Kalshi’s Tarek Mansour Wants to ‘Depolarize the Conversation’

Sign up for the daily CJR newsletter.

In recent months, Americans have wagered more than 1.3 billion dollars on the outcomes of the midterms via Kalshi, the world’s largest prediction market by volume and valuation. Kalshi already symbolized late-stage capitalism’s love affair with sports, but it has increasingly become a source of information—or speculation—about electoral politics, offering a market that priced every campaign event and polling blip. In the Michigan Senate race—which appeared to be among the most hotly contested in the country—Kalshi’s odds predicted a convincing victory for the Democratic candidate, Abdul El-Sayed. (He won, and Kalshi’s overall forecast of a blue tsunami proved accurate.) Adding complexity: Kalshi users are informed by—and its markets shift according to—news reports, but Kalshi also contributes to the news through partnerships with CNN, CNBC, and Fox, which use its numbers as forecasting data points. 

On Tuesday night, Kalshi turned into a midterms hub—a one-stop shop for election data, sourced straight from the Associated Press and ready to be bet on. Unlike voting booths, the betting remains open until elected officials are sworn in. “Our view is to just give as much information as possible and let people make their own independent judgments,” Tarek Mansour, the company’s cofounder and CEO, told me. “Show the polls, show the voting data, show the market odds, show the campaign fundraising, and so on. You’re not just relying on a few DC operatives telling you who’s going to win. It’s millions of people trading on it.” 

Mansour is convinced that Kalshi’s accuracy as a source of information and its freedom as a marketplace make it a force for good. Its critics have a different view. As a federally regulated exchange, Kalshi users trade against one another instead of betting against a house—a loophole that allows the company to bypass gambling regulations imposed by states. That means Kalshi and other prediction markets, while technically not gambling platforms, carry the risk of both gambling addiction and insider trading, as virtually any yes-or-no question can be traded on and information on anything can constitute insider knowledge. This year’s midterms have proved crucial for Kalshi, not just for the revenue the races generated, but also because multiple states are going to court for the right to regulate prediction markets. Mansour personally donated hundreds of thousands of dollars to candidates on both sides of the aisle. 

I spoke with Mansour at Kalshi’s headquarters, a twentieth-century printing plant and warehouse that has been converted to office space in Manhattan’s Meatpacking District. The company had just raised a billion dollars at a forty-billion-dollar valuation—yet Mansour still wanted to make a case in its defense: “There’s some sort of presumption that because it’s new, it’s bad,” he told me. “I just think people don’t understand it.” Our conversation has been edited for length and clarity.

ILN: I recently spoke to polling experts who noted that the use of Kalshi’s figures in TV programs like CNN’s The Odds can be misleading to audiences. What do you think about that?

TM: I think Harry Enten does a pretty consistently great job at saying that those are probabilities of success. Clearly, Kalshi works better than the polls. You can pull the data, play with Claude a little bit, replicate the study, and get the same results. It shows how calibrated it is. It actually works. It’s amazing.

But it’s not a crystal ball; it’s a probabilistic model. It doesn’t mean that if something has an 80 percent chance of happening, it’s going to happen. It’s going to happen probably four out of five times. 

What are the financial details of Kalshi’s partnerships with CNN, CNBC, and Fox?

We’re not going to disclose any financial terms.

Is there any money involved?

We’re not going to disclose anything about our partnerships. (Editor’s note: CNBC’s relationship with Kalshi includes customer acquisition and a minority investment.)

What does Kalshi get out of them?

What do you think?

Exposure, certainly. 

Yeah. We have a lot of exposure on Twitter, but we want exposure from the place where people consume information. When we started the company, in 2018, the thing that really got me over the fence was seeing polarization going rampant in the country—and by the way, that hasn’t gotten better. We are in a situation where we have more information than ever, but less truth than ever. People believe things that are just straight-up untrue. People can promote some story on social media, and it flies—by the time somebody corrects it, it’s too late. And people make shit up pretty much all the time, because the things that get promoted on social media are the things that are extreme in nature. The incentive structure is clickbait.

The incentive structure for prediction markets is very clean, very elegant, very simple: if you’re right, you will get rewarded, and if you’re wrong, you won’t. I’ve always said that the mission of the company was, over time, to depolarize the conversation. Bring it to the middle, to the boring land of nuance. In my view, it’s the only way to incentivize people to go out and seek that middle-ground truth.

So the only way to get to some sort of ultimate truth is through these yes-or-no questions on prediction markets?

What I’m saying is that you make money if you’re right and lose money if you’re wrong. On social media, you don’t get more likes if you’re right—you get more likes if you’re extreme. Cultural success for us is when you’re sitting at Thanksgiving and you’re debating with the uncle or cousin who is so certain about everything, and I always say, “You should tell them to put twenty bucks on it.” All of a sudden, they will get more careful. They’ll lean out a little bit. They will be like, “Okay, maybe now I have to think about what I say, because I may actually get punished.” 

Is Kalshi a source of information? Is it a news source?

I think it works with the news. Think about it this way: Is the stock market a replacement for market analysts on Wall Street? It’s not, but they’re not competing with each other. I think it’s the same thing here. 

What about Kalshi News, your news service that is based on your markets? You also have a newsletter of stories you can trade on; you have Twitter accounts that post news headlines and get great engagement. What’s the purpose of those?

I literally don’t even look at it. It’s one out of a million other things that we do. It’s not our product. Kalshi News on Twitter is, like, memes, or oftentimes we just take news headlines and kind of recycle them—you know, “Bloomberg said this.” It’s just an amplification of our markets. 

Looking at Kalshi’s markets, it feels like you don’t need, say, financial literacy, like you do for the stock market—you can just read the news and have a decent chance of doing well.

A few years before we started the company, in 2015, Philip Tetlock wrote a book about superforecasters that got me very excited. He brought together a bunch of geopolitical experts and put them up against a bunch of essentially random people, with one requirement: that they read the news. Guess which group outperformed the other.

If you look at our top ten thousand customers, there are no specific characteristics they share with regard to age, gender, or geography—the only thing that comes back is that they like to think critically about the world because they have this incentive to be right. They read the news a lot; they know how to ingest copious amounts of information and filter it over time.

Midterms-related trading on Kalshi has been close to one and a half billion dollars in volume. How does trading on the news compare with Kalshi’s other markets? 

It’s like two-thirds sports now. But people are so set on this question. Yes, there are more sporting events per week, but to me there’s no difference between the weather, economics, and politics. They’re the same customers. It’s the same structure.

What I will say about this is that people like and understand sports, and that’s a very good educational tool. When they come and trade on sports, they actually start trading on a lot of other things. Our politics category has grown faster than football this season.

Two years ago, on a Reddit “Ask Me Anything,” you wrote: “It’s election season in America, and you might have noticed something different about the 2024 election cycle compared to the others—this time around, you can bet on the outcome.” How is this different from gambling on the news?

When your friends tell you that they just bought Tesla stock, they can say “I bought Tesla stock” or “I just bet on Tesla.” But let me read something to you that’s even more interesting. Grain futures got legalized one hundred years ago. No one here is sitting around thinking that buying grain futures is gambling today. They’re the most boring form of financial instruments. This is from the Illinois Supreme Court at the time, from a suit against the Chicago Board of Trade: “It is not only contrary to public policy, but it is a crime—a crime against the state, a crime against the general welfare and happiness of the people, a crime against religion and morality, and a crime against all legitimate trade and business. This species of gambling has become emphatically and preeminently the national sin.” It was described at the time as a crime against humanity, but society evolved, and we enacted the Commodity Exchange Act to differentiate between what is an exchange and what is gambling. Sportsbooks set the lines themselves in a way that is adverse to their customers, and if you do something smart, you get banned—whereas in an open financial market, the smart traders, the people who do research, are critical for the model. The incentive is that I want truth.

Which then begs the question of what makes an insider. For instance, journalists often possess information about future events.

Can Wall Street Journal journalists trade on the stock market? They cannot trade on companies they cover, because they often have access to insider information.

Can they trade on Kalshi?

It depends on the market. On things like sports they can.

How is that enforced?

We don’t have rules against them. The Wall Street Journal has rules. In theory, they don’t have access to insider information because they don’t work at, say, Tesla, to know what their next quarterly earnings are going to be. But if you have access to material, nonpublic information that you are not supposed to disclose, then you cannot trade. 

But in theory, insider information would make your markets more efficient.

In theory it would, but in practice, it’s bad. The same way it would make the stock market more efficient, but it’s banned because you’d be creating an unbalanced marketplace where people with unique information have an asymmetric advantage. That’s what you don’t want—that’s when other people stop wanting to participate. And 80 percent of our customers don’t trade—lots of people are coming in to see how others price what’s been happening.

You say that people come to Kalshi as a source of trusted information. As with a news site, they expect you to present information without an agenda. But there’s a lot of political influence around Kalshi. Donald Trump Jr. is a strategic adviser. If I come to Kalshi for my information, how can I be certain that you are not going to show me something that politically benefits you?

If that were true, would we have been saying that the Democrats have 92 percent odds of winning the House of Representatives in the midterms? The point is that you don’t have to rely on anything. You can see every single dollar that’s ever been put in a market. I agree with you in the sense that it’s similar to a news organization, but Kalshi is not running the news. It’s not top-down; there are no editors who decide what goes and what doesn’t go. 

About the Don Jr. thing—we have copious advisers across the board, people who like prediction markets and see the vision, and we’re excited to work with all of them. And yes, we are a politically engaged company, there is no question about that, because the people on the other side of our fight—the casinos or the bigger exchanges—have been politically engaged for decades.

But my answer is no. There’s no conflict of interest. There’s no influence. The White House teleprompter operator—we caught him and enforced our rule uniformly. Anyone can participate, no matter how small, and the same rules and obligations apply. There’s no house. It’s a free, competitive marketplace. And it has to be fair markets. It doesn’t matter who cheats—they have to be punished, because otherwise the market doesn’t work. 

What’s your news diet? What’s the first thing you read in the morning?

I look at Twitter to see what people are saying. There was all of this news about James Talarico in the first week of October, that he was “missing” or something, and now the election is over. The first thing I did was I looked at the odds for the race on Kalshi. He had a 64 percent chance earlier that week, then it was down to 62: only a 2 percent change. The market hadn’t priced it as a real narrative. 

Journalists have always struggled to generate engagement because reading the news is a passive thing: you consume the information, you move on. Is Kalshi gamifying the news?

It depends on what you mean by gamification. I don’t think this is gamifying the news any more than, or certainly not as much as, social media. Does it feel like as much of a dopamine hit as opening Instagram? I don’t know. To me, it feels mathematical and statistical. It’s doing something else, and we can see it in the data: It’s actually increasing civic engagement. It pushes people to do more research, to go out and seek knowledge in a world where most incentive structures are based on dopamine, clicks, and impressions. 

Has America ever needed a media defender more than now? Help us by joining CJR today.