Robinhood CEO Vlad Tenev 谈股票代币化、扩大非上市公司股份可及性与金融科技未来

I am Vlad Tenev, the founder of Robinhood. We're talking about Robinhood. The stock's more than doubled since its last report. Its stock surged 180% this year after nearly doubling in 2024. The shares of the trading platform are now up more than 400% in the last year. Glad your presence there speaks volumes. Robinhood Gold hit a record 3.5 million subscribers. Most financial services get worse the more money you have, but we wanted to kind of invert that. Ladies and gentlemen, please welcome Robinhood CEO Vlad Tenev. [Music]

My guy.

Good to see you, brother.

Good to see you. You're the reason JCal's here. I mean, it is a great story: JCal bumbles into 8 shares of Robinhood, and Vlad builds a $100 billion company. It's unbelievable.

People know me for the Uber investment at a $4 million or $5 million valuation. But—

When do I flip that?

Well, yeah, it's going to take a little more. I think you have 20x left to go. But wait, was JCal the third or fourth investor in Robinhood?

It was at a $20 million valuation, but it's a good story, I think, because you hadn't launched.

We were at Antonio's Nut House. I went for a drink with my friend Adeo Ressi, and he brought his college roommate, Elon Musk. We were hanging out at Antonio's Nut House in Palo Alto. Rest in peace, Antonio.

Keep dropping these names. One second.

Vlad and I knew each other a little bit. Vlad pitched me on this idea and said, "I'm a quant."

I said, "What's a quant?"

He said, "Quantitative analyst."

I said, "Yeah, I've heard of it. Hit me with the idea."

Then he goes, "Is that Elon Musk?"

I said, "Yeah, just hit me with the idea. I know you've got a startup."

He said, "I want to get this generation—these millennials, these Gen Zs—to trade stocks."

I said, "Love it. They don't care about getting a driver's license. They're still on their mom and dad's Netflix. You're going to try to get people who don't care about the future to trade stocks?"

He said, "Yeah."

I said, "What's the business model?"

He said, "This is the best part. We're going to let them trade for free."

I said, "Okay, let me repeat this back to you, kid. You want to get a group of people who don't have any interest in the future to trade stocks, and then—"

We have 30 seconds.

You're going to make money. You're going to make money. Thank you, Vlad.

I said, "I'm in. I'm in."

Not only that, but he said, "This is probably the best idea you'll ever have."

I did say that to him, too. I was like, "This is the best idea. What if it works?" And here we are 10 years later. What's worked?

And last week you were added to the S&P 500.

I was—

Last week or yesterday?

Two days ago.

It was Friday.

I mean, what a huge accomplishment.

Thank you.

I think it was because I rejected you for a job, right?

You heard about that.

This is a series.

Thank you guys for upgrading me, by the way. I guess that's been the best part of being added to the S&P—going from just a Jason interview to the whole squad.

We were doing the rehearsal yesterday, and everyone wanted to do the interview, so we said, "Let's all do it together."

Why don't we all get in here? Let's maybe start. Look, you've built an incredible business. There's a part of it that looks like what comes after the E*TRADEs of the world, and so on. But there's an enormous other part of your business, and there are all these emergent paths.

I want to start by double-clicking on something that you announced a few months ago in France. Maybe you can talk us through what the goal was. You got a lot of support, but you got a lot of blowback as well. There were a lot of people who said, "Wow, this is a little too disruptive—tokenizing these stocks, putting them on the blockchain." Maybe talk us through the business, and then double-click on that narrow thing so we can understand what you're up to.

We had an event in the south of France that we called To Catch a Token. The idea behind that event was that we wanted to show what Robinhood—the app, the platform—would look like if it were built from the ground up on crypto technology.

What that looked like was stocks on blockchains. Obviously, we added a bunch of crypto-native features, like perps. We launched in 31 countries. We also wanted to demonstrate to the U.S. the power of putting traditional financial services on blockchains.

To me, a lot of people talk about 24/7 stock trading and instant settlement. These things do have real value, but I think the most powerful thing is taking inaccessible, illiquid assets and making them available. We were actually, I think, the first to tokenize OpenAI and SpaceX and make that available to our retail customers in Europe in the form of a giveaway. That was very exciting—not without its controversy, but I felt like it was such a powerful thing.

How do you do it? How did you enable that?

It's actually very similar to a stablecoin, in a way. This is a little oversimplified, but if you're a stablecoin issuer, you can think of it as keeping some dollars or treasuries in a bucket over here. We mint and burn tokens against that bucket, backed 1:1, and the tokens can trade publicly on a variety of blockchains.

It's just extending that tokenization concept from stablecoins to public and private securities.

So you had to go and secure your own block of SpaceX and OpenAI stock and then put it somewhere. Were the companies okay with it?

It depends. I think a lot of people are okay with it in principle, but if you're a company focused on your mission, like OpenAI is, and you hear about some new thing, it's a distraction.

I don't really blame them for tokenization or private access not being their top priority, but I did want to be the first to tokenize OpenAI.

Did Sam give you a call?

I've had a couple of conversations with Sam before and after.

He's a spicy individual. What was his take? Did he tell you to stop?

I'd like to think we get along quite well. I think he understood why we were doing it. Again, the distraction aspect, when they have so much going on, is a real thing. But at the end of the day—

Where do you take it from here? Are you going to go and get 50 or 100 of these well-known private companies? Is that the goal? Is it every private company? What do you do from here to build on top of it?

We've been hard at work trying to figure out how to do it in the U.S. I think that's what everyone's interested in since the France announcement. Obviously, we're expanding what we do in Europe as well, and there will probably be different mechanisms in the U.S. and Europe, at least for some time.

You should expect that we go bigger and deeper into the space and have plenty of things to do in the future.

I'm curious what the relationship has been with the new administration. The last administration was not very pro-innovation or pro-crypto, and now you've got David Sacks running that specifically. How has the change in administration changed how you look at innovation at Robinhood? And then, Sacks, I'm sure you have some follow-up questions here.

It's been very positive. Just by nature of how many times I've been to Washington, the last administration didn't invite me to the White House once. I asked for meetings, and they wouldn't even meet in person. They were all working remotely until, I think, 2023, early 2024.

It was funny how remote work kind of broke down along political lines. It's sort of like the Republicans wanted to get back into the office.

I think you're referring to remote work as not working.

I mean, that is—I wasn't going to say it, but, yeah.

It was funny how that worked out, but it's been very positive. During the last administration, we were playing a lot of defense. It was one enforcement action after another, and we had a Wells notice. All aspects of our business were sort of under assault.

The most direct thing was that, when all of that went away, we had to think, "Okay, well, the administration now wants to work with us rather than just trying to attack us from all these angles." For a while, we didn't even know how to operate in that environment because we were completely unused to it.

Have you met with Elizabeth Warren?

I have not. No, I just receive letters from time to time.

She really—she hates you. She told me personally. She really can't take it.

No, the reason I asked is: What is the core motivation behind the idea that we need to enforce, restrict, or prohibit? Is it consumer protection? Is the belief that systems like yours, which are more open, accessible, and usable, will cause more consumers to trade more and potentially lose money, and therefore they have to play a role in restricting consumer access to these markets and marketplaces?

Is that what they're ultimately driving toward, or do you think there's something more vested-interest-wise motivating them?

I think there's probably both. Certainly, consumer protection is the stated reason, but obviously these folks have funders, backers, and lots of interests.

I mean, there are powerful financial services companies in the state of Massachusetts. So, I don't know what's happening behind the scenes, but I do think the consumer protection angle is what they're pulling out.

How do you look at this type of innovation and your role at the White House to support it and foster it while still having some rules on the field?

David Sacks

Well, I think Vlad's vision around tokenization is very exciting. I like—you wrote an op-ed, I think it's in The Washington Post, that I thought was very good on this topic.

We now have a regulatory framework in place, the GENIUS Act, which the president signed in July. It creates the set of rules for stablecoins, which are just tokenized dollars. Like Vlad's saying, if you can tokenize a dollar, you can tokenize anything. You basically create a reserve of that asset in a secure account at a bank or broker, and then you mint tokens on a one-to-one basis.

I think it's very exciting. There's no reason why we can't tokenize—let's start with public securities. I think that's the easy case, because with public companies, there's already disclosure requirements. There's an abundance of information, and anybody can buy a public security because of those disclosure requirements. The companies don't really care who their stockholders are, because they know that the public owns these securities.

What we could get right away with tokenized public securities is, like you were saying, a 24/7 global marketplace with instantaneous, blockchain-based settlement. That could be really exciting. There's no reason why trading has to be on this 9-to-5 exchange with all this overhead. We could enable stocks to trade as easily as you transfer a stablecoin.

Now, the private securities part is interesting. That is more complicated because, first of all, the companies, like you're saying, do care who their shareholders are, and they generally restrict those things. That's why you probably got the phone call from Sam. The regulators care also because there's not as much public disclosure, so there's more of an impetus to protect the public.

By the way, I'm not saying we can't get there on private securities. About a decade ago, I founded a startup to tokenize real estate called Harbor, and we were just way too far ahead of the curve. That was basically to tokenize private real estate securities. I think we can get there, but I think the place to start that would be really exciting would just be public securities first, because it's easier from a regulatory standpoint, and then we can work our way into private.

Yeah, it's certainly easier technologically. We've made both available to some extent in the EU. I think private could be more meaningful long term, and I'll tell you why I think so.

If you look at the technologies that are transforming society right now, and that we feel so optimistic about over the next 5 years, it's AI and, to some extent, space exploration. With AI in particular, there's a lot of fear right now. You talk to a random person on the street, and more than half the time, they're a little bit nervous about what AI is going to do to them.

Now imagine the scenario if 20% to 30% of someone's net worth is in AI companies. Suddenly, they're not fighting against this thing. They want it to succeed because, if AI succeeds—

Entrepreneurship is a way to let more people participate in the boom.

Yeah. Because I worry about the status quo. These AI companies in particular are getting into valuations of hundreds and hundreds of billions with zero retail ownership. That technology could completely disrupt how normal people live their lives.

We actually expect it to drive that sort of disruption because, if you look at Cathie's presentation, you're talking about negative inflation, high GDP growth rates, and giant productivity improvements. I don't think you're going to get there without some significant labor force disruption.

Okay. So what do you need from the U.S. government broadly, whether it's the SEC or maybe new legislation on Capitol Hill? What exactly do you need to bring about this revolution?

I think relaxation of accreditation standards toward more self-certification. You mentioned a test. I think a test is one form of self-certification, but the simplest form is just someone saying, “I understand the risks. I understand I could lose 100% of what I put in this investment.” You could even put a skull and crossbones.

No crying in the casino.

Yeah, exactly.

You could put somebody crying in the casino.

Yeah.

Literally in the app.

No crying in the casino. I think—

I guess the point you're making is you can't, on the one hand, cry for access and, on the other hand, cry in the casino.

Exactly.

Can't do that.

But the executive order on 401(k) access, I think, was a step in the right direction. We could also ease into it by extending that to individual retirement accounts, which are great short-term vehicles.

You need a security interest. Why can't you just create a synthetic, or a futures contract, saying if and when OpenAI goes public? You can see how many shares there are. You know what the legal registration of the corporation is. Can't you create a synthetic contract that just trades the value of the stock and ultimately needs to settle at some point after the company goes public?

We can't do that currently, and OpenAI in particular is a tricky one. Pick any other LLC or C corporation, right? But we're continuing to look at all angles. I think some clarity would be helpful.

Because this is the whole value of futures markets and prediction markets. Sorry, Chamath, but you can effectively create a synthetic on some underlying without actually having ownership or a security interest in the underlying, or delivery of the commodity. You could basically just say, when this thing goes public, is it above $20 a share or below $20 a share on some number of days after something like that? I guess one question for you is: Is that where prediction markets can take us?

The difference with prediction markets is you can create a prediction market, but it has to have an expiration date for the contract. For example, we have a prediction market live on the platform now about which companies are going to IPO. You could do something like that.

There have been prediction markets in the past—not on our platform, but on other platforms—that make a market around the IPO price. But if you just want exposure to the underlying equity in a private company, I don't think we can do that.

How do you think about the criticism people have had: We have a young generation. They're frisky. They want to take all this risk. They want to bet. And your responsibility as a platform that is giving them that access—if you're the on-ramp, the education you give?

I remember with options, and people being able to short, you came up with an incredibly elegant solution. When you try to short something, you give people a test and education in that moment before they do it. So how do you think broadly about young people getting into wagering? They're playing cards, they're betting on fantasy football and doing sports betting, but they also want to have their hand in crypto and in puts and calls, and pretty sophisticated stuff. What's your responsibility as a platform in introducing them to those sophisticated ways of betting and investing?

I have a lot of thoughts. When you interviewed me for a job in 2008, which, by the way, was one of 2 final job interviews that I got—most people just rejected me. I never even got a call from Google or any of the others. It was either WeatherBill, later The Climate Corporation, or Optiver, where I interviewed to be an options trader. I got very, very close.

We had a math team.

Yeah, I got very close.

Was he in the interview?

He—you probably don't even remember. You remember Alex Machulka?

Yeah, I think he was—

I forget my recruiter's name, but that guy was great. Anyway, then I became an entrepreneur.

It might not surprise you to know that, personally, I'm sort of averse to controlling the level of risk that I would take, because my entire career path was sort of a maximally leveraged bet on one company, which is the one that I started.

I would be reluctant to discourage people from being entrepreneurs or doing what they want to do with their money or time. Of course, I'm in favor of reasonable things, like it should be clear to you what you're investing in. But generally speaking, I think if it's available to wealthy people—high-net-worth individuals—it should be made available to retail as well.

Let me broaden the conversation. Historically, we would have banks, brokerages, payment processors, and merchant acquirers. They were all disaggregated. They could all be public; they could all build thriving companies.

Now, with stablecoins and everything else, there's this creeping convergence. You're issuing a credit card. Coinbase has a credit card. SoFi has a federal banking license. Stripe just launched a new L1 called Tempo.

Everybody's competing with everybody. Tell us the scope of where you think Robinhood goes in the next 4 or 5 years, and what the financial landscape and infrastructure looks like. The Visas, the Mastercards, the JPMorgans—what roles do these companies play as you become more and more ambitious, girthy, and big, with your market cap and all that stuff?

I think the industry goes through periods of consolidation and then divergence. Robinhood has a unique advantage, which is that our customers put an increasing amount of their dollars into Robinhood.

What we're thinking about—and it became pretty clear to us as soon as we rolled out our second product—is that customers spent more time on Robinhood. The 2 products help each other. For example, with retirement, the big question was, if we launch retirement, is it going to cannibalize the core brokerage business? What we saw was the opposite. If someone opens up a retirement account, they tend to actually increase the amount they put into their individual account, and we saw that again with the credit card. If they're a credit card primary user, their top-of-wallet card, they actually put more money into Robinhood.

That gets us to a future where we ask ourselves: Can we be your comprehensive financial platform? Can you put your direct deposit into Robinhood? Can you put all of your money into Robinhood? Can you get to Gold subscriber premium status as soon as possible? Can we get all of your family members onto Robinhood as well, including your kids?

I don't think anyone's really thinking about it from that angle, but I think there's going to be over $130 trillion that changes hands from the Silent Generation and baby boomers to younger people. I think Robinhood is actually very well-positioned to be one of, if not the number one, primary institution that benefits from that transfer. We've got over a quarter trillion dollars in assets on the platform already, which seems like a big number, but it's actually just a drop in the bucket compared to what's going to happen.

Where do you see the JPMorgans and the Mastercards and Visas? How do they compete with an elegant product with hundreds of millions of users, the product velocity that you have, and the risk you're willing to take?

If you think about an incumbent, they have certain benefits. They're very muscular from a regulatory standpoint. They know how to deal with regulators. They've got global scale, tens or hundreds of millions of customers, and lots of assets.

But the disadvantage is that they're sometimes slow to adopt new technologies. They don't have the best engineering teams, they can't move very fast, and they can't hire the best talent. We don't have those downsides. We have great talent, we move really quickly, and we use the best technology.

We haven't been super acquisitive historically, even though we're doing more now, and that prevents us from being bogged down by these massive integration things that take multiple years. It's a question of whether we can get the benefits of scale while also maintaining the nimbleness of a technology startup.

Before we run out of time, maybe one last question: Can you tell the audience about the LLM you guys are building? This is a different project for you.

Oh, yeah.

What's the goal of that, and why did you decide to fund it outside the scope of Robinhood?

He's talking about Harmonic, which is a company that I started 2 years ago and am chairman of, completely separate from Robinhood. Basically, the goal there is to build what we call mathematical superintelligence. This is mathematical reasoning that exceeds the capability of any individual human researcher.

We had a pretty cool result a couple of weeks ago, where we announced gold-medal-level performance at the International Mathematical Olympiad, which is the biggest mathematics competition in the world. To my knowledge, we were the only formal model.

You're the only formal one that got IMO gold.

Yeah. OpenAI and Gemini—

OpenAI and Gemini did it with informal—

Gemini's informal model got a silver last year.

Explain why it's going to be so critical to have a mathematical superintelligence model.

Two reasons. One has to do with how these models are trained, and the other is more of a consumer pain point.

The thing that we've figured out with formal is how to verify that a statement is true very precisely. When you're doing reinforcement learning of these models, having a strong reward signal is very helpful, because you can just discard all the data that's not helpful and train on the high-quality, correct data.

When you're a user of these AI models, sometimes they hallucinate. This is not just a consumer problem, but also an enterprise problem, because if you're a software engineer using a coding model, your job has become, over the past couple of years, less about writing a whole bunch of high-quality code and more about reviewing LLM-generated code and making sure that it's correct.

In a world where you've got LLMs producing thousands and thousands of pages of code, human verification just doesn't scale, particularly for back-end systems. We want to solve that problem.

Okay, give it up for David Sacks's second favorite. [Music]

Good to see you guys.

Thank you. Thanks, brother. I'll see you in Vegas.