CEOInterviews.AI
Start App
Tarek Mansour
Co-founder & CEO, Kalshi

Explaining Prediction Markets with the CEO of Kalshi

📅 Nov 03, 2026 Natalie Fratto 11 MIN 21 VIEWS 73 SEGMENTS · 2 SPEAKERS
Natalie Fratto (Creator of Charts & Crafts) sits down with Tarek Mansour, the CEO of Kalshi to better understand how prediction markets actually work, the counter-intuitive impact of prediction markets on elections, and why he believes they can make the world a better place. We cover: How Kalshi works (hint: a lot like the NYSE) 60¢ = 60%: how event contracts price probability Prediction markets vs. sportsbooks Gambling or market? Why federal regulation beats a state-by-state patchwork Polls vs. markets: skin in the game & the pursuit of truth Why markets reward the boring take What you can...

What Tarek Mansour said

Written from the verified transcript and checked against it. Every figure links to the moment it was said.

Tarek Mansour, co-founder and CEO of Kalshi, explains that prediction markets function like traditional exchanges, with event contracts paying $1 if an event occurs and zero otherwise, prices set by traders competing, and no house counterparty. He contrasts this with sports books, where users bet against the house, prices are set by the house, and winners are often banned. Mansour emphasizes that Kalshi is federally regulated under the Commodities Exchange Act (1974), which he argues provides a nationwide uniform framework and better prices for users. He defends the use of markets over polling, citing their incentive for truth and accuracy, and highlights Kalshi's role in prediction during elections like 2024. He explains hedging with a concrete ice cream shop example, showing how buying weather contracts can stabilize daily profits. The conversation also touches on forbidden markets like wildfires due to moral hazard risks. Mansour closes with a personal reflection on his Lebanese upbringing and his admiration for anime character Goku.

Key takeaways

  1. Kalshi event contracts pay $1 if the event happens and $0 if not, with prices set by traders, not by Kalshi itself.
  2. Sports books ban or limit winners because they bet against the house; prediction markets reward winning traders.
  3. Kalshi is regulated under the Commodities Exchange Act of 1974, ensuring a nationwide framework with uniform rules.
  4. Prediction markets outperform polls because traders have skin in the game: they are financially rewarded for correct predictions and punished for wrong ones.
  5. Kalshi avoids markets like wildfires because they could create moral hazards, potentially incentivizing the adverse event.

Numbers and commitments

FigureWhat it refers toTypeAt
$1 Payout per event contract if event happens price 1:18
60% Market-implied probability of Fed rate hike example metric 1:18
60 cents Price of yes contract in Fed example price 1:18
1974 Year Commodities Exchange Act was enacted timeline 4:41
200 Number of cold event contracts purchased in hedge example metric 9:10
$20 Cost of buying 200 cold contracts at 10% each price 9:10
$200 Payout if cold day happens in hedge example price 9:51
$80 Stabilized daily profit after hedging metric 9:58

Chapters

  1. 0:00How event contracts work on Kalshi
  2. 2:57Difference from sports books
  3. 4:41Regulation and the Commodities Exchange Act
  4. 6:29Value of prediction markets over polling
  5. 8:14Forbidden markets and moral hazards
  6. 8:52Hedging example for businesses
  7. 10:16Personal background and anime influence

Questions asked in this interview

12
  1. 0:18The diversity of participants and rules that foster competition, fairness, or are you betting against the house?
  2. 0:34I'm opening KHI for the first time. What am I seeing?
  3. 1:08On Kalshi, you see a market like will the Fed rate hike next quarter?
  4. 2:08Why wouldn't you just use the stock market to place a futures contract?
  5. 2:48Is the bank the broker in this situation?
  6. 2:57How is that different than a sports book?
  7. 6:16If Kalshi is the most successful you can possibly be, what world do we live in?
  8. 6:36Why does it need to be a market though?
  9. 6:51The 2016 election is a funny example of this, right?
  10. 7:17One of the challenges I saw in the 2016 election is as we continue to have algorithmic feeds that feed our echo chambers, on a prediction market that doesn't work the same way, right?
  11. 8:14You can trade on a vast majority of things. What can't you trade on?
  12. 8:52Make it simple for me, maybe pretend I'm an ice cream shop owner. How would hedging help me?
Tarek Mansour 0:00 ↗
Financial markets have always been complicated to society. Doesn't mean they're not important. Actually, they're incredibly important.
Interviewer 0:04 ↗
One of the fastest growing companies in the world, but also one of the most controversial.
Tarek Mansour 0:08 ↗
We want to discover the price of something. So, we want the winners because winners are smart people. You know, I'm trying my best right now. It's like so bad.
Interviewer 0:14 ↗
Here's how I understand an event contract. Check my understanding.
When you put money on something you don't control to make more money, that seems and looks and feels like gambling. But are you trading on an open marketplace? The diversity of participants and rules that foster competition, fairness, or are you betting against the house? That is really the foundational difference.
I'm opening KHI for the first time. What am I seeing?
Tarek Mansour 0:39 ↗
You're basically seeing something that functions exactly like the New York Stock Exchange, but it's a very large range of topics that you might relate to or care about. Whether it's economics or sports or weather or politics, it essentially works exactly like the NASDAQ. People are buying and selling these contracts, these shares of events from each other, and the prices are set by people competing against each other to figure out what the price of something is. And you can essentially decide to go yes on an event like I think Brexit is going to happen, or I think this team is going to win the next game, or no. And if you're right, you get paid at the end. If not, well, you don't get paid at the end.
Interviewer 1:08 ↗
On Kalshi, you see a market like will the Fed rate hike next quarter? However many people are voting, I suppose with their trades, makes it look like a probability.
Tarek Mansour 1:18 ↗
That's right. Each contract is a bit like an option. It pays out $1 if the event happens and zero if not. In this example right now, it seems like the market thinks that there's a 60% chance of the event happening. That is the same thing as 60 cents on the dollar. You can buy it for 60 cents on the dollar.
Interviewer 1:35 ↗
So, if I buy a yes event contract and the Fed does hike next quarter, then I would gain 40 cents.
Tarek Mansour 1:41 ↗
Yes. You got to pay one and your net profit is 40.
Interviewer 1:43 ↗
And then let's pretend it doesn't happen. I have paid...
Tarek Mansour 1:46 ↗
60 cents and then you get zero. Right now you're buying for 60 cents. This person is selling the yes to you and then they're putting up the 40 cents that would be required for you to get paid out. Those prices are 60 cents and 40 cents. They're not set by Kalshi. That 60 cents is person A saying I'm willing to buy at 60.
And then they wait for someone else to say I'm willing to sell it to you at 60 cents, or I'm willing to buy no for 40 cents. Same thing.
Interviewer 2:08 ↗
Why wouldn't you just use the stock market to place a futures contract?
Tarek Mansour 2:11 ↗
The stock market usually is to buy and sell shares. And then the exchanges that are regulated by the CFTC, the Commodity Futures Trading Commission, is not buying shares in companies. It's buying derivatives, right? Whether futures or swaps or other on commodity, something that impacts more than one company like the price of grain. It could be whether a Fed raise interest rates, a hurricane hitting in a certain town, a game going one way or the other, but it's actually the same exact underlying mechanic.
Interviewer 2:33 ↗
So in a prediction market, you have two people who trade against each other. Underneath that call sheet is a platform where you facilitate that transaction and then you take a fee.
Tarek Mansour 2:44 ↗
That's right. That structure that you see in the New York Stock Exchange is the exact same one.
Interviewer 2:48 ↗
Is the bank the broker in this situation?
Tarek Mansour 2:49 ↗
Yes. The bank is one of the types of brokers. Yes. What we call broker-dealer.
Interviewer 2:53 ↗
So always, even in the stock market, it is a person-to-person trading.
Tarek Mansour 2:56 ↗
Yes.
Interviewer 2:57 ↗
How is that different than a sports book?
Tarek Mansour 2:58 ↗
In the sports book, none of what I just described exists. You're betting against the house. You're betting against the sports book itself. The prices are set by the house. There's no open price discovery going on. It's a bilateral bet. They just bet against you against a bunch of different people. And if you win too much, if you're doing a lot of smart research, they generally ban you and limit you because you're making money against the house, right? They're losing to you. So their incentive is to limit the winners and over time figure out who are the people that are losing consistently and sometimes may have a problem or addiction signs, and then we figure out how to give them incentives to keep coming back to lose more to the house.
Interviewer 3:31 ↗
What happens if you win too much on a prediction market?
Tarek Mansour 3:33 ↗
Nothing. They keep winning. They keep trading. I mean, prediction market is one of our key functions. The reason to exist, we want to discover the price of something. So we want the winners because winners are the smart people. The smart traders are doing a bunch of research, gathering information, and then servicing that information by trading it.
Interviewer 3:47 ↗
I've heard you talk about your two tests to see what gambling is. So in the stock market, supply and demand are setting the price. In a prediction market, supply and demand are setting the price.
Tarek Mansour 3:57 ↗
That's right.
Interviewer 3:57 ↗
In the sports book, the house sets the price.
Tarek Mansour 3:59 ↗
That's exactly right. There's no price discovery. Think of millions of people basically trading against each other and bumping against each other, competing to discover the price of something. Let's say you disagree with the price. It's at 60, you think it's 50, you trade in a way that moves it to 50. And someone else disagrees, boy, it trades in a way that moves it to 54. And that constant real-time bumping between different traders is the notion of price discovery.
Interviewer 4:19 ↗
So in a prediction market and in a stock market, anyone can take either side.
Tarek Mansour 4:22 ↗
They can enter a position and exit it whenever they want. But that's the notion of a liquid exchange. If you come and say I want to buy 100 shares on something and there's no one else on the other side, then no transaction happens.
Interviewer 4:32 ↗
Right.
Tarek Mansour 4:32 ↗
Sports book, the house is always there. It's one counterparty.
Interviewer 4:35 ↗
When I think about my determination of what's gambling, a lot of it comes down to what's regulated. Stock market is very regulated.

33 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Mansour, T. (2026, November 3). Explaining Prediction Markets with the CEO of Kalshi [Interview transcript]. Natalie Fratto. CEOInterviews.AI. https://ceointerviews.ai/interview/3188028/

MLA

Tarek Mansour. "Explaining Prediction Markets with the CEO of Kalshi." Natalie Fratto, 3 Nov. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/3188028/.

BibTeX
@misc{mansour2026_3188028,
  author       = {Tarek Mansour},
  title        = {Explaining Prediction Markets with the CEO of Kalshi},
  howpublished = {Interview transcript, Natalie Fratto. CEOInterviews.AI},
  year         = {2026},
  month        = {nov},
  url          = {https://ceointerviews.ai/interview/3188028/},
  note         = {Speaker-attributed transcript with timestamps}
}