Robinhood’s Vlad Tenev on AI, Prediction Markets, and the Future of Trading | Ep. 33

Transcript

We call it vibe trading. So, there's vibe coding. I think there will be vibe trading, which sounds a little flippant. Well, but the thing you said, that we're not a trading platform, it's a financial home. It's vibe finances. Because you have all these other products, I assume it's going to vibe set up my kids' 529 account, which is not so bad. Here's the areas that I'm most interested in. All right, Vlad, I'm super excited to be here. Thanks for doing this with me today. Well, glad to have you. Okay, so I wanted to start by getting your historical lay of the land of online brokerage. And maybe we don't have to go all the way back, but if you could just go back to maybe the early days of online stock trading into Vanguard and Schwab and Fidelity. And now you've got online platforms for trading that are modern, like Robinhood. What's the history and the narrative as you see it? There's a great book about this, A Piece of the Action by Nocera. It covers the modern, the rise of the modern financial industry, which is super interesting. It's a story of a big wave of democratization associated with lower costs. So Charles Schwab began in the 70s. I think there's a lot of similarities if you look back between what Charles Schwab was to begin with and what Robinhood is sometimes accused of being, right? Before Schwab, Merrill Lynch was the big broker. And really, they had brokers that would sell you trades and call you and convince you to buy stuff. And they would charge hundreds of dollars per trade. And then after that, there was the event that led to the creation of Charles Schwab. It was called Mayday. I forget the exact date, but it was in 1972. And up until that point, commissions were regulated. So you could not charge under a certain amount per trade for commission. And Mayday led to deregulation of trading commissions, which allowed Schwab to enter the business. And Schwab said, you know what? We're going to cut costs. We're going to make it as efficient as possible. So we're not going to have branch offices. You're going to call us on the phone. We're not going to try to sell you anything. We'll just process your ticket over the phone, make your trade, and we'll do it for, I don't know what it was, call it $75 per trade. And so they were in the news in the early days for creating this extremely sophisticated phone dialing system that could handle lots of simultaneous, they innovated using the tools they had at the time. And then in the 1980s, the Apple II came out, right? And actually in Palo Alto, not very far from here, two guys, Bill Porter and I forget his partner's name, Randy or something, they met at a party and Bill Porter had just bought an Apple II computer. And he had this idea of what could I, what if I use this machine to trade stocks from my home? And the two of them got very excited about this idea and they decided to make a business out of it. And E-Trade was born. And of course, that was the origins of it. It went public 10 years later during the dot-com boom. And it became one of the first profitable dot-coms, right? For a long time, they were accused. The whole internet thing was dismissed as a fad. Nobody thought that there would be a way for many of these companies to make money. But online brokerage was, I think, one of the first, if not the first, proven business model that could be profitable in the internet era. So that was E-Trade. And you probably remember, maybe when you were a kid, they were heavy on marketing. They had that baby. It was a cultural moment. The dot-com boom was when I started trading myself. I opened my first brokerage account. And then Robinhood started, what would that have been? 2015 was when we launched. 2012, 2013 was when I got the idea. And our big innovation, I think there were three innovations that led to Robinhood. First was obviously mobile. Everyone else had ignored mobile. And we made the bet that people were saying mobile's a toy. Nobody would do serious financial transactions on their smartphone, but that would become primary. And so we built for that when everyone else was ignoring it. The second bet we made was that high-frequency trading had completely changed institutional finance on an infrastructure level. So entire rooms that would have been full of trading desks, people picking up phones and making trades, were being disrupted by five to 10 kids out of MIT creating HFT firms. And so we took the technology from a sophisticated high-frequency trading firm, and that became the backbone for our retail product, which allowed us to lower costs and offer commission-free trading. And at that time, commissions were still between seven to $10. So that was a huge disruption. And the third thing, the company was formed in the wake of the global financial crisis and the whole Occupy Wall Street movement. And if you think back on the global financial crisis, I had graduated Stanford in 2008 and gone to grad school. My first month in grad school, Lehman Brothers went under. It was an interesting thing because a lot of my friends, the ones that were most secure in their financial careers, just all of a sudden, as quickly as they came into their jobs, and they had their internship at Lehman Brothers, they got their return offer, they were very proud, they had three years of job security in front of them, and then they were going to go get an MBA. There was this well-defined path. And then it evaporated. Evaporated, and you had the images of them packing up their cubicles with boxes. There was a lot of disillusionment, particularly among young people, among millennials with the financial system, because some of this happened to them, they saw it happen to their parents. It felt like they were the victims, young people in particular, of forces beyond their understanding that were packaging products and taking risk, making things complicated, and that there was no accountability. No one went to jail as a result of crimes committed in the global financial crisis. Nobody big anyway. Yeah. And I think this really, this led to the Occupy Wall Street movement, which was really a young person, a millennial thing. And I think it opened the door to a new brand in the space. No love was lost between young people and the big financial companies. And Robinhood, along with mobile, along with the great technology that enabled commission-free trading, I think hit a particular resonance because it was a very optimistic message in a way. It was the anti-Occupy, whereas Occupy said, the system doesn't work. We just have to reset and start it all over, which I think was not really actionable. And that's why the whole movement ended up fizzling out without much impact. And we were saying, why don't we plug you into the system? If we get more and more people to become investors, to own the best companies, and we want to make it easy and engaging and actually enjoyable to do so, then we could solve a lot of these same problems. And I think at the time, there was a need for a new brand. And the story of Robinhood resonated just as much as the product and the technology. When I think about Schwab, Fidelity, Vanguard, for some reason, I feel like those brands were always not deeply connecting with the younger generation, or they weren't maybe going out of their way to be as loud of a brand as they could have been. Yeah. When I think about the big banks, there was definitely a lot of animosity between the population and the big banks. For whatever reason, I feel like there was neither animosity nor love towards those brokerages. Did you experience it that way? Or is that maybe just I wasn't paying attention? I think a lot of people lump the brokerages and the big banks into one thing. You think they can get lumped into one? I think so. Yeah. And not even those guys, but you look at the market makers like Citadel. Yeah, for sure. You saw this during the GameStop era. There's just an association that they're these puppet masters that are moving things around in the dark and nobody understands. And I think trust in general tends to be quite low. Yeah. Now, that was at the time. Now, I think Gen Z and Gen Alpha, there's almost this opposite thing happening where the old, big, storied incumbents are cool again. Interesting. And so we're trying to figure out how to navigate this scenario. What is that? I didn't know. Can you share more about that? I think there's a broader trend of things that are old and maybe that your grandparents and the big banks into one thing. You think they can get lumped into one? I think so. Yeah. And not even those guys, but you look at the market makers like Citadel. Yeah, for sure. You saw this during the GameStop era. There's just an association that they're like these puppet masters that are moving things around in the dark and nobody understands. And I think trust in general tends to be quite low. Yeah. Now, that was at the time. Now, I think Gen Z and Gen Alpha, there's almost this opposite thing happening where the old, big, storied incumbents are cool again. Interesting. And so we're trying to figure out how to navigate this scenario. What is that? I didn't know. Can you share more about that? I think there's a broader trend of things that are old and maybe that your grandparents would use being cool again. So Gen Zs are really into buying vinyl, and cassette tapes are selling again. I like vinyl. 10 years ago, maybe it would have been very hard to get a cassette tape. But now people are buying them. And so my daughter asked me if she could have a Walkman. And I think in the same way, financially, the younger generation is interested in retirement. You and I are probably close to the same age. When I was graduating from college in 2008, I wasn't really thinking about retirement that much. Yeah, I was 2011. Yeah. Retirement seemed really, really far away. Now Gen Zs are opening retirement accounts at 19 years old. So they're thinking a little bit more conservatively, I think, than prior generations. I'm not a cultural historian, even though I try to be. It's important context because I assume that for a business like yours, I would think that capturing the hearts and minds of people between 18 and 30 must be super important for the long term. Yeah. And there's a lot of counterintuitive things there. For example, for older people, I think we've had a lot of success. You would think with older customers, we would emphasize how stable and how long we've been around and all these things. But actually, that resonates very strongly with young people. And older people want to be that, you could be plugged into the cool new thing. Yeah, it's innovative, easy to use. We have all these features. Is that new for you, or was it this way for you all along? I think it's relatively new for us. You had to invert those seven years ago or something? Yeah, I think we have to constantly change the way we think about our products, communicate our products, what we build. And I think the trap that we would like to avoid is, to some degree, these companies that came before us, Schwab, E-Trade, they're stuck in a generation. E-Trade was very popular with Gen X. It became a Gen X broker. Schwab, average customers in their 60s, maybe 70s now. So it's very much the broker of the boomer generation. But what we try very hard to do is always pay attention to the next generation while also moving up market in the sense of being able to serve older customers, people that are in their 60s and 70s, because they have needs and they also have a lot of assets that we could benefit from. So we've been trying to do both over the last 10 years with, I think, some success. How do you think about, if at all, index fund-type investing versus a more hands-on active style of trading? And I'm asking because when I think about Vanguard or Schwab, I think of them as pioneers and champions of the passive investment mindset. Absolutely. And when I think of Robinhood, I think of you can trade for free, and so therefore you should trade. Yeah. How do you think of those things? Does it matter? People should do both? Do you strongly advocate for one or another for young people? When I first started raising capital for Robinhood, there was this perception that ETFs and index investing had eaten the active trading market and that there was no growth in trading and there was all this growth in ETFs, and that was just going to continue and young people wouldn't want to trade. They just wanted to buy ETFs. And I think that that's not exactly true. And I think we've demonstrated that, but it's not a graduation. A lot of people ask, okay, well, do you first start out trading individual stocks and eventually graduate to buying and holding ETFs? What we've seen is it's not that people go from being a trader to being an investor. As their money grows, they just have more and more buckets, right? They have your retirement, long-term money, which for many people is passively managed, but they always want to have a little bit, at least, that they're making decisions on. And so we've tried to conform to that. It used to be that there was only one account in Robinhood. And I think one of the big unlocks for us in the past four years, I'd say, is figuring out how to actually conform to people's mental accounting of their finances. And now you can have over 10 accounts and many different account types. Do you think of your role as just give people the tools to do whatever they think is best? Or do you think of your role as we're going to help create certain healthy patterns and pathways for people to do the right thing at the right time? Yeah. So at the top level, we are aligned. We make money generally proportional to our total assets under custody. Do you want your customers to do well? We want our customers to do well. We don't want them to send their accounts to zero. The best behavior for us is if someone's account balance with us just monotonically and continuously grows over time. I think the best way to do that is to actually conform to how customers like to use financial products. And this has pushed us to, we don't really think of ourselves as a trading app anymore. The way to describe what we are and what we're increasingly becoming is a financial super app. We want to be your primary and your secondary financial account. So we have Robinhood Banking, which is rolling out right now. I think it's the best banking product on the market. And if we serve customers well there, their direct deposit comes into Robinhood. And then once we've got the direct deposit, it's sort of like our assets to lose. And we look at all the ways in which customers take money off the platform. We ask ourselves, is that something that should be happening on platform? And I think that's what's pushed us into this multi-product line conglomerate. It's like a financial conglomerate. Across some of our businesses, we do have a fiduciary responsibility where the leaders of those businesses, for example, Robinhood Strategies, they spend day and night thinking about how they can make as much money for you as possible. Other businesses are more about selection. For example, prediction markets. What do customers want to trade? How can we make sure we have the highest market share relative to our peers? Yeah, actually, this is a good moment to transition to that. Robinhood has a crazy number of big businesses. I think the stat I heard was 11 businesses at over 100 million of revenue. So could you just talk about, at a high level, what the portfolio of businesses are? We organize it into three arcs. So one of those arcs is what we call number one in active traders. So we want to be the place where, if you're an active trader, you would feel like you're at a disadvantage using any other platform. And that's where you have our options business, our crypto trading businesses, Robinhood Legend, which is a new product that we have for active traders on web. I was talking to you a little bit about how there needs to be a Cursor for web, if that makes sense, an AI-driven active trading home for the people that are extremely sophisticated. That's what Robinhood Legend is. And that's where our prediction markets business is as well. It's like, what are the things where we can track our market share? We're, for now, sticking to regulated financial products. So that's kind of the organizing principle. And we look at our market share and we want it to grow over time. The second bucket is just, we call it number one in wallet share. All the things that lead to all of your finances being on Robinhood. The credit card, our Robinhood Gold program, which is kind of the subscription wrapper that gives you a better deal on everything. We have retirement, which is now 25 billion in assets. We've made some acquisitions there: TradePMR, which is, you can actually get a human advisor, which will eventually grow into hopefully your point person for your family office, for those that have one. a AI driven active trading home for the people that are extremely sophisticated. That's what Robinhood Legend is. And that's where our prediction markets business is as well. It's, what are the things where we can track our market share? We're for now sticking to regulated financial products. So that's the organizing principle. And we look at our market share and we want it to grow over time. The second bucket is just, we call it number one in wallet share. All the things that lead to all of your finances being on Robinhood. The credit card, our Robinhood Gold program, which is the subscription wrapper that gives you a better deal on everything. We have retirement, which is now 25 billion in assets. We've made some acquisitions there to trade PMR, which is, you can actually get a human advisor, which will eventually grow into hopefully your point person for your family office for those that have one. But the end game there is how can we get all of your money in Robinhood and how can we get everyone to be a Gold subscriber so that you just get incredible value from everything. And then the third bucket is we call it number one global financial ecosystem, which is us growing our business across two linearly independent vectors. One is from retail only to business and institutional. The other is from US only or US primary to fully global. And we think, in 10 years, over half of our business can be ex-US, and another way, over half of our business can be non-retail. So I think there's multiple vectors by which we can grow at 10X, and we're trying to go after all of them. Can we talk about prediction markets a little bit? I think that's been a very interesting new category that has grown crazily. Yeah. And obviously that's been a big success for you all. What has made prediction markets take off in the way that they have, do you think? This is a funny one because there's actually a very simple answer, and it's just one thing. The big bang of prediction markets was the presidential election last year. In my opinion, I've been a student of prediction markets for many years. I've just loved them. And I remember in 2016 with the presidential election, I was voraciously looking at the prediction markets, trying to figure out who was going to win this thing. And at that point, there was nothing in the US. It was basically Betfair, which is a UK-based exchange that was pricing the odds in real time. And you could tell you're watching the news. They didn't have an answer on that election until the next morning, when Clinton conceded. At 7 p.m., 7:30 Pacific at night, Betfair showed Donald Trump had a 95% chance of winning. So that was, I think, the first instance where you could tell that this was a valuable new forecasting tool and source of information. And then in 2020, it was the same thing. PredictIt was probably the main platform, which was run as a research experiment. But I remember everyone, they had this nice electoral map. They had prediction markets, not just for the election as a whole, but for each state. And their website crashed from so much use on election night. And for the longest time, having prediction markets for the US elections was impermissible. The CFTC said no. And then there was a Supreme Court case that got resolved literally a month before the election last year that said, actually, it's okay. You can roll out federally regulated CFTC presidential election markets. And that surprised everyone. Nobody thought that this result would shake out this way. So credit to Calci for actually taking on that fight and going to the Supreme Court and arguing for their business. When we saw that, we basically mobilized our entire company, and we were like, we have two weeks to get this ready. We've got to be ready for the presidential election. And that week, it was a very high degree of difficulty because we began integrating with Calci for that. And then Calci didn't get the approval to allow other brokers to connect. So we integrated with ForecastX, which is another DCM. But our team sprinted toward this thing. Then we had to change vendors, so we had to sprint again, and we shipped it with about a week to go before the presidential election. And we still saw, I think, over half a billion contracts traded, which is a big business for us. And then what happened was essentially the same argument that you could make for why the presidential election has a lot of economic value could be adapted to the Super Bowl, like a big sporting event. Clearly a Super Bowl, huge event, lots of economic value. And then you could imagine, if the Super Bowl has economic value, every football game has economic value as well because the Super Bowl is just a function of what happens the rest of the season. And it allows you to make those trades more granularly. And that has essentially unlocked the sports prediction markets industry, which has been a big thing. Which is enormous, right? Yeah. And if you look at it, it's a disruptive force to traditional sports betting because sports betting is state-by-state regulated. It's a patchwork. Taxes are quite high. You have this annoying effect where, if you're using one of these sports betting apps in New York, you get shut out. Yeah. And you have to drive across the bridge to New Jersey, and maybe there you have different products and services. Brick-and-mortar rules intended for brick-and-mortar casinos have been applied to digital platforms, which I think makes a little bit less sense in this case. So it's been a disruptive force. And for Robinhood, if you look on a contracts-traded basis, it's been doubling quarter after quarter. Wow. And in October, which is just one month, we did more than all of Q3 put together. And I think that's a function of more contracts, not just in sports, but diversity across the board. We have really interesting AI contracts too, where you can trade what's going to be the best AI model at the end of the year, what's going to be the AI coding model that wins. It's just more people finding out about it, figuring out what prediction markets are. So it's getting more adoption within our user base. We're also getting new customers coming in. But yeah, I think at least definitely on the retail side, we're the only broker that offers prediction markets to a significant degree right now. One of the things I'm very proud of is how quickly we jumped on it. There's some of our competitors that have been integrating and trying to do this for years, and they still haven't shipped. I think it's not just a trading product. This is a unique one where there's actually a huge use case almost as a media forecasting tool. I did a tweet storm about this a couple months ago. I said, prediction markets, a way to think about them is they're truth machines. We're bombarded by all this information constantly. Anyone can write anything on Twitter. It can go viral. How do you sift through what's real and what's not and what's actually going to happen? Well, now we've created a tool that lets you do that. And I think the benefit is not just for the folks that are trading. It's almost like the trading and speculating is doing the work so that all of us have this reliable source of information. I should have looked this up before I came here, but I feel like I remember seeing that if you ask a million people to guess the weight of a cow, or something ridiculous like that, it's actually very close. The middle is actually very close, and I feel like that is basically what prediction markets are doing, right? The wisdom of crowds betting with skin in the game leads to truth. I think the skin in the game is the big distinguishing feature rather than just a poll. You can run a poll, but yeah, it's different because it's a price. This is a price. It's not a poll. It's not a guess. It's from people with real skin in the game, and I think that, there's empirical evidence that they're super accurate even one month before the event in question. To what extent do you think of prediction markets as economic value via this truthy outcome from the wisdom of crowds with skin in the game versus just a cool form of entertainment for people? Yeah, we get asked this all the time. And neither is that, by the way, I love poker, for example. Isn't it just gambling? But the thing is, every tradable asset has had that criticism over time, right? Futures contracts, where people have been speculating over the price of oil or the price of gold. This was a huge fight back in the, I think, '60s, where a lot of people were on both sides saying, retail shouldn't have access to these products because it's just speculation. But I think, yeah, another word for gambling is speculation, right? Well, but also, there's nothing wrong with gambling, really. Put aside that, and I agree. of prediction markets as economic value via this truthy outcome from the wisdom of crowds with skin in the game versus just a cool form of entertainment for people? Yeah, we get asked this all the time. And neither is that, by the way, I love poker, for example. Isn't it just gambling? But the thing is, every tradable asset has had that criticism over time, right? Futures contracts where people have been speculating over the price of oil or the price of gold. This was a huge fight back in the, I think, 60s where a lot of people were on both sides saying, retail shouldn't have access to these products because it's just speculation. But I think, yeah, another word for gambling is speculation, right? Well, but also, there's nothing wrong with gambling, really. Put aside that, and I agree. Without speculation, you can't have a functional financial market. So we need the speculators. Otherwise, things just don't work. You need people that have a view on what's going to happen in the future to create the market. And I think most people would agree we need the market. But that said, I don't think everyone should be speculating with 100% of their money. That's an extreme case. I have no, I'm just thinking out loud here for fun. But I suppose maybe the counter there would be it's useful to get price discovery on an equity asset or something like that. Yeah. But there's not particular value to getting price discovery on the outcome of the Ravens game or something. I imagine there'd be some counter like that. Yeah. And it's definitely the case that that has value because any forecast has value. And politics is a good example where there's obviously value. Well, just think about how much money is there in informing people and commenting on what's going to happen to that Ravens game. Totally. People buy the newspaper to read the sports section and see the commentary. You're listening to SportsCenter. And I think that's just going to become a bigger and bigger part of the economy. Let's say we get to this future where AI is automating more and more jobs, right? One of the things that probably will be difficult to automate is entertainment and sports with real humans. So I would make a bet, actually, that this just continues to grow. And, the lion's share of jobs and job families that we consider in the future probably look to us like some form of entertainment today. Totally. One of the things that I asked Dylan Field about on the podcast the other week that I think maybe relates to prediction markets, it relates to some of the things we've been talking about, is I have this sense that I think to some degree with today's young people, but I think it's actually with all of us. There's some growing degree of financial aggression, or the negative way to say it might be gambliness, a positive way to say it might be risk-taking. Yeah. I think it shows up in a lot of ways. Obviously, prediction markets are an easy thing to point to. I also gave the example when I was talking to Dylan about collectible cards and NFTs, and you have these small things that people start speculating on like crazy. I think a lot of behaviors that people have, just in general, have gone the way of a lot more risk, I would say. And I don't think that's a bad thing, but to me it seems like a thing. And I'm just curious if it's something that you've noticed, if it's something you've thought about at all, how you think Robinhood's suite of products can channel that in positive ways. I don't know, I'm just curious how you think about it. Yeah, I actually think that it's not a huge fundamental change in people's mindset, but more what they have access to. So, for example, it used to be very difficult and cost-prohibitive to invest in individual stocks. I think there's a big reason why ETFs had extreme product-market fit, because before the ETF came along, if you want to be invested in 500 companies, you're not going to buy 500 individual stocks. You could get that diversification quite simply and easily. We're on a trajectory where people want more granular ways to express their opinion about things. So one example I can give you is how many times have you seen a company just blow out their earnings estimates? They blew out EPS, they blew out revenue, but the stock goes down. Yeah. And folks get frustrated because they're like, I'm a student of this company, I figured out, I have a pretty good model of how well they're going to do, but the stock might not always be a perfect representation of that. So then you have prediction markets for EPS and revenue where, if you have a point of view of how something's going to shake out, you can actually trade that more directly. Options trading, I think, is part of this trend too, where maybe you do want to hold a company for a very long time, but you have a view that it's going to do well this quarter. So you can design a contract and trade it profitably based on that viewpoint. Then you have zero-day contracts, which take that to the extreme, which is, I have a pretty good idea of how something's going to do today, and maybe it's an earnings day and you don't want to take the time decay risk. So, I think in general, even though to some degree these are called more complex products, there's a certain simplicity in that so many things drive the value of a stock, but these contracts let you express a more granular point of view. And I think the other thing is they just weren't available easily. You would have had to be very, very sophisticated to have unique access to them in the past. And now, through tools like Robinhood, we make them available to traders with costs that are really low. And then I think the third thing is you hear a lot about it in the news predominantly because it's more interesting than buy-and-hold passive investing. Everyone wants to write about options trading and prediction markets and crypto because it's speculative. You have big swings. Nobody's interested in writing the story about how index funds are gaining AUM or money market funds are seeing huge inflows, even though that's also happening. So I think it's the combination of these three things that makes it seem like there's a huge increase in speculation. But I think on a relative basis, the story is much more tempered. I mean, it's probably also true what you're saying, where if in the year 1950 you gave people a smartphone with an app that let them bet on the baseball game they were watching, a lot of people would have bet. And so I think the tooling probably is just expressing human interest to do various things. I guess, if they respect, there has always been speculation. I mean, you look back to Isaac Newton. There's just been a lot of friction in the past, though. Yes. The EPS example is really good because I think a lot of times people work really hard to take a certain view on a company and have confidence in it, and then you bet on a proxy, which is share price goes up. Yeah. It's a good example. Totally. There was a trend over time that I do think is harder for young people. It does seem like, compared to, and maybe this is a not truthful nostalgia, but I think it probably seems to be the case that our parents' generation had an easier time buying a first home than today's recent college grad or something like that. So I do wonder if there is a long-term trend that does make it harder to get on the financial ramp for people out of school, whether it's to do with home prices going up on a relative basis or something to do with inflation or I don't know what else, but it does seem like there is a long-term directional trend where it's harder for young people to get into the place in their financial life that they want to get to at a young age. I think that's true. I think the social safety net has frayed a little bit. You have homeownership becoming a little bit more difficult, a lot more difficult in the Bay Area, perhaps. Yeah. You also have the student loan burden. Which is a disaster. Yeah, the cost of education has just continued to inflect. Yeah, it's gone nuts. I went to Stanford, and when I was there the tuition plus room and board was 40 to 50, right? And now I think it's like 80 to 90, which is insane. And state schools have encroached upon the 40 to 50 level for in-state tuition inflation or I don't know what else, but it does seem there is a long-term directional trend where it's harder for young people to get into the place in their financial life that they want to get to at a young age. I think that's true. I think the social safety net has frayed a little bit. You have homeownership becoming a little bit more difficult, a lot more difficult in the Bay Area, perhaps. Yeah. You also have the student loan burden, which is a disaster. Yeah, the cost of education has just continued to inflect. Yeah, it's gone nuts. I went to Stanford, and when I was there, the tuition plus room and board was 40 to 50, right? And now I think it's 80 to 90, which is insane. And state schools have encroached upon the 40 to 50 level for in-state tuition in some cases. So, yeah, I think there's a lot of problems. It's probably a lot of these common goods have just gotten hard to get. There's a lot of that. That's a huge problem for sure. I would think that would be a big opportunity for you, though, to engage with people young and find some way to give products that can help them get there. Well, what's gone the other way? Trading commissions. Yeah. In our generation, trading commissions were $100 per trade when I was born, and now there's zero. So, when you hitch it onto the technology wagon, yeah, things get taken to zero pretty quickly. Another thing along these lines I'm really interested in is now, as a, I suppose, as a venture capitalist, a lot of value gets created in private markets. And I think as companies, you're public, but a lot of peer companies that are of similar scale are private, and a lot of value compounds in private markets. And I think that's becoming more true every year. And so maybe this touches on the tokenization of everything type of idea, where one way that you could help bridge some of these gaps would be to give people access to some of these hard-to-access equities. Yeah. This is something that I'm personally very passionate about. I think that it's one of the biggest inequities in capital markets. It wasn't too long ago. A few years before we were born, Microsoft and Apple went public at valuations in the hundreds of millions. Right. Right. Or maybe low single-digit billions. And then 99.9% of their value is created in the public markets. Yeah. And you still see examples of that. Right now, you're seeing a lot of companies that are worth hundreds of billions at the frontier of their industries, not just OpenAI and Anthropoc, but also SpaceX, which is leading the space revolution. Yeah. Before they're available to retail, it's not hard to imagine that they're going to be in the trillions. Right. And then what kind of appreciation, to get a Microsoft IPO till now appreciation, they're going to have to get to a quadrillion of value. It's tough. It's going to be tough. Without hyperinflation. Right. So, yeah, I'm very motivated. I think ex-US, we're pushing on tokenization, which I think is probably something like the end state. A lot of people have to get comfortable with the implications of it, and it's really going to be disruptive. Can you explain how it would work or works? At the fundamental level, it's the same idea as what stablecoins are. You have some traditional financial stuff that you put in a bucket, and in stablecoins, it's US dollars and Treasury notes and bonds, and you mint and burn tokens against that. The traditional stuff stays in the bucket, and the tokens are freely tradable on blockchains. So, you can imagine 24/7 liquid trading around the clock, just like a stock. And the only time tokens change hands with shares is in a mint-and-burn scenario. So, another analog is you imagine it's kind of like an ETF or an ADR. You have a bunch of stuff here, like S&P 500. An ETF warehouses the underlying stocks. The ETF is traded freely, and then at some point, very rarely, you have a mechanism that exchanges the underlying for the ETF, which is more of an institutional product. But yeah, you can extend this concept with blockchain technology to anything. So, you can put private stocks or SPV LP interests in the bucket. Do you think that's where things go in the US over time? I think so. Yeah. I think that it's really hard to imagine a cleaner solution. I think a lot of people talk about on-chain issuance. I don't think the end user gives a shit about on-chain issuance or really any of this. The end user just wants economic exposure to things in a way that works. And I think some mechanism of this, and there's a lot of questions to be answered, right? So, currently in Europe this structure is a derivative, but this is more of a paperwork thing, right? The underlying mechanism could stay the same, and the regulatory description of it and what it is and whether you get ownership or voting rights, that could all change. So, I think there's going to be questions of, is it a derivative? Is it a tokenized stock? Is it a new thing? Do you get to vote your shares? But we'll be able to make it at least as good, from an experience standpoint, as the existing things that are out there. Obviously, everybody talks about AI now. I'm curious how much of an impact it has been for you to date and where you think it's going to be the most impactful in the future. A lot of people speak about this in generalities, don't they? I think that... Well, if you just speak in generalities, then everybody just, you know, value goes up, so you just say AI and it's good. Yeah, I like to measure things. I think we're very... Yeah, we pick the areas internally that we thought AI would make the biggest difference for us, and then we spent a lot of time trying to... We continually spend time thinking about how to measure it and how the progress should go. Did you have any guesses that you thought would matter and they didn't? And then were there any where you were surprised in the other direction? The two areas where we really invested big time in were customer support and engineering, which I think were the two most... If you had to think about where human capital really matters, writing code determines how fast we ship products and fix bugs and improve infrastructure. Customer support is the frontline interface with our system. Yeah, I think outside of chat, those are probably the biggest two AI markets so far, so that makes sense. Yeah, and so customer support, we measure AI deflection rate. It's worked super well. Worked super well. And with deflection rate, which you probably can understand, but I don't know if most viewers would, you kind of look at it's a measurement of what ticket would have gone to a human but instead has been fully solved by AI. What'd you go to? I don't know if we've announced that publicly. But something good. It's relatively, it's very high. I think it's the best in the industry, actually. And we do all this work in-house. We don't use vendors. Have you measured if it doesn't just save costs by not using a human, but have you measured if there's also higher satisfaction on the way the ticket was handled? Is it both a better experience and cheaper? We do measure that. And I think the story is complex. And the reason is there's some percentage of people, call it 10 to 15%, that just don't like talking to an AI agent. They have a predisposition to not like that. We actually track that number, and that's going down over time too, which is very good. Yeah. And then we have to do all this work to make the agent try to convince them, right? It says, actually, I'm very helpful. Give me a try. The wait might be quite long. Shut up, agent. Yeah, that's good. Yeah, it's getting better and better. And there was an incident a couple of weeks ago where our AI customer support agents actually went offline for a few hours. Immediately, the phone started ringing like crazy. So then we have clear, direct evidence that the deflections are actually real. We haven't talked much publicly about progress there, but we're doing an AI event in a couple of weeks, and I think that one will be good. And then engineering, we look at, obviously, number of lines of code that are contributed by AI. And we also look at overall, commits per engineer per month. And both, basically, have been going up and to the right. And I think for those areas, those are probably the important things to track if you're an entrepreneur, if you're looking for metrics. And I think they're controversial because the engineers will say, oh, well. Yeah, that's good. Yeah, it's getting better and better. And there was an incident a couple of weeks ago where our AI customer support agents actually went offline for a few hours. Immediately, the phone started ringing like crazy. So then we have clear, direct evidence that the deflections are actually real. We haven't talked much publicly about progress there, but we're doing an AI event in a couple of weeks, and I think that one will be good. And then engineering, we look at a number of lines of code that are contributed by AI. And we also look at overall commits per engineer per month. And both have been going up and to the right. And I think for those areas, those are probably the important things to track if you're an entrepreneur. If you're looking for metrics, I think they're controversial because the engineers will say, "Oh, well. It's not about the lines of code, it's about the quality of each line." And you're like, "Yeah. Yeah, exactly. But let's look at the data and just investigate more deeply, right?" I remember we found that it was not just, we did a core sampling on this. We found that engineers who contributed more lines of code also contributed higher quality per line of code. We've seen the same thing. Yeah, yeah. The two aren't as contradictory as you might think. And then I think the next one is just marketing and creative output. You should measure the total throughput of marketing creative, and also the lion's share of that should be generated by AI. I think the tools at this point are good enough to where you can actually move the needle in a big way. I feel like the cursor idea you mentioned should be very interesting. If at some point you get to a place where you trust Robinhood enough to just tell it, here's what I'm looking for both in the short term and the long term. Here's my risk appetite. Here's the products I want. Just go make it all happen. We call it vibe trading. So there's vibe coding. I think there will be vibe trading, which sounds a little flippant. Well, but the thing you said, that we're not a trading platform, it's a financial home. It's like vibe finances because you have all these other products. I assume it's going to, vibe set up my kids' 529 account, which is not totally. Yeah. Here's the areas that I'm most interested in. So I think vibe trading and going down that path of helping first active traders, but eventually seeping into more general use cases with indicator analysis. Right now, these guys are writing code in languages that you and I have never heard of to write these custom indicators and different scanners. So we can turn that to English. And we announced that through our Robinhood Cortex for Legend product at our active trader event a few weeks ago. We have stock digest and crypto digest in app, which actually are really cool, incredible feedback from users so far. The way it works is you get a notification saying some stock or crypto is up 5%. You immediately want to find out what's going on. And so we actually give you a digest in the product that explains, and it's accurate up to seconds or low single-digit minutes. And that's been really good. So you could imagine AI unobtrusively being this assistant and, in some cases, autonomous financial agent for many of these things. I also think that there's just some really, really boring stuff that's very hard right now that we're going to push to fully automate. So moving your bank account. I don't know if you've ever moved a bank account, but it's a pain in the butt. You have all sorts of bills that are hooked into it, all sorts of subscriptions. Totally. And I think at some point you'll just press a button and that bank account will move. And the first companies that figure out how to do this are going to have a pretty big advantage because it's considered a very sticky thing precisely for this reason. Nobody wants to go to the effort of changing their credit card number or their ACH information for 20 to 30 different things. Yeah, you don't even know what you forgot. Yeah, you don't want to even think about it. And yeah, that's why it doesn't happen. I think if we can, if it becomes a button, I think it's headed there, right? Then it's going to be very disruptive to a lot of incumbent firms. Moving brokerage accounts, same idea, very complicated. They don't want to make it easy either. Your counterparties are not interested in having a streamlined integration. Totally. But I guess an agent that read all of your documents from the old brokerage could figure out everything that you needed to know, I suppose. Yeah. Yeah. Yeah. It could be, if you have a person doing this on your behalf and they make the calls, they talk to customer support, they have access to your emails. Yeah. There's no reason why an AI agent won't be able to do all that work for you. Yeah. Maybe just to finish up, I guess you've been doing this for over 10 years now. And obviously, over a $100 billion public company, but you still have, I know, an intensity, a drive to keep going. So I guess one thing I'm curious about is where have you found the drive to stay hardcore, detail-focused, founder through both a lot of years, but also you've gotten to a level of success where it would be easy to not fight so hard. But obviously, you're pushing the boundary on so many things at once. I think that the one amazing thing that you have as a founder of a company that I don't think is as easy if you're an employee or even a senior executive is ultimately you have a lot of control, right? So if there's something about your company that doesn't work well or that you don't like or you find yourself having to say things to the employees that you don't agree with, I think that's a good opportunity to reset and to change something. You either have to change something about yourself or about your policies or what you're saying. I think I learned this during COVID, which was a hard time for us for many reasons, but also for me personally. Some of the folks in the company, maybe also the press and what you're hearing from the internet, pushed me in directions where I was just saying things that I didn't really believe in. Like, "Oh, we really care about having the best health and wellness benefits so that our employees can feel good, and you can bring your whole self to work." And I noticed that I was saying things that I didn't really believe in. And I think that's the source of ultimately a lot of unhappiness. And at the end of the day, as a founder, I think the company values have to be your values. The way you run the company has to be the way