Invest Like the Best with Patrick O'Shaughnessy
· William Hockey
[00:00:00] S1: One of the things that I think about – AI is here. We're probably much closer to ASI than people think. And it's gonna be like a tidal wave through the economy. You have to think about what are the implications for your business. And I think in my perspective – if you're not a researcher, if you're not a lab – how do you play? I actually don't think "AI companies" are very set up for success. That's actually probably not where value is going to accrue. The value is going to accrue in two areas. Most important area I think, do you have massive distribution? Because if you have massive distribution, if you have massive costs, AI is going to be a massive boon for you. And so the thing I think a lot about is, how do you think about – not just from a software perspective, but from a distribution perspective and a brand perspective, what you would best capture to utilize AI? Because I think people talk a lot about the railroads. Yes, the value accrued to the railroads. Yes, the value accrued to the ISPs. Yes, the value accrued to the people like building mobile phones, but the value actually accrued to the people that could harness that the best. The value in railroads accrued to the oil companies. That's a little bit hard to fundamentally understand, that Standard Oil was the biggest beneficiary, biggest boon for railroads by an order of magnitude. What is the equivalent area for AI? And I sometimes don't think we're focused on that enough. Hot take here, I think like the biggest, fattest, most inefficient brands are going to be the best beneficiary of AI, because brands have a massive moat and man, there's a lot of costs to cut there.
· Patrick O'Shaughnessy
[00:00:00] S0: How do you think AI will most affect financial services specifically? Maybe even just like your own products and services?
· William Hockey
[00:00:00] S1: Financial services, especially large banks – they suffer from the fact that the business model's too good. Banks are pretty profitable. Financial services, it's like, if you own a bank and you aren't profitable, that's a problem. You've clearly done something wrong. The business model fundamentally is really good. That's made them laggards in a lot of ways. But it also, I think, makes them a massive opportunity, where I think financial services and legacy banks tend to be the largest inefficiencies out there. They also are very hard to take over. There's not a lot of private equity activity in financial services because it's highly regulated. And so I think the large banks that can actually effectively harness that are going to be some of the largest beneficiaries, because banks don't have a lot of physical assets. Places that have massive capex and physical assets, it's hard to tell a good AI story there. How's AI going to make railroads more efficient? If 99% of your money is spent on fuel, track maintenance, then people – the human cost in railroads is de minimis. You take conductors down from 1000 to 500, it doesn't like change the equation at all. But if you think about a traditional bank, it's pretty headcount focused and it's pretty technology focused. That's where a majority of the money goes to. The actual physical branch infrastructure is such a minority of your balance sheet. I do think they will be either the most disrupted or probably actually like the largest beneficiaries. The banks that are like most effectively run, have the largest distribution of largest cost structure are going to be the largest beneficiaries. I think the UX of financial services will change a lot because people think, "Oh my gosh, my bank is hard to use. My bank is hard to move money. It's like slow to move money." That's actually a feature, not a bug. I think what fintech sometimes starts with is they start with, "Hey, we're going to make it like super easy and super free to move money all over the place." Then it actually starts to slow down. The reason it slows down is because fraud is super expensive. If you make it really easy for your grandma to send money to somebody in Nigeria – yes, that's great for remittances, but that's really bad for romance scams. That's really bad for fraud.
· William Hockey, Patrick O'Shaughnessy
But I do think with AI, we can actually build those detection models a lot better and so it's probably less likely your grandma is going to fall prey to elder abuse. And if the banks don't have to optimize so much for that, and it comes a little bit for free, we can actually make the UX better for everybody else. Because we have the technology right now to make financial services almost entirely instant and entirely friction free. All the friction is actually built to protect the 5 to 10% of consumers that can get hurt. People forget that. But I think if we can build models that are just as good as humans at detecting that stuff and that can happen instantly, we can actually take that tail away and it actually is massively beneficial for everybody else. [00:00:00] S0: Do you think it's a good time to be a new entrepreneur in financial services?
· William Hockey
[00:00:00] S1: Anybody who tells you that it's a bad time to be an entrepreneur, that probably means it's a good time to be an entrepreneur. If you look at – there's all these stats on Twitter that I'm sure people have seen where the best companies are created in the worst environments. And I think that's generally true. It is cheaper than ever to be an entrepreneur. It's probably the least risky time to be an entrepreneur. It's also pretty crowded. But, you probably look at the last YC batch, I'm sure 90 plus percent of them were AI related. So yeah, I think it's actually probably a pretty good time to be a founder in a non-AI related place right now, because there's probably less competition, less smart people. If you want to be successful, you can go look at every single industry and you can say, okay, who are the dumbest people and what makes the most money? And if you go attack that area, it's probably a pretty good space. The problem is sometimes Silicon Valley or founders, you look and say like, Hey, we're all like the smartest people. That may be a cool space, but that's probably like the most crowded with the smartest people. And so your ability to compete is like, like your competition's pretty intense. YC puts out this request for startups. My recommendation is like, that should be a list of startups you should not start. Because by the time it's so consensus that this is a good area or super interesting, the amount of capital and the amount of smart people – it's like moth to light. I would almost go the opposite way. Hey YC, I love you guys. Please continue to fund our customers, you guys are amazing. But as a founder I would maybe be a little bit skeptical.
· Patrick O'Shaughnessy
[00:00:00] S0: I think your perspective is so unique and interesting. I always love talking to you. I always find it very inspirational. On the dimensions of just really going your own way, but also just the willingness to fall in love with some part of the world and get devoted to it and, and just outlearn everybody and stick with it. I love your diversification strategy of illiquid Plaid and illiquid Column. I think you know my traditional closing question. What is the kindest thing that anyone's ever done for you?