucks and then you go to streamers and say, hey, we made this collateral. If you run it, every time you get somebody, you're going to get 75 bucks, we're

SPEAKER_01: going to get 25. Are you interested? Have you thought about that? Like affiliate network dynamics?

SPEAKER_276: 100%. And we already have two affiliate campaigns running right now. Uber Eats was like a one

SPEAKER_272: off campaign where an agency came to us, here's the campaign, here's the four creators that

SPEAKER_241: needs to run it. And we acted just like a tech platform that created technology and the visuals for that. So we were kind of like just bought as a technology at that point. But we also run affiliate campaigns with others.

SPEAKER_44: Does it have to be a live stream? Or can you go to my previous videos that were posted

SPEAKER_00: in, in some way, put a layer on them? So if I did a deal with Budweiser, you know, Bud Light, you could go like if I'm Dylan Mulvaney, you could go to every previous TikTok I did and somehow insert it in there. I guess it wouldn't be possible. You have to repost the video, but

SPEAKER_280: it'd be kind of cool, you know, if that was possible.

SPEAKER_272: Yeah, it's a great question. So we right now integrate with OBS and mostly live streamers, but we ran our first YouTube campaign last month. And we're now looking, it's okay, how would this work with YouTube, right? Do you as a content creator upload your video? And then we can add those after automatically, like we're now looking into this YouTube market, like how will it look

SPEAKER_00: like for a video that was pre-recorded? I think actually, sometimes closing the loop is a really interesting concept to get the flywheel going. So here, you know, you had the one off from Uber Eats, but what would you say is the open affiliate product that has the highest commission rate? Is it still credit cards or an E-Trade Robinhood account? Do you know that answer?

SPEAKER_241: It's a good question. I heard that in the gaming area, there's these like energy powders that have

SPEAKER_00: great commission base. Perfect. So let's take the energy powder as a concept. Here's what I think would be an interesting test for you, since we're doing office hours here. You make a relationship with

SPEAKER_08: one of those. You say, hey, listen, I got this thing. You're going to get not only the affiliate,

SPEAKER_44: but you're going to get some shine, right? You're going to get some CPM campaign. I'm wondering if you

SPEAKER_08: would hire us to go and do these campaigns with influencers on their streams. But we will also,

SPEAKER_45: for this fee, flip their best of, and then we'll make shorts for you. And then we will run advertising of the shorts against their audience and retarget them. And then we will share the results with you, and it will be a win-win-win situation. The streamer doesn't have to do any additional work because we're doing all the clipping of those videos and posting them with them. You don't have to do any additional work. We outsource it all to us and we'll get you more sales and more eyeballs on yours. And we'll do it exclusively with you. So we'll give you 12-week exclusive. We won't do any

SPEAKER_00: other energy products. We have 20 different streamers we work with. We're going to get as many of them involved as possible. And then we win because you used our tool. We want you to be a Lighthouse customer. And then we want to obviously generate revenue because we're a startup. We need revenue to raise more money. It'd be like a very interesting packaging where you sell them a group of 20 people and you get some CPM. You say, hey, $500 per streamer upfront. So we need $10,000 for that. We need $10,000 to run the campaign. And then we want whatever percent of the sales. What do you think?

SPEAKER_241: I think that's a great idea. And as you said, these brands and campaigns often want to have this

SPEAKER_290: full all-inclusive solution. So that would actually be a great upsell potential at that point.

SPEAKER_08: So anything with your business that you need help with? Now you're in the 34th accelerator class. I'll see you tomorrow. So we'll have some barbecue. We'll talk some more. But just here for the sake of the audience, anything you're struggling with, any blockers, anything confusing in your business as the

SPEAKER_272: founder? I have one particular question. When we run these campaigns, the brands are pretty happy. The creators are happy. And we can measure in absolute terms, engagement range. But for us, I'm always trying to say, okay, how do we actually do a fair one-to-one comparison with traditional formats? And I always come back to the idea of, do we have to run a case study, a very official blind test to run this? Or are there other ways to really show them the benefit really in numbers?

SPEAKER_00: There are firms that will go do recollection kind of things. So if you've ever been on a website and

SPEAKER_08: it's like, hey, will you take a survey for us? And it's like, hey, you know, we're just doing some advertising surveys. And like they do a post survey. Hey, did you see any of the following advertisers on this website? And like Samsung did some campaign and it's like Apple, this, this, this, and Samsung. And they're just trying to see if it lifted that. Those are like for when people are spending millions of dollars, maybe tens of millions of dollars with a publication, et cetera.

SPEAKER_00: You're outside of that. People are going to do this for the vibes. And you can come up with your own metrics, which are, uh, you know, we know, uh, we have a list of a hundred streamers and we approach all hundred on your behalf. 40 of them that, you know, who are in the top category said no, because they want five to $10,000 in advance. So we'll manage that relationship for you if you want. But of the 60 that didn't want an advance, we were able to hit 37%. So we hit 37% of the top 100

SPEAKER_08: streamers. And so you had a 37% saturation rate, uh, or completion rate. And so you're making up this

SPEAKER_00: metric, but of available, so of available top 100. So it's not the 100 it's the available 100 who do these type of deals. You know, we got your 57%. So you can kind of come up with your own

SPEAKER_08: metric. And I think reach is always the issue with these. So this is where it gets really compelling. You know, you could then make a super cut of 10 different, oh, this would be great. So let's say you land the 37 streamers out of the hundred, but they represent 37 of 60 available or 30, let's say, let's say 35 of 70 that are available. So you have 35% of overall, you got 50% of what's available. Then you say, we made a super cut for you. This is a, uh, a seven minute cereal. This is a seven minute reel of the best moments of those 35 with your product that you can put on your website,

SPEAKER_00: 20 seconds each, 10 seconds each. And, uh, we'll also do marketing for that. So you're creating this

SPEAKER_08: extra collateral for them. And you say, Hey, you can use these clips on your social media for one year, and then they have to come down and you can put up to a thousand dollars in boost against each one, whatever you spend in boost. You just have to share it with us so we can share it back with them. And if you boost, you know, $10,000, uh, we just want 10% of that going back to the streamer and us. So that's where, like, if somebody wants to do a deal with me, I'm working on a celebrity deal or micro celebrity deal with somebody for GLPs, right? I've been taking these GLPs to help me lose weight. So now somebody wants to do a deal with me, um, to do this. Basically, I think what my people will negotiate is, Hey, if you spend a million dollars on advertising, it's 15% back to J Cal for appearing in those. So it's 150 K let's say those ads really perform and they do 5 million of them. Whoa, Whoa, yum, yum for J Cal. Maybe it's 750 K is my fee, but that makes sense because they wouldn't run extra ads if it didn't. So there's just like a ton of opportunity there. And I think owning a vertical is also great because then you can replicate it. So is, is gaming the one you want to go after? Is it dating? Is it consumer goods? Is it product on back? Is it people who are doing,

SPEAKER_04: what's that called? Muk? Mukbang. That's where they eat a lot of food on camera. Oh,

SPEAKER_261: I've seen that. Yeah. Well, I didn't look at you for any personal reason when I said mukbang.

SPEAKER_02: I've never done. People don't want to watch me. It's like an attractive person eating. It's not like, Oh, is that what it is? You're beautiful to me, Lon. It's not, it's not just you watch somebody eat.

SPEAKER_21: It's watch somebody charismatic and charming eat, not yeah. Watch the big tubby guy eat.

SPEAKER_311: Kev, don't grow a beard. What do you think? What's the, what's the ideal vertical for you to

SPEAKER_272: really get the ball rolling here? For us, we've been doing all kinds of stuff right now. Like we haven't focused on one particular area because that's maybe also another thing. We, we work a lot with agencies and those agencies are often spread across different verticals. So we're like offering that tool and they cross sell us, which is for us very nice reoccurring revenue stream. And we've

Chamath Palihapitiya: been working efficient for that relationship, but it's inefficient for raising prices and your

SPEAKER_00: offering. Right. So I would, you know, I think a homework assignment might be to find an area unique to you, like a secret. And then you've got this secret weapon. You don't publicize it. You don't talk about it publicly. You don't put it on your website, but you secretly build your little consortium. We mentioned video games here, but it could be stock traders. It could be fashion, whatever it is. You get your little collective going and say, Hey, we would like you to be part of the secret collective to do this. And then you can market them. So you get a little community going,

SPEAKER_08: you know, maybe you start a group chat with them, you know, or you just try to start those

SPEAKER_45: relationships, but it feels like you're onto something here. Most people want you to sell a platform fee and charge for just the software, but kickstart this and get it going. I think maybe

SPEAKER_08: having little networks where you can bring them volume. Cause I I'm, I'm perceiving what will happen in year two of this business, which is you're doing a bunch of one-offs. It's not profitable. You're losing money on each customer. One of the great things about this week in startups, as an example is, you know, we got up to at 1.6 days a week during peak Zurb because we kept selling out. And I told the ad sales team, Oh, if we sell out, just add another day. Problem with that wasn't burnt me out, but we had something scalable here. I sense that I just can't do as many. Look, we could be five days a week again, no problem, but we're doing three

SPEAKER_00: because I don't want to burn out. So we're leaving money on the table, but you want to get to that

SPEAKER_316: point where you've productized it enough that you can take orders, right? And they, and they are

SPEAKER_290: profitable orders. So you're sustainable. Yeah. No, it's a great point. And you actually like said what we are experiencing. We have these one-off projects that show product market fit. People are willing to pay, but it's about like really scaling it and making it like profitable

SPEAKER_08: in every transaction. Yeah. All right. You'll get there. Keep grinding and we'll see you tomorrow in Austin for our Tuesday founder jam session. It's the jam session where we, you saw here in office hours, Alex is like the light version of a jam session. Imagine, you know, 11, 12,

SPEAKER_00: 15 companies in a room, each going through their product and then helping each other and what

SPEAKER_08: they're struggling with becomes really dynamic. You're gonna have a great time tomorrow and we'll

SPEAKER_322: see you 10 AM tomorrow in person at our offices at the capital factory. Shout out. Thank you so much.

SPEAKER_325: Okay. Thanks Kev. Jason, I want to throw in streamfog.com. I forgot to say that in my intro. I just want to make sure I gave them that shout out. What do you think about affiliate links? I feel like

SPEAKER_02: he could do a pretty decent test with those, like, like a regular ad where you just read with those powders, right? Like powders. Those are affiliate. Yeah. A little guy holding up a sign telling you the website and then you just compare contrast. There's a little guy get us more than just a person saying it out loud, because I really do feel like that's one area seeing the link,

SPEAKER_21: reading it might have a bigger impact than just hearing a person say it out loud. Absolutely.

SPEAKER_27: Yeah. I want to throw in just one tiny thing perplexity, the AI search company, the latest bit of news adopted just before we went to air today, Jason, is that they are raising

SPEAKER_67: another $500 million at a valuation that could be $14 billion, up $5 billion from $9 billion last

SPEAKER_217: November. That's quite a lot of capital. I just wanted to throw a question. Is there a revenue

SPEAKER_67: number, by the way, associated with the firm? Yeah. So Perplexity's CEO, Arvind Srinivas.

SPEAKER_335: I think Arvind Srinivas, yeah. We should have him back on the program. He's great.

SPEAKER_66: Practice that. Sorry, Arvind. 100 million ARR, that was announced about a month ago. So the company is theoretically at 140X ARR multiple. It's a hot company. Ridiculous, but there was just

SPEAKER_00: this talk of Apple maybe talking to them. Eddie Q said he thinks search is moving to maybe, and I think he name checked in his thing, Perplexity, that he had met with them. So that is a little bit of catnip for investors. Oh, there's an Apple purchase, perhaps.

SPEAKER_27: The question that I have is just, you know, why do you think they need another $500 million? They've raised a billion now in a couple quarters, if this deal lands where we think, well,

SPEAKER_285: straight up, just that? Yeah. I mean, if you can raise and everybody's raising,

SPEAKER_08: you want to be opportunistic. They probably also want to invest in not their own infrastructure, because there should be plenty of that available to them, but perhaps they want to do some infrastructure. That would be a great way to raise money because investors love investing in infrastructure because they consider that a defensible, even if it's not, but they do consider it defensible. And you can ask for a big number and a big valuation if you were going to buy, let's say the $500 million was going to be $300 million in data center and H100s. You'd be like, hey, listen, we want to get those because we think it's a competitive advantage.

SPEAKER_00: The other one is maybe they have new products or services they want to invest in or explore. Why not build a cash position if AI is this hot? Also dissuades other people, capital as a weapon, dissuades a competitor from emerging.

SPEAKER_137: All right. Well, I appreciate that. I was just curious because I'm like, wait,

SPEAKER_106: half a billion again? I was thinking acquisitions. The other possibility, somebody came to them, a sovereign wealth fund, they met at an event and

SPEAKER_00: they're like, we really would love to be involved. And you say, okay, yeah, well, do you have a ticket size? And they said, yeah, we like to put in 500 million as our ticket size. Okay, would you like to join the board as an observer? And it could have been opportunistic that way. Could it be that

SPEAKER_72: opportunistic in that they think the market's hot and their revenue growth is good? So why not?

SPEAKER_347: Marvin did say 6.3x growth year over year. So like, clearly this thing is scaling.

SPEAKER_316: 15 million to a hundred million is pretty, pretty good time to raise is when you have performance.

SPEAKER_02: I am curious about if they're going to have their sort of breakthrough moment, because even, even now with me spending all day in this world and thinking about it,

SPEAKER_21: using all of these, like you see the one that we went to just now to do our consumer reports. It was, I used Grok, I used Claude, I used OpenAI, Google Gemini would have been fourth. So perplexity, they've got ground to make up. As good as the product is, people just, it's not top of mind, even for me.

SPEAKER_00: And there's a Google labs. I didn't, I forgot to put it on the docket, but I was playing with this week and there's a Google labs for like a new LLM first chat thing, which we'll talk about on Monday. So I think the big question for that company would be is how defensible is it? Like if, and I remember they were using other people's LLMs and then, you know, Sam Waltman said they wanted

SPEAKER_44: to do a search engine. They had done a search engine at some point. I don't know whatever happened to the ChatGPT search engine. You can still use it. I use it all the time.

SPEAKER_00: Oh, okay. Yeah. So I guess there's a search engine from ChatGPT and it's, you know, when you're doing a format, like a design, like Naver's comprehensive search design or the design we did from Mahalo, the design can take you only so far, but there are people who are addicted to perplexity. I do hear

SPEAKER_08: some individuals who just love the perplexity response, but they don't have distribution. So they have to fight to build that distribution now. Meta and Grok have this incredible built-in debt, you know, in the language model. And I don't know why Reddit doesn't have that yet. The fact that Reddit doesn't have the Grok-like feature where you can press on a thread and have it summarize the thread is like, what are they doing over there? Yeah, they're behind.

SPEAKER_27: That is a really good thing that they do now on X. I see people pinging Grok all the time. Oh, all the time. Every thread. Such a coup. So I wonder what percentage of Grok queries come from people on X asking it because

SPEAKER_217: on one hand, hella distribution, native usage, hell yeah. Also not cheap probably to run all those queries.

SPEAKER_02: Oh, it can't be.

SPEAKER_361: They have a certain amount of infrastructure, so they might as well use it.

SPEAKER_02: True, true, true. But even like I posted the perplexity, it's in Twist Taping, if you want to take a look, like even the front page, when you first go there, it looks exactly like ChatGPT. And ChatGPT is already a name brand. If you think about like the kids are like, oh yeah, I'm on ChatGPT all day. Like, how do you compete if you're basically doing a similar thing? Hard to change consumer behavior.

SPEAKER_00: Yeah. Hard to change. I think the results look different. I think they were, you know, using a much more rich result on mobile specifically to, to do it. But you know, it's, it's, this happens every time we have a new vertical, there's an Instagram, and then there's the 20 other sharing apps that nobody remembers. There's the Pinterest, and then there was this next and 20 other social shopping sites that nobody remembers. So it is who can find a unique, sticky way to acquire customers and keep them.

SPEAKER_93: So like, those become the two things. How do I acquire customers? And then how do I keep engaging them?

SPEAKER_02: So you're going to be announced, Humane, which is owned by the PIF, the, sorry, public investment fund, they're going to quote, operate and invest across the artificial intelligence value chain as a unified operating company. So what do we actually know about Humane? They're going to offer services, products, and tools powered by AI, including data centers and infrastructure, AI models, including a high powered Arabic LLM that's in the works. And it's going to bring a lot of government help to make Saudi Arabia a more of a center for AI. Here's the quote, Humane will also streamline various data center initiatives, procure hardware,

SPEAKER_21: and accelerate the adoption of AI technologies.

SPEAKER_76: All right, Jason, this is just a Contra G42 Emirates strategy, right?

Chamath Palihapitiya: What it is, is you, if you believe that controlling the LLM in your native language and in your geographical region is important, you would best be served as a nation state to control it and to

SPEAKER_00: invest in it. So that's, I think, what we're seeing here is, do you want Sam Altman, Microsoft, Google coming into the region and saying, we're putting up data centers, we're going to control the language model, as we just talked about in the previous perplexity segment? Okay, people are doing their searches, who controls it? You know, okay, well, he who controls the result controls society, right? Like this is, the results could have dramatic impact on the society. And that you probably do not want to leave up to another nation state. If you look at the impact of Google or Facebook globally, Facebook would be the better one. Because Facebook, you know, was used in a bunch of revolutions, it was used politically, it was used subversively, it's used culturally,

SPEAKER_08: TikTok would be another example. So if you draw that parallel, yeah, you're going to want to probably

SPEAKER_00: not have TikTok controlling what people think about, I don't know, what's the number one political issue

SPEAKER_08: they had impact on? Probably Gaza, sure. So if you, if you're in the United States, do you really want the Chinese impacting how your populace, especially young people feel about geopolitical issues? I'm putting aside how you feel about Gaza or Israel or the whole conflict. Just do you want them doing

SPEAKER_226: that? The answer is no. All right. Google lost monopoly case last year about its search market share.

SPEAKER_67: There is now the remedies part of the case, Jason, when we figure out what the government's going to tell Google to do to fix this problem. Amicus briefs, friend of the court briefing, essentially, you send in a letter expressing your views with a little bit more context. It's a way to weigh in. Then Y Combinator, the famous startup accelerator, weighed in on the Google antitrust case. So what did they say? Well, generally speaking, my favorite quote was YC supports plaintiff's proposed remedy package as a whole. Essentially Reddit, sorry, Y Combinator made my argument, which is that startups should love antitrust because it prevents major companies from crushing huge parts of the market. Then Gary Tan did go on Twitter X later on and did say that they're not directly calling for Google to get broken up, maybe as a later on penalty. But he's like, look, YC backs companies. These companies are taking up all the market and they're being unfair about it. I thought it was a surprising position for a tech leader to take, given that there's often

SPEAKER_217: a anti-antitrust vibe, Jason, from folks with love like yourself.

Chamath Palihapitiya: And why isn't Gary owning it? I mean, this takes substantial work to create this document and

SPEAKER_08: substantial costs. This was done at a cost of millions of dollars, a million dollars. Some legal group was engaged by Y Combinator to explicitly say that Google should be broken. And now Gary's back, walking it back. That makes no sense to me. This is a premeditated expense. This took 12 months, six months at a minimum, more likely 12 months to create this in my mind. And I can't see a law firm doing this for less than seven figures. So Y Combinator took their management fees, a million dollars in management fees in six to 12 months of time to create a, how many pages is this report, this brief? It was hundreds of pages, I think. The Amicus brief itself is, and I'm scrolling

SPEAKER_27: through it now. Everyone bear with me. Only 14 pages. Oh, it was only 14. Okay. So maybe that's,

SPEAKER_08: maybe that changes it. Maybe it was a quarter million dollars, 500. Maybe they did it in three to six months. But this is an explicit act that requires engaging a law firm. And to make a very thoughtful argument, they must have had hundreds of hours of meetings. It's not, my point is,

SPEAKER_376: this isn't a tweet. This is a legal paper. Yes. Yes. I mean, it could, I, my, my own,

SPEAKER_21: you know, take on it would be like, my guess would just be that it's political for him and that he doesn't want to be seen as somebody who's publicly attacking Google. He's the CEO.

SPEAKER_02: Even if the organization, I mean, it is a, it's a fine line to walk for certain.

SPEAKER_261: He made the decision. Right. He's the CEO. So why would you walk it back? It's probably,

SPEAKER_176: there was probably an, probably somebody from Google is like, Hey, why did you do this?

SPEAKER_02: You know? Yeah. I mean, that, that is what I'm thinking. It's political for him personally

SPEAKER_21: to sort of be like, Hey, this is what, as an organization, we sort of aligned this way, but it's not me being an enemy of Google. I don't think this is a good position for

SPEAKER_102: an incubator to take and just put it out there. And he probably regrets doing it. You think that's why he's walking, walking it back a little bit?

SPEAKER_93: Probably negative reaction. And probably people were like, I'm trying to have a partnership with Google. I'm trying to collaborate with Waymo. We're trying to, you know, get investment from

SPEAKER_00: Google venture. I don't know what, but you know, it just, it feels like too sharp elbowed.

SPEAKER_21: I would say maybe damage control for VC companies that are working on Google deals right now.

SPEAKER_93: It could be. Yeah. Maybe they're like now, maybe Google's like F by comedy. Right. Yeah. And now it's going to have blowback on their portfolio.

SPEAKER_334: That's exactly what I was thinking. How hard is it to be pro antitrust as a startup investor, when you're trying to sell your companies to the big companies you're antitrusting against? Like that's a really thin, that's not very wide.

SPEAKER_256: This maybe feels like, you know, like you're going, what this is, is the equivalent of like,

SPEAKER_08: you go in the prison yard and there's like, you know, five big guys and you're like,

SPEAKER_00: I'm going to walk up to one and that, you know, just recently got jumped. Google got jumped by the DOJ and like the person's on the ground and you're like, yeah, F this guy and you come in the teeth.

SPEAKER_46: But like the person's already on the ground, they're getting beat up and you just pile on. This is like a pile on move. Yeah.

SPEAKER_389: My commentator wants to join the Latin Kings.

SPEAKER_46: It's literally like, it's not like you do this. If you did, I'll tell you what, I would respect it.

SPEAKER_45: If you did it and there wasn't an antitrust case, you said, I believe that these are antitrust, these are behaviors I think are anti-competitive, which I did. I said, I thought it was anti-competitive to put Google local above Yelp. And, you know, I was very clear about that. Like, this is my belief. You don't have to like it, but I don't think Google should be able to put the one box. I complained about the one box as well back in the day.

SPEAKER_08: I said, I don't think they should be able to put the one box at the top and take the number one coffee maker off wire cutter and put it in the one box.

SPEAKER_00: I think they should ask for permission for that, you know, or that's what the ads are for. And I don't know whatever happened to the one box experiment. I think it's gone and now it's the LLM. But the point is, this feels like a pile on. It feels like there's some weird agenda here that maybe wasn't well thought through and just own it or don't like maybe withdraw it. So, you know what, we put it in, you know, but the whole thing was saying they should be broken up, that they support the breakup. And now he's saying they don't support the breakup.

SPEAKER_27: I want to be super clear that they support the overall remedies package as a whole. And they do note later on page 12, an effective remedy order should, in our view, also include

SPEAKER_67: the government's proposed contingent spinoff requirement for Google's Android platform. So what they're saying here is maybe you don't pull that trigger first, but that should be on the list. And then Gary is saying, let's not break up Google. So you could thread the needle. It's a different statement.

SPEAKER_08: It's not just no threading of the needles. They said they support it. The judgment said to spin out Chrome, and then they're saying spin out the store. And so that's, that's the breakup. Either you are for the breakup, Gary, or you're not for the breakup. You file an amicus brief. That's a serious thing to do. You ask for a breakup. Now you're saying you don't want the breakup. I'll be honest. Breakup is good for startups. Breakup is good for startups. Obviously it is. Unless, you know, you spin it out. And then that new boss is even more cutthroat. Maybe they'll have sharper elbows. So there is the risk of that too. Like if you were to spin out Android, maybe Android would be more cutthroat with startups. If you were to spin out Chrome or you would, maybe Chrome comes up with a, you know, their own paid ad system and they're going to charge more for it. You know, so everything you, you don't be careful what you wish for you spin out YouTube. Maybe YouTube becomes a really powerful presence in the world or Android becomes a really powerful presence that says, you know what, we're going to just make our own apps. That's actually the perfect example, Alex. Android spins out the new CEO of Android. She says, you know what? You see, I put the word she in there to get virtue signaling points along. She says, uh, I thought you said for the YouTube CEO previously was a woman.

SPEAKER_195: Yeah, very much a lady. Doesn't that have to be virtue? Yeah, he's just accurate. New CEO of ChatGPD.

SPEAKER_45: Um, you spin it out. And she says, you know what? All these like apps you have out there. Oh, there's a fitness app. Oh, there's this.

SPEAKER_01: You know what we're going to build a better version and put it free on your desktop.

Chamath Palihapitiya: We are going to sweep through. We're going to make our own game studio. We're going to make our, when you buy an Android phone, it's going to come with so much free stuff. You never have to buy an app again. That's our new value prop.

SPEAKER_01: And yeah, we're going to lower the fees in the app store to 10%. So it would be a better deal for you. But we are going to compete with what's in the app store.

SPEAKER_402: Like, what if that happens? That's a distinct possibility.

SPEAKER_404: What if Google made their own maps functionality? Or maybe Google made their own email functionality. That would be, that would be tricky.

Chamath Palihapitiya: Well, they did it very slowly. And you know, uh, the Apple was very, very clear about this.

SPEAKER_00: At some point they said, you know, we have notepad and we still promote Evernote. We have them at our keynotes. We very slowly add features to the notepad app. It's considered like a basic app.

SPEAKER_08: If you're a power user, you're not going to use notepad. And then they're like, but we had it, they had it. Did you notice they added collaborations to notepad like last year in 2023? Or when did they add collaborations to notepad? I think it was 23. Like really? 15 years after like collaborate and word and Google docs and Evernote, you know, collaborative document sharing. They added it 10 years, at least 10 years after Google doc, which was bought by the way.

SPEAKER_322: Like that was how Google got into the sharing space. Okay.

SPEAKER_67: Apple will boil you alive slowly. Anyways, I can't find that directly, Jason, but we should wrap anyways. We're back on Wednesday.

SPEAKER_407: We're back on Wednesday. See you next time. Bye bye. Bye everybody.