SPEAKER_00: When Meta launched its Vibes product, we all mocked it. It's another feed of videos generated by AI, everyone called it slop. But when it comes to Sora 2, everyone on X and even yourself are kind of demanding access to it, fighting for it. And so the question we had was, what's so different about Sora 2 from OpenAI versus Vibes from Meta? I'm curious what you think.

SPEAKER_02: Well, I think...

SPEAKER_05: This Week in Startups is brought to you by Perspective.ai. Surveys? They never capture what customers are really thinking. That's why we use Perspective AI. The candor we get back is eye-opening. Real insights stray from your customers. And the first two months are on us. Just go to getperspective.ai slash twist. CLA. Innovation takes balance. Our CPAs, consultants, and wealth advisors can help you get from startup to where you want to end up. Get started now at claconnect.com slash tech. 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.

Chamath Palihapitiya: All right, everybody, welcome back to This Week in Startups. I'm Jason Calacanis. You know me, angel investor, host of All In podcast.

SPEAKER_09: Maybe you know me from back in the day, Silicon Alley Reporter, or Weblogs, Inc., or Severian High School. Who knows, Fordham?

SPEAKER_11: And with me, my co-host, Alex Wilhelm. How are you doing, Alex?

SPEAKER_00: I'm doing fantastic, Jason. A couple of IPOs and some big government news. Lovely news day for us. All right, let's get started then. What's at the top of the docket? The top of the docket is what the government shutdown means for startups. Jason, I wanted to tell people out there who are building stuff not to worry too much. There are a couple of things to keep in mind. For context, the U.S. government shut down at midnight after the two American political parties couldn't find common ground on how to keep the government running. Now, Jason, this is not really our domain, but I did find a couple of things that matter. So first of all, the IPO market is expected to stall or slow as the SEC kind of slowly shuts down. The Department of Labor may stop processing certain things that you need for H-1B visas. So the immigration world could also slow down a little bit. Startups that sell to the government, defense tech, I'm thinking cybersecurity, companies like that may see slower procurement and slower payments.

SPEAKER_05: And also, we're going to have less economic data. But I thought we'd take a moment and say, don't panic, founders.

SPEAKER_09: Yeah. And if you look at the history of this, I think these typically get worked out in days, right?

SPEAKER_15: I think the longest one, at least Producer Claude is telling me this, Producer Claude says the longest one in U.S. history, 35 days from December 22nd, 2018 to January 25th, 2019. And that was an impasse over the border wall funding. I remember that one. Previous record to that, 21 days during the 1995-1996 budget standoff under President Clinton. And then there was a 17-day closure in 78 and a 16-day shutdown in 2013.

SPEAKER_09: Yeah, I mean, this is part of our democratic process. It's messy. The media would make you think that, like, it's the end of the world. And I guess if you work for the government, not getting paid could be the end of the world. But if you're a government employee, you anticipate this stuff, I think. And it's not that they typically lose their jobs or lose that money. The money gets paid back. So a lot of them, I think, still come to work. At least that was my understanding. And then they just don't get paid for two weeks. And then that pressure causes folks to work it out.

SPEAKER_15: So I think it's great that we – I didn't know that about the SEC. I should have assumed that in the H-1B. So that's a really good catch. I guess if you were in the middle of something, you could experience a 10-day delay. Worst case, a 30-day delay. So if you just say – you know, a lot of times when I have to think about these things, I just prepare for double the last worst one.

SPEAKER_16: So if the last worst one was 35 days, I'd say, okay, I guess the worst possible case scenario would be 70 days, maybe double what the last one was. And then you just deal with that. Can you weather that storm? And clearly you can.

SPEAKER_18: 70 days would be a long time to sit here and listen to the debate about it. Next up on the docket, Jason, Wealthfront is going public.

SPEAKER_00: Now, Wealthfront is a robo-advisor, very popular and well-known in the pre-ZERP era. Quieter since then, lots of venture capital backing. And the company's S1 filing shows a company that's growing quickly and has quite a lot of profitability and is finding a foothold in the Gen Z market.

SPEAKER_04: I'm very bullish on this, Jason. Your thoughts?

Chamath Palihapitiya: I'm a shareholder. Andy Ratcliffe has been on this program no less than five times.

SPEAKER_09: He was the co-founder of Benchmark. He teaches a product market fit class at Stanford.

SPEAKER_15: And he's really smart. And, yeah, I became a shareholder. Man, maybe in year one or two, I've gotten all my family on Wealthfront.

SPEAKER_09: It's really a great way to build an optimized portfolio and pay the most reasonable fees you could ever do. I think they charge 50 bps, 25 bps. It's something ridiculous. It's 25?

SPEAKER_20: Yeah.

SPEAKER_09: So, you know, if you have a money manager, it's typically 1 to 1.5. Now, that doesn't seem like a lot. But if your returns are 4% a year after taxes and you're paying 1.5 to your money manager, guess what? Now you're, you know, giving a third of your profits every year to your money manager.

SPEAKER_16: And what they did was they said, well, you should make this robo-advisor. You say what your goals are. And then on a scale of 1 to 10, how much risk you want to take. And they do a blended portfolio. They also do things like tax loss harvesting, which rich people have access to, but poor people don't. What does that mean? They look at your portfolio. They individually trade stocks.

SPEAKER_22: If you lost a bunch of money on your Intel and made a bunch of money on your NVIDIA, you could sell the Intel to get a loss to offset the gain, right? As you move your portfolio around.

SPEAKER_09: And they do that for you all the time. You know, traders do that inherently, but most people don't have the time for it.

SPEAKER_15: And so, yeah, I've tried to get every family member I know using a money manager. And the money managers are annoying. They try to put you into products you shouldn't be in.

SPEAKER_09: And Wealthfront tries to put you into the most balanced, rigorous portfolio. It's kind of the opposite of Robinhood, where people are day trading and doing, like, individual stuff. This is for people who, like, don't even want to look at it. Although Robinhood does offer portfolios now. They offer everything.

SPEAKER_15: And I think they might have added crypto to a blended portfolio here because I think they had so much demand for that at Wealthfront.

SPEAKER_23: I think that's becoming the standard everywhere.

SPEAKER_00: I think it's interesting to talk about the two parts of their business because we discussed the bits they charge against your, you know, portfolio. But when you take a look at their actual income statement, what you see here is that their cash management business is actually quite a lot larger than their investment advisory business. Because no matter how big your AUM is, 25 basis points is 0.25%. It only scales so quickly. So what I find very interesting about Wealthfront is that it's a very good company. It's profitable and it's growing. And there's much to like here, but it is sensitive to interest rates. And so if we do see interest rates decline in the U.S. over the next, I'd be picking number of quarters, three, four, five, it could dramatically harm the company's ability to grow. So keep in mind that the bits business is good, but the cash management business is better. Got it.

SPEAKER_26: Got it. Okay.

SPEAKER_09: So if they have cash and they're sweeping it around from different accounts, they make some sort of spread on it, I guess. Yeah.

SPEAKER_15: And this company's been around for a while, over 10 years, certainly. Fintech's a great category. We've done very well for it, you know, as an investor. So in our firm, Robinhood, Wealthfront, two great positions for us.

SPEAKER_09: And, uh, yeah, I had the opportunity to sell before the IPO, you know, there's always like these moments where you can sell and I elected not to, because I just think there are so few public companies and people really want to have access to Fintech. And with stable coins, new crypto regulations, with boomers, my thesis is boomers are going to die. I hate to be cruel about it, but everybody dies. It's true. And then those boomers are going to, you know, have their money sitting in some, I don't know, you know, what, what account, but, you know, some Goldman Sachs account, right? And grandma and grandpa die, and all of a sudden $3 million goes down to their three kids. They each have a million dollars.

SPEAKER_15: Are they going to keep it in Goldman? No, they're going to put it into the Wealthfront. They're going to put it into their Robinhood. They're going to put it into their Coinbase account. That's why I think these companies are doing so well is because there's just a generational shift. Most people don't want to use these old banks. They want to use the new banks. Most people don't want to use the last paradigm. They want to use the new paradigm. The interfaces on these are incredible.

SPEAKER_22: If you open up your Bank of America, it's just disgraziad when compared to using Venmo or using Robinhood. It's so crazy to me that these big companies can't just hire elite designers to refresh their app.

SPEAKER_28: If you were working at Bank of America or any of these major banks, like literally just go find a design firm to tell you what to do and just spend a million dollars having three different design firms make you three different versions each. Now you've got nine versions and just have the CEO or the president, whoever's got taste, say this one is the best.

SPEAKER_16: And let's go with that. But it's always organized. This is the problem with committees and design. You never get great design with a committee, period, full stop.

SPEAKER_31: You need to have an artist. You need to have somebody with taste and chat GPT and all due respect to LLMs.

SPEAKER_16: You know, they do interesting things, but they're not going to nail it. They're just not going to nail it on design. They're going to give you a lot of, you know, four, five, six and sevens. They're never going to get you to a 10. You know, maybe we'll be here in five years. It will. So I think they're great for mocking things up, Alex, but they're not great for the finished product.

SPEAKER_24: I like the idea of people playing and coming up with ideas, but then you've got to have a really great designer finish your album art and your podcast or your logo.

SPEAKER_28: Hey, listen, we meet a lot of early stage founders here at Launch, my investment company, and some, they don't have a lot of traction yet. They just have an idea. Maybe they haven't even finished their product. They've just got an MVP. But they still need investors and accelerators like ours to take them seriously. And you know what? We can't just wire money to your Gmail or your PayPal. That's not how it works, folks. We need to know that you're a legit and official business. We need to know your company is incorporated. That's why you need Northwest Registered Agent. It's the service that will help you run your business the right way from day one. In 10 clicks and in under 10 minutes, you're going to file for your LLC or a C-Corp if you're a startup, get a domain name, launch your official website, claim your business email, and even fast track your trademark application, which some people forget to do. We're talking about more than just company formation. This is your entire identity as a business. Go to northwestregisteredagent.com slash twist and show the world you're in business.

SPEAKER_34: And make sure you use that URL slash twist so they know that we sent you.

SPEAKER_18: The generational point you made, Jason, is actually very salient because if you take a look at who is signing up for Wealthfront,

SPEAKER_00: and on the screen right now, I have a chart that shows new clients by generation over at Wealthfront going from 21 through 25, and there is a steady increase in the number of Gen Z customers that are signing up. So clearly, they're the next generation of people that are saving money. Do they have boomers? They do. That's the lightest one at the top.

SPEAKER_41: Yes, up here at the top. Not really a big category for them, so Goldman's probably safe in the short term.

Chamath Palihapitiya: I can tell you that really matches mine because when I try to get, you know, the boomers in my extended family, like, to get on this,

SPEAKER_09: they're like, I don't want to change. You know, people don't like change. And, you know, it's like, do you want to live in the same house you've always lived in? Or, you know, now that it's just two of you, would you like to be on a one-level house that's new and brand new and is built to your specification, your lifestyle? Maybe you want to live in an apartment where it has the concierge and a door person and everything. And so, nope, I want to stay in my rickety old house until I die. And that will be me on my ranch.

SPEAKER_22: I'll be on my ranch with my chickens and my longhorns, and they'll have to get past the chickens and the longhorns and my bulldogs to get to me.

SPEAKER_00: Actually, it doesn't sound so bad. Jason, one more thing on this. On the venture capital front, there's a number of firms that own a lot of this company. Tiger Global owns 22 million shares, or about 20% of equity as far as we can tell.

SPEAKER_05: DAG Ventures, nearly 14 million shares. Oh, Dad. DAG Ventures, my mistake.

SPEAKER_51: No, I think you can call them DAG. I call them Dad. Keep going.

SPEAKER_00: Index Ventures, 12.9 million shares. And Ribbit Capital, the well-known fintech-focused venture firm, 9.8 million shares.

SPEAKER_53: So, here's hoping to price as well when it does trade, maybe after the government's done being shut down.

Chamath Palihapitiya: And this is where having a theme-based firm like Ribbit Capital has can pay dividends if you get it right.

SPEAKER_09: Now, if you went for climate, and we tried to do a climate practice here, actually with Molly Wood. We're very public about it. And, you know, the reason she's no longer here is because I had to shut down the climate business. It was a really hard decision for me because I just, you know, after over a year of doing it, I was like, none of these firms have the founders, or very few of them have the dogged founders. They're all overpriced. Gosh, I mean, who's going to make money here? And I'm an LP in Chris Socker's lowercase carbon funds. I don't, you know, I don't know how they'll do. Hopefully, they do great. But, you know, it's just a constant problem with certain verticals. Now, if you hit the vertical, like fintech, and you hit it at the right time, these can be incredible. But then sometimes you're in fintech, let's say you have a vertical, but you hit all the, you know, C and D players. So you pick the right vertical, but you pick the wrong companies. And then, you know, some firms don't like to invest in, you know, competing products. So, you know, you might be in a, you know, whoever, you know, you might be in Lyft as opposed to Uber or Postmates as opposed to DoorDash, whatever it happens to be. Not that Lyft and Postmates weren't great companies. They just weren't the big economic winners.

SPEAKER_58: So venture then is right theme, right companies, and right time.

SPEAKER_00: So you have to kind of hit all three of those to have an outlier fund. Jason, why is fintech having such an amazing year? We've had Klarna, Circle, eToro, Chime, and now Wealthfront trying to list. That's at least five IPOs that we've talked about.

SPEAKER_48: Is this just these companies reaching maturity, or is there something else going on here that I'm missing? I think it's the incompetence of the incumbents.

SPEAKER_61: I mean, the JP Morgans, the Citibank's, the... E-Trade. Oh, yeah.

SPEAKER_09: Even PayPal's an incumbent to Venmo, you know, and then just, yeah, I think the incumbents, don't seem to have the product velocity of these new players. And man, you watch Wealthfront add features, Robinhood add features, Coinbase add features. And these are all founder-led companies. So I'd say it's both of those things. It's the legacy sucking and the founder-led companies, you know, hitting their stride. This is why I tell all founders, if your product velocity isn't high, then you are the problem.

SPEAKER_15: The CEO is the problem. The designer is the problem.

SPEAKER_16: The developer is the problem. Everybody's the problem. You have to be shipping early and often. And I think for some of these companies, their product velocity, while they hit scale, is increasing.

SPEAKER_22: That makes you twice...

SPEAKER_09: That makes you doubly difficult to deal with. Like, Google now is releasing products at a pretty fast clip again. Sure. So Sergey's back in the saddle, I notice, kind of correlates. So now, like, Chrome has a Google Assistant. I think we're going to do a demo of it. All of these things keep happening where if a company can release product at a fast pace and they have scale, right now, if Chrome...

SPEAKER_15: How many Chrome users are there, Producer Claude? I'm going to guess there is 2 billion Chrome users globally. I think there's probably 5 billion people using browsers.

SPEAKER_09: So maybe they have 40% of the market. I'm just remembering probably old numbers, but that would be roughly 2 billion people using Chrome. If there are 2 billion people using Chrome, let that sink in. That means 2 billion people are going to have access to Google Gemini and be upsold on it.

SPEAKER_00: Jason, according to producer Claude, based on the latest data, approximately 3.98 billion people use Chrome. So call it 4 billion, roughly half the planet, give or take. That's a lot of people regularly using the browser.

SPEAKER_41: But I think probably Android's market share around the world helps that because Chrome, of course, is also on the Android operating system.

SPEAKER_00: If you want to use Claude like we do, you can go to claude.ai.twist and save, I think, 50% off your first three months of a paid plan. We live and die by Claude here.

SPEAKER_67: 50%? That's a very generous deal. Thank you to our friends at Claude. Absolutely. What's next on the docket?

SPEAKER_00: Next on the docket is OpenAI debuting Sora 2. OpenAI, of course, is the American foundation AI model company. They released Sora a couple years ago, a video generation model, Jason, that had a lot of hype behind it but didn't quite meet expectations. Sora 2, they say, is much, much, much better. And they're baking it into a new social app called Sora, which you can download over on iOS.

SPEAKER_22: If you were on X yesterday, you couldn't miss it. I mean, there was slop everywhere. And all of the slop seemed to be based upon Sam Altman. So is that, is just like Sam the default character in Sora 2?

SPEAKER_18: I think what you're seeing there is that people in the technology world are trying to signal to one another that they're in the in crowd, so they're doing Sam Altman clips.

SPEAKER_70: It's an insider joke.

SPEAKER_00: Jason, I have a number of clips from Sora here that are non-Sam Altman, if you'd like to take a sample.

SPEAKER_70: Okay, sure.

SPEAKER_73: Yeah, show us some slop.

SPEAKER_00: All right. From the Venture Twins, I give you a Sora-generated video of, well, Jason, I think you can just see what this is in one second. Observe, my friends, Pikachu at D-Day.

SPEAKER_76: Pikachu, stay!

SPEAKER_27: All right, that's IP. That's somebody's IP.

SPEAKER_11: That's $250,000.

SPEAKER_05: You sent it to the Venture Twins.

SPEAKER_00: And then we also have a series of interviews between different historical figures presented in a news format. Jason, take a look at this.

SPEAKER_78: I've been treating you today.

SPEAKER_79: Ah, busy as ever, my lady. Was up before the sun, mucked out the goat shed, fetched water from the well, then mended the cartwheel that split yester even.

SPEAKER_82: Sounds like a full day already.

SPEAKER_83: You've been in quite a few duels.

SPEAKER_84: How many victories are you up to now? 37. Each one witnessed. Each one alive in memory. And looking ahead, how many more do you expect to win? Every duo that finds me. Antirazat.

SPEAKER_86: Centurion, how long do you think...

SPEAKER_09: That's pretty good. Yeah, that's pretty good. Yeah, very creative. I wonder what the prompts are on these were. And I'm noticing also these are longer. Is there an upper limit on the number of seconds you can do? Is there a cost to this?

SPEAKER_22: Because my understanding of generating video is it's always been limited to like five seconds unless you're paying because you're basically like burning a hole in the ozone layer every time you make an hour of video.

SPEAKER_89: Maybe that's changing. I don't know what the...

SPEAKER_93: One of the themes we talk about over and over again on This Week in Startups is making sure you do your chores. I'm no expert on these things. I have some experience. Steven Estes from CLA is an expert. Let's talk about being cash efficient. Tell us about efficiency and what you see in the top tier startups in your practice.

SPEAKER_95: We're seeing kind of an interesting trend out there where companies aren't needing to raise quite as much as they had in the past. You really have to be careful as a founder to only take on as much money as you really need. You've got to do the forecasting. You've got to do the modeling. And you've got to dial in and get it right. Otherwise, you're going to end up either not raising enough capital to get to where you're going. And you're going to have to go get venture debt or go back and have an extender to the round.

SPEAKER_96: Or you're going to give up too much of the company because you just didn't recognize how much money you actually needed.

SPEAKER_98: Yeah, very important to get this stuff right, folks. And that's really a bummer when startups don't do things in a button-up way. I always have a great partner, a good partner to have on this adventure. While things change, my friend Steven over at CLA.

SPEAKER_100: Visit claconnect.com slash tech. And don't forget to mention that your boy Jake Al sent you. That's claconnect.com slash tech. Start today.

SPEAKER_18: So I think nearly every clip that we've seen has been nine or ten seconds. I have one of Bob Ross painting a fake Godzilla being attacked by sticks.

SPEAKER_00: The Pikachu one was nine seconds.

SPEAKER_18: So in the case of the last one, I think what they did was they took them and then stitched them together to make it look like a longer overall video.

SPEAKER_73: So maybe nine seconds is the limit here. And it's free? Or is it for only paid users?

SPEAKER_18: My understanding is that it's free. And I think that's probably why they're limiting how many people can go into it right now.

SPEAKER_00: Because remember when they launched the AI Studio Ghibli moment that we talked about on the show and they were melting their GPUs? I presume right now they're trying to keep demand at a moderate level to see what the load looks like.

SPEAKER_27: They should have just made it for paid users.

SPEAKER_09: I don't know why they didn't do that. Because then it would, you know, it's fairer to the OpenAI ecosystem. And then we would have it right now because we pay for it. I pay for a corporate account. This is really infuriating. I don't, somebody at OpenAI, just think of how you, what you're signaling to us is like, if we pay you, we don't get it. And it was the same thing with Google. They always give their free products to Gmail users as opposed to Google Docs users. These are two different namespaces. It's the exact opposite. If I'm paying for Google Docs, then give me the Gemini Assistant in Chrome first because I'm paying you cash. Both of these companies have it backwards. I don't understand this philosophy.

SPEAKER_72: We're worth the paid users.

SPEAKER_05: Do you think they're just fighting for overall mindshare in the AI world? So maybe OpenAI here wants to attract the every man, the everyday person?

SPEAKER_09: In Google's case, I think it's a technical limitation. They have to go rewrite everything to go work in that other namespace. And so that's what I think is happening. They make it for the big giant namespace. Then they eventually will bring it to Google Docs. I think they just got to figure out how to make it work in both places quickly and just give the paid people the first 60%.

SPEAKER_00: I think your enthusiasm here, Jason, actually underscores an interesting question that we had on the production team today, which is when Meta launched its Vibes product, we all mocked it. It's another feed of videos generated by AI, everyone called it Slop. But when it comes to Sora 2, everyone on X and even yourself are kind of demanding access to it, fighting for it. And so the question we had was, what's so different about Sora 2 from OpenAI versus Vibes from Meta? I'm curious what you think.

SPEAKER_09: Well, I think maybe the success of the first Sora makes a built-in audience for the second, and Vibes is a first-time product. That would be, I think, why people really want to see the sequel to Terminator versus whatever the new film is, because they fell in love with the first one.

Chamath Palihapitiya: It's probably as simple as that, but also the reputation of Zuckerberg and Facebook and Meta is, you know, to just create consumer-based stuff, and the reputation of OpenAI is to create industry-leading stuff. So if Facebook releases something, it's typically not innovative. It's typically copied from somebody else. When OpenAI releases something, you're going to assume, oh, this is the best-of-breed product. So I just assume Sora, if it's coming out of Sora, it's the best one. If it's coming out of Grok, it's the best one. If it's coming out of Anthropic, it's the best one. If it's coming out of Gemini, it's the best one.

SPEAKER_09: If it's coming out of anyone else on the long tail, you know, Mistral or, I don't know, what's the one in China, DeepSeek? DeepSeek, Z.AI, sure. Or Apple or Microsoft or Amazon right now.

Chamath Palihapitiya: I think they're behind, so I'm just like, okay, yeah, it's probably not going to be super impressive.

SPEAKER_00: It's pretty good. It's pretty good. I'm actually impressed with this, much more so than I was with Sora 1. I think the consistency is pretty good. I think the fidelity of the physics representations are pretty good.

SPEAKER_18: And I got to think, Jason, studios are going to use this ad nauseum because it's got to cost, what, 1% of using human?

SPEAKER_73: Let me bring in producer, editorial director Lon for a second.

SPEAKER_09: Editorial director Lon could one of these short videos, like three seconds clip,

Chamath Palihapitiya: but if you were to put it into one battle after another, Paul Thomas Anderson, we went to see as a group, would anybody in the audience notice it if there was like a five-second or nine-second clip of this?

SPEAKER_116: I don't think so.

SPEAKER_118: I think we're already at the point where you could slip in a few seconds of AI. I mean, we saw it, there was that Eternaut. We looked at that clip of that Argentine sci-fi series where they use AI for a little bit. I don't think people would have any idea if it was faster. Maybe you left it linger on screen long enough for people to notice glitchiness, maybe.

SPEAKER_120: But three seconds, four seconds? No, I could already pass.

Chamath Palihapitiya: You know, I can tell this because I was watching, re-watching Ashoka, the Ashoka Tano, Anakin Skywalker,

SPEAKER_09: aka Darth Vader's Padawan with my daughters last night. We re-watched the first episode, which was great. And I looked at it through the lens of AI and she goes like into a temple and, you know, it's like some ancient Jedi ruins and et cetera. And I was like, yeah, this could have been done with AI. This feels AI-ish and it is, in fact, taped on those giant screens.

SPEAKER_118: Right. The volume, that's what they call it.

SPEAKER_126: The volume.

SPEAKER_118: So the volume. It's not just, it's backlit, it's LED and it's projection so that it matches, it gives it more of a 3D feel.

SPEAKER_127: So you can move the camera around and the background stays consistent even though it's green screen.

David Friedberg: So when you see the volume set, somebody did a drone shot over it.

SPEAKER_09: It's literally like somebody's ranch or like a parking lot in, it looks like a parking lot in the valley somewhere with giant LED walls around it. And then the ground was made of sand, like Tatooine. And they were showing like somebody had gotten a shot of like what they were filming. And it's like, oh, they just do the ground and some, you know, whatever the plant or the, you know, broken down droid on the ground is. But the backdrop is this incredible high fidelity LED wall. And like Lon's saying, you move the camera around in it, but it's fixed. Right. And the camera is able.

SPEAKER_127: The camera and the background are synced. So as you move, they're all moving in unison.

SPEAKER_118: And it gives you this, this, it's 3D effect of the, the, the environment is real. And I just dropped an image in twist taping. If you take a look at that, where you could see how they try to make it match the actual sky behind them. And so it gives you this really impressive three-dimensional feeling, even though you're shooting a green screen.

SPEAKER_132: Hey, you know what?

SPEAKER_48: People absolutely love those shows. So whatever they're doing, it's working really, really well. Shout out to them.

SPEAKER_09: Except when you watch Andor. And then when you watch Andor and you see like true production value where they make giant sets and they spend real money. This is going to be the big question when the new, the Mandalorian and Grogu movie comes out next year, it's looks like the volume, not like Andor.

SPEAKER_22: And the fan base now is involved in a big debate. And this is going to be a big debate that keeps going. Do you want more of this IP, but that is made in, I wouldn't say slop, but more towards the slop side, or do you want the classic, you know, really refined sets and the classic CGI? This will be the, the next big, um, debate we'll be having.

SPEAKER_09: That feels like, I think we'll go into pop culture. They're going to look at Mandalorian and say, yeah, it feels a little sloppy. It looks like AI slop.

SPEAKER_27: And then they're going to be looking at Andor, Rogue One and say, oh yeah, that's got real production value.

SPEAKER_00: Yeah, Lon, stick around for a second because next up on the docket is our main character of the day, Jason. And in this case, it's a woman named Tilly Norwood, but she's not actually, yeah, I remember

SPEAKER_136: her.

SPEAKER_00: Not actually a person, uh, actually a synthetic AI actress. And she's the creation of an AI talent studio called, and I'm guessing here a little bit, um, Sequoia, uh, which is part of the, uh, particle six group. Their brain person in charge here is Eileen van der Velden. Sorry, Aline, if I just ruined your name. But essentially the gist is there's now an AI actress that is supposed to be signing a talent deal for representation.

SPEAKER_05: And people are very, very unhappy. Lon, why are they unhappy?

SPEAKER_28: Shipping is so hard. But knowing what your customers want, that shouldn't be. And that's the magic of perspective AI. Take it from the world's greatest moderator. Our friends at perspective AI have built the world's greatest AI interviewer. All you do is you type up a simple prompt for your potential customers. Maybe, Hey, what are your thoughts on this new feature? Or maybe the specific pain points they're confronting in your service. Anything you want to know about. It doesn't matter if you're a hotel or if you're building a dating app. This product is so incredibly powerful. Within just hours, you're going to start getting useful, high level feedback with accurate emotional responses and instant summaries. All pulled together into research grade reports. We've been using perspective AI here at this week in startups, and it has been a game changer. I got to tell you, speaking to the AI interviewer, people are more honest than they would be with a written survey and that candor is more helpful. You want to know the truth.

SPEAKER_26: So sign up today, get perspective.ai slash twist to get two months free.

SPEAKER_142: To clarify first, we don't know for sure that she's definitely about to sign.

SPEAKER_118: That was this, this woman, Aline van der Velden, gave a presentation in Zurich where she showed off Tilly Norwood, her creation. And she said, I'm already talking to talent agencies. We're very interested in signing her. We don't have actual confirmation. In fact, the only person we've heard from a major Hollywood talent agency so far, Leslie Siebert, the head of the Gersh agency, gave a quote where she said she thinks Tilly Norwood is frightening and vowed not to sign synthetic actresses. So this might just be hype or they're putting this out there. Here's Tilly. She's not real. She is an entirely synthetic AI created performer. Now, what I think is most interesting about this story is that a lot of actors thought after the 2023 actors and writers strike that one of the big things they were fighting for was protection from exactly this sort of thing. But as it turns out, Justine Bateman, several other people on social media are pointing out that actually the 2023 SAG-AFTRA agreement doesn't actually, it expressly permits studios to work with AI actors. It says, here's the rule. A synthetic performer created through generative artificial intelligence can be used provided, one, parties acknowledge the importance of human performance in motion pictures and the potential impact on employment. So that's just literally an acknowledgement. And then two, notice to union and an opportunity to bargain in good faith over appropriate consideration, if any, if a synthetic performer is used in place of a performer who would have engaged under this agreement in a human role. So if they tell the union what they plan to do and agree to negotiate non-binding and they acknowledge that humans are good, they're actually allowed to use AI actors in whatever capacity they please. So a studio could make a movie starring Tilly Norwood and there would be no recourse from any actors or guild.

SPEAKER_09: All right. Do it. Yeah. Stop talking about it and do it. No, the whole thing is, this is all planted. The Gersh Agency, everybody's in on it. Gersh Agency probably has signed her already. They're going to do a whole announcement. Then they're going to announce her movie.

SPEAKER_27: This is just silly. I mean, how does this relate to Yoda? How is Yoda any different?

SPEAKER_118: Well, you know, Frank Oz, Yoda's voice by Frank Oz, the memorable Muppet performer who used to use his hand to move Yoda around.

SPEAKER_09: Yes. But there were people who were like, oh, they don't need an actor to do that. You don't need a literal person to do it anymore with makeup.

SPEAKER_22: And there was like a whole brouhaha about that, right? Sure.

SPEAKER_118: I mean, we've gone through many different versions of this. There's, of course, mocap where an actor can put all those sensors and then they can play an ape or themselves at age 10. And I think the interesting thing is that they were what SAG was very focused on in 2023 during their AI negotiations was actually like, could I use Tom Cruise's AI likeness? Could I use a real actor? Like, could I make a synthetic Angelina Jolie and put her in a movie? Yes, of course you can. Well, with that person's permission, but they didn't really deal with this whole area of, could we create a completely synthetic performer and begin casting them in movies? The answer is apparently yes. Yes, you can.

SPEAKER_65: There was a very interesting case of this, actually. Crispin Glover was in Back to the Future.

SPEAKER_52: Back to the Future 2. George McFly. Yeah. Right.

SPEAKER_28: He wasn't in Back to the Future 2. He was in number one.

SPEAKER_09: He didn't like, for whatever reason, he didn't like the screenplay of number two. Yeah. He didn't want to come back. And it wasn't that he didn't want to come back. He just had like, because he's like a pretty unique individual. Right.

SPEAKER_125: And they recast him in the film, but did somebody who looks exactly like him. Right.

SPEAKER_161: And the guy is doing an impression of him, too. He's speaking purposefully like him.

SPEAKER_53: Oof. Yeah.

SPEAKER_05: I would be very mad if I declined to do a movie and they found an Alex lookalike who talked just like me to replace me. That would feel scummy, I think.

SPEAKER_09: So they used makeup, camera angles, recycled footage to obscure the recast. Uh, and, um, they, uh, went to the screen guilds and they changed the use of people's

SPEAKER_28: likeness. And I think he got a huge settlement.

SPEAKER_118: It was settled in a collective bargaining agreement with the Screen Actors Guild. Uh, there are clauses now that forbid producers or actors from utilizing methods to replicate a performer's physical appearance if they're not in appearing in the film. So, yeah, you couldn't create like a digital clone of Timothee Chalamet and put him in your movie if the real Timothee Chalamet is not being compensated for being in your movie.

SPEAKER_00: Can I just ask who actually wants to see this Tilly Norwood in a film? Because to me, it's all a downside because people will boycott it because there's an AI character. Everyone will complain. And is there anyone who has a positive feel about going to see that type of movie? Because to me, there's just no one who is more excited about a film because it has

SPEAKER_132: a fake person in it than they would be if there wasn't.

SPEAKER_118: I think the idea is most people at this point probably would not care. And it's so much easier to make something with an AI person rather than a real one. Real actors, you know, get sick. They don't like the line. They don't like the costume. You know, like you just do it in an instant. So I don't think there's so much a desperation among audience members to see an AI movie.

SPEAKER_176: There's just most audience members might not care one way or the other. And it'd be so much easier. Everybody could make their own movie if we could just do it with AI.

SPEAKER_00: Wow. All right. Moving on. Today's Polymarket is all about OpenAI building a browser. The question, Jason, is will OpenAI announce a browser in 2025? And the reason why I chose this Polymarket, I'm going to show you the chart right here, is simple. There's a big jump at the end there.

SPEAKER_170: And when I was looking through Polymarket, picking out today's market, I wanted to figure out why. Why did the odds of OpenAI dropping a browser?

SPEAKER_09: Let's pause for a second here. Is this, they announce one or they release one? Because we have to look at how it resolves under rules, right? Because the title here is OpenAI Browser in 2025? So that's like very general. Anytime there's a general Polymarket, you got to get into the details. And this one says,

SPEAKER_00: Release is a standalone web browser that is intended for general web browsing and is available for use by the public in at least one country or region by December 31, end of the year. Okay, so it's released, not announced. Got it. Sorry, if I said announced, I mean-

Chamath Palihapitiya: You said announced, yeah. That's why I wanted to be like, because I was like, did they announce it? I thought they may have announced they're working on one.

SPEAKER_00: No, it's still roughly under wraps. As of July, we knew the code name for it, which is Aura, I believe. But no one's seen it yet. But the thing is, I saw this jump in probability. I know why I jump. Should I guess? Did you guess?

SPEAKER_36: I had a guess, but I was wrong. So I'm curious what you think. Well, the obvious guess is that Google released theirs and that puts pressure on them to release theirs because they got to play catch up, right?

SPEAKER_00: That's a much better idea than the one that I had. And it helps explain this. But I was thinking about there was no actual news showing that OpenAI had actually started to drop breadcrumbs or whatever. And so I was thinking, is this an example of Polymarket being ahead of the news, being a little bit smarter than what people might know based on the headlines? Because you always like to say the sharps are on Polymarket and it's a useful tool to figure out what's going on. I like your point. And it explains a bit of the mystery that I saw in the data.

SPEAKER_09: Well, what we're doing here is now we're trying to figure out how to leverage prediction markets. And so looking at this, going to the news and looking for people, looking for a news story would change the odds. Looking for a thread in a forum like Reddit, Hacker News, Quora, that would be another way to do it. And then the other way to do it would be is if there were any breadcrumbs in the code of say a new product like Sora, then you would maybe even get more. So if Sora said open in the OpenAI browser in the code base or in some notes somewhere, there are these people who study the code changes in the Tesla dashboard to look for the next car or the name of the next or the code name.

SPEAKER_22: They look inside of Apple's products for new hints and breadcrumbs of like, they'll be looking in some code base and then find something about Vision Pro 2 or something.

SPEAKER_00: So very interesting. But this worked out because I had a question. I didn't have headlines to back it up. I brought it to conversation and I figured out what's driving it thanks to you with just the Google and competitive pressure pushing OpenAI forward. And if you're curious, people now think there's about a 70% chance that OpenAI will release a browser by the end of this year, which would really shake up the market, Jason. So big things to come. But it has to be open to the public, but that could be paid users, right?

SPEAKER_31: So it doesn't say if it's paid or not. So, but it can't be in beta in a closed beta.

SPEAKER_16: So that does change my betting on this. I would have a hard time making this.

SPEAKER_190: No, it says, oh yeah, a closed beta will not suffice. Yes, closed beta won't suffice.

SPEAKER_31: Because, you know, then the problem with closed beta is, is like, does the team trying it at OpenAI, do they count? Like, does that count as a closed beta? No. Because it would be a closed beta would be the public, but in a closed fashion by invite

SPEAKER_22: only. Anyway, this is where it really, these prediction markets are going to be great because you

SPEAKER_09: could create a second prediction market here. Will they demo a browser before 2025, right? So I just love the fact that these exist. Well done to our friends at Polymarket. This one to me, I don't have enough insider clarity to place a wager.

SPEAKER_00: Ah, well, we'll see if they're right by the end of the year. All right. So for everyone knows Spotify, it's the well-known global music streaming platform pioneered the way that we now listen to music that has spread to Apple music and YouTube music, et cetera. Longtime CEO, one of Europe's, I think, most famous business executives. And he's not stepping away from the company entirely, Jason. He's handing off the CEO role to two lieutenants, and he's going to become executive chairman of the board, which he says in a note is a more active role than Americans are accustomed to. And he wants to go help build more big companies in Europe. So good. And I'm glad he's doing that. Shout out to him. Long tenure. Well done. But what I wanted to ask about, Jason, is when is it time for a founder to either change their role or step back from a company? Because we don't see this often. We like to talk about founder-led businesses. Founders Fund, of course, likes to back companies that are only led by founders. And in this case, he decided it was time. So in your experience as an investor, when is it the right time for a founder to shake up or step back?

SPEAKER_16: Well, let's pull up Spotify stock chart just while I collect my thoughts here and take a look at the, you know, max chart here. Um, this has been an incredibly successful company.

SPEAKER_15: Um, but you know, it took a long time.

SPEAKER_09: I was on Spotify when it was not legal in the U.S. yet. Somebody had gotten me an account. I won't say who, um, that I could use at my VPN. And it was a mind blowing experience to have all you could eat streaming music for one price. And man, there were like, I don't know, a dozen of us who had these hacked accounts, uh, would say how we got them, uh, cause I wouldn't get anybody in trouble, but statute of limitations,

SPEAKER_27: or we might have had these hacked accounts. I don't know. Allegedly we might have, maybe I'm misremembering actually now that I think about it.

SPEAKER_17: It was actually my cat. My cat had the account.

SPEAKER_27: Sure. Yeah, it was, uh, we were hallucinating.

SPEAKER_09: Um, and so then they had a great second act, which was adding podcasting. So really two great acts and they bought a bunch of podcasting assets, did a great job on that. The amount of time, the amount of paid users, the amount of free users, the advertising base, um, really a revolutionary product.

SPEAKER_15: Some people don't like the product or the history of the company, because as Oasis was saying, you know, like people will go out and buy two cups of coffee for 40 bucks in

SPEAKER_16: a cafe with a croissant, and they won't spend 10 bucks on a CD and have it for life. It's very frustrating for musicians that that era ended, but it also for the up and coming musicians who weren't accustomed to selling DVDs and having that revenue stream where, Hey, you know, a million people bought your DV, your CD, and you got $4 as the musician, $3 net net at the end of the day as the musician. It was like, okay, yeah, you just made 3 million bucks, right? You sold 10 million, made $30 million. This was big money back in the day, especially for double albums.

SPEAKER_09: Uh, brothers in arms did incredibly well by dire straits with the CD revolution, um, smashing pumpkins, infinite sadness, a double album, which then you make twice as much money on. Um, there was, this was an incredible money printing machine and, and it went down to a trickle, but for the next generation, they got incredible, uh, fame instantly. They could release a track on a Friday and then be headlining a music festival three months later. Uh, like we saw at Chapel Rhone, uh, Pink Pony Club and these songs, she was singing Pink Pony Club two years before she was the second headliner at ACL, Austin City Limits here. I saw it with my daughter and my wife, who's incredible. She was like literally playing malls two years earlier with the same song. So the velocity at which you can become Chapel Rhone went back. So there's that, it really did democratize, um, being a musical star and it changed how you make money. You had to be resilient. You had to have some inner resources to say, okay, I'm not going to make money off the CDs, but for Chapel Rhone, she never did.

SPEAKER_192: I'm going to make it touring, which is why Oasis just did a tour.

SPEAKER_22: And when they ended it at Wembley, Liam said, see you next year, because he knows that if they want to make money, they're not going to go into the studio and do an album. They're just going to show up next year and sell out Wembley for 10 shows and bag, uh, you

SPEAKER_16: know, hundreds of millions of dollars, like the billions of dollars Taylor Swift did. So Daniel Ek, if you look back on it, he, he, he basically reinvented two industries. Podcasting also changed. If it wasn't for Spotify doing these huge money deals, uh, with Joe Rogan, Bill Simmons, buying up all this stuff, they really did popularize and maybe, you know, accelerate the podcasting space by 30%, 40%, perhaps I would say something there and the music industry, they completely revolutionized. Yeah. So what an incredible run to impact two huge industries like that.

SPEAKER_22: Again, disruption, never easy. Uh, but what, what an amazing career.

SPEAKER_09: When should you hang it up? Um, you know, when you get to 50 years old, uh, or so, and you've got a ton of money in a bank account and you're like, what's the next 10 years of doing the same thing I've done for 10 years going to be like? And the answer is arduous, uh, or more of the same. And there's all these other opportunities. I respect somebody saying, you know what? Yeah, I'm good. Uh, Steve Jobs never did that. He worked till he died, but Bill Gates did. Yeah. Bill Gates and Jeff Bezos were like, you know what? There's a huge pile of money here. I could do interesting things with it. I'm going to start a space company. I'm going to start the Gates Foundation. I'm going to, you know, and, and valuation is part of this. Net worth is part of it. It does screw with people's brains. So take a look at this. When you look at that chart, I don't know what percentage of Spotify Daniel still owns, but by the way, this is $143 billion company.

SPEAKER_16: He has to own 5% of it. He might own as much as 15% of it. Um, but let's just say like Larry and Sergey, he wound up with 10%.

SPEAKER_31: He's worth 10 or $15 billion.

SPEAKER_196: Yeah.

SPEAKER_31: Um, let him cook.

SPEAKER_09: Yeah, that's it. This all just has to do with really personal stuff. You know, maybe you get a divorce. Maybe, you know, you got a sick kid, God forbid. Maybe you've got, uh, exhaustion. Maybe you want to try a different career. You know, all these things can happen. So I see it happen with my friends. You know, Chamath had his own venture firm. He had LP money. And then at one day he was just like, I'm not enjoying this. I'm going to just invest my own book. I own social capital, the brand. I'm, I think that's what went down. I don't know all the details, but anyway, he's like, yeah, no more partners. Just me. Everybody's out or, I mean, I kept some people, I guess, but he just changed the way of doing business. Sure. And I've looked at that myself. I'm like, I wonder if I would enjoy better. And, uh, I was looking at like how QSBS works, qualified small business stock, you know, that

SPEAKER_22: rule where the first 10 million you invest as an angel investor, you don't have to do. Well, that got raised to 15 million. So I was looking at my book of business and I'm like, wait a second, wonder if I should raise another fund or I should just do a stacked QSBS trust. You can make a trust for like each of your kids.

SPEAKER_16: So now let's say you put $25,000 into Uber or Rob, let's say Robinhood or like a pick open AI or something, right? You put 25 K into open AI when it was a hundred million dollar evaluation. And you had, uh, you know, a spouse and three kids, you put $5,000 into each one. The 5,000 turns into 15,000. You pay no tax on that 15,000 when each of the five different investments goes. And so they call it like a stackable trust. So things can change and people go, huh, maybe there's a better way to architect my life. And you should always think about that. You, you did that. You were work, you were working for people. And he said, I want to work for myself. So you have two customers right now, yourself with your newsletter and then me with the podcast, right? I think those are your two.

SPEAKER_09: You really need to have financial education at a young age, which is why I created kids investment club.com. If you pull it up, anybody who wants to work on this with me, I'm looking to hire a full time teacher and I'm looking to hire a full time teacher to work on curriculum for me. This being the tip of the spear.

SPEAKER_16: So I'm looking for a teacher who teaches, uh, I don't know what a teacher gets paid in Austin. Somebody asked producer Claude what a teacher gets paid in a high school science, a high school math teacher. What is a high school math teacher? There's some award-winning high school math teacher. So I created kids investment club.com. I had like 2000 people sign up for this, um, already. And so I want to hire a teacher who teaches math, finance, et cetera, who has a passion for this and then start this as a rolling curriculum. Like, you know, maybe we'll just start with Mondays after school, but then do a second cohort on Tuesday, Wednesday, Thursday, and Friday. So now you've got five cohorts, 500 people in each, you know, uh, and each parent pays $500 for a course or a hundred dollars, but you teach kids on a zoom call, how to invest

SPEAKER_31: money and how to do personal finance. And man, kids love money and they love investing. I think this could be a hit, this could be a hit of a startup.

SPEAKER_202: Uh, do you want to grab a question from the audience before we go? Sure. All right.

SPEAKER_53: Um, thank you, Brandon, Alex Medic, Henry, and other people for sending these in. Jason, I'm going to pick Brandon Buchanan from YouTube. He says, Jason, what platform is best for buying secondaries?

SPEAKER_00: Hmm. Which platform is best for buying secondaries?

SPEAKER_204: Which one currently advertises with us?

SPEAKER_27: I don't think we have one advertising with us. Um, and I don't buy secondaries.

SPEAKER_09: I'm a seller of secondaries. So I actually don't know. I apologize. I think there are things for you to look out for, which is the fees. Uh, some of these are now being sold with, uh, five or 10%. So if you want to buy open AI, uh, you might need, or Stripe, you know, you want to buy a hundred thousand dollars worth. You have to pay 10,000 on the way in to the person who's selling it to you. And you don't know what class of shares you have. You're flying blind. You may love Stripe. You may think it's the greatest company in the world. But I think not having information is a big challenge in these investments. Not knowing where you are on the preference stack, man, it's very dangerous. It's a very dangerous game to play unless you really know the promoter, the person who's running the SPV and you trust them and they have access. They're an insider. So, you know, they know the Coulson brothers and they worked there or they were an early investor and they're clearing their position. Like, okay, you, there might be some mitigating service there. I wouldn't get too involved in the secondary market unless you have an information edge.

SPEAKER_206: Yes.

SPEAKER_09: Information edge could come from, uh, knowing people who work at the company, um, being a venture capitalist or knowing a bunch of venture capitalists who can advise you on the state

SPEAKER_16: of the company, knowing board members in the company, uh, or being a user of the products in some deep and meaningful way. So if you really used Wealthfront, you had 10 people on it and opportunity came up, you're like, I think Wealthfront's a sleeper, you know? Okay, great. I could see doing it in those circumstances, but I would rather see people, um, invest in

SPEAKER_09: public market equities where the information is pretty crisp and clean and set it, forget it, or low, low, low, low fee index funds. Um, you have to have an edge and that's like, I think the problem. Um, so I started to build my edge as a public market investor doing J trading. Um, but man, it it's, it's hard. It's hard to do these things. So you have to be really honest with yourself. Now, if you're learning and you want to come to angel.university, which I'm hosting in a week or so, um, yeah, come to angel university, all the proceeds go to charity. You can learn about angel investing, read my book on angel investing. You could read about venture capital and there's like, um, Brad Feld's got a good book on VC, you know, go do all those things and try to build an advantage and learn the discipline. Like I did with public markets. I started really reading about public markets, how people value companies, you know, over the last two decades and really in the last couple of years, really trying to sharpen my knife by doing it every day for half an hour, an hour here on the program and in

SPEAKER_15: other places. But, um, bet small while you're learning, bet small while you're learning.

SPEAKER_16: If you're learning to play PLO and you're Alex and you've played no limit for 10 years and you want to learn PLO, you don't walk up to the highest stakes. You don't put 10 grand on the table. You go find a PLO tournament. That's $50. And that has a $50 rebuy. And you just, you know, that's your entertainment budget for the weekend. That's it. Uh, all right, everybody. Another great episode. Go to thisweekinstartups.com slash docket.

SPEAKER_09: You can read the docket and you can follow us on Instagram, YouTube, uh, and, uh, x.com slash TWI startups. And it's TWI startups on most platforms are just searched for this week in startups on the Instagram. Please do me a favor and just say, Hey, Jake, how in the comments, we are starting to do the docket, which we do publicly at thisweekinstartups.com slash docket. We're taking the top stories in the docket and we're doing a little storytelling experiment on Instagram. So we can pull that up for a second here. If somebody pulls up the Instagram account, uh, this is like a little, I don't know, like a little way for us to communicate with you. If you're part of the Instagram gang, you can go to our Instagram and you'll see us talking about what's going to be on the docket before it happens. So if we just pick one of those stories that relates to a story we did here, here's the Wealthfront IPO story. Um, and it explains Wealthfront. And then if you go to the next story, uh, with that arrow on the right, you can see, uh, here's us talking about the Sora app. You go to the next one. Uh, and we're just, just starting to do this last couple of days.

SPEAKER_15: Here's, uh, the Polymarket CEO, my guy, Shane Copeland getting, uh, flipped off at a meeting with regulators. Here's, uh, that one got 113 likes. So go ahead and just go over there and you'll be more informed. Here's the story about Daniel Ek, uh, stepping aside as, um, CEO of Spotify. All those stories we talked about today in real time, we're putting them up on the TWI startups account on Instagram.

SPEAKER_16: But, um, I just really want you to say, Hey, Jake, how, uh, in that group there, uh, and I'll, I'll reply to you if you do that. Okay, everybody, we will see you next time on this week in startups, which will be Friday. We'll see you Friday.

SPEAKER_151: Bye everybody.