Khamenei Is Dead, but Kalshi Won’t Pay: Inside the $54M Lawsuit That Could Break Prediction Markets

Ayatollah Ali Khamenei died in a barrage of American and Israeli ordnance on February 28, 2026. Thousands of traders on Kalshi — the CFTC-regulated prediction exchange valued at $11 billion — held “Yes” shares in a market titled “Ali Khamenei out as Supreme Leader?” They expected $1.00 per contract. Instead, Kalshi invoked a clause buried in its technical documentation that excluded death as a valid resolution trigger, settled positions at the pre-strike price of $0.02 to $0.29 per share, and told everyone they should have read the fine print. On March 5, a class action lawsuit landed in federal court in Los Angeles, turning a contract-language dispute into the first major legal test of whether prediction markets can promise one thing and deliver another. The outcome will shape how $44 billion in annual prediction market volume gets regulated, settled, and trusted — or doesn’t.

A binary question with a hidden asterisk

Kalshi’s Khamenei market launched around January 8, 2026, under the internal CFTC product certification name “WLEADEROUT” — a generic template for world leader “out of office” contracts. The user-facing title was clean and intuitive: “Ali Khamenei out as Supreme Leader?” Traders could buy “Yes” or “No” on rolling monthly expirations — March 1, April 1, and later dates. The primary rules summary displayed on the market’s webpage stated simply: “if Ali Khamenei leaves office before March 1, 2026, then the market resolves to yes.” Binary. Clear. One dollar if you’re right, zero if you’re wrong.

Except it wasn’t. Buried in a PDF filed with the CFTC — the WLEADEROUT contract terms document, accessible only by clicking through a rules link on the market page — sat a provision that would become the most controversial clause in prediction market history:

“If [leader] leaves solely because they have died, the associated market will resolve and the Exchange will determine the payouts to the holders of long and short positions based upon the last traded price (prior to the death). If a last traded price is not available or is not logically consistent, or if the Exchange determines at its sole discretion that the last traded prices prior to death do not represent a fair settlement value, the Outcome Review Committee will be responsible for making a binding determination of fair allocation.”
— WLEADEROUT Contract Specification, filed with CFTC

This was the “death carveout.” It meant that if Khamenei died — the single most probable mechanism by which an 85-year-old theocratic leader with a lifetime appointment might leave office — traders would not receive $1.00. They’d receive whatever the last pre-death price happened to be. A market framed around the question of whether an autocrat would lose power quietly excluded the most likely way he’d lose it.

Kalshi’s rationale was straightforward: U.S.-regulated exchanges cannot legally list markets that directly resolve on death, assassination, or war under the Commodity Exchange Act’s Section 5c(c)(5)(C)(i) and CFTC Regulation 17 CFR 40.11. The death carveout, Kalshi argued, was what made the market legal in the first place. Without it, the contract would be a prohibited “death market.” CEO Tarek Mansour would later write on X: “We don’t list markets directly tied to death. When there are markets where potential outcomes involve death, we design the rules to prevent people from profiting from death.”

The problem was disclosure. The death carveout appeared nowhere in the user-facing rules summary. It existed only in the CFTC-filed technical specification PDF. No pop-up warned traders. No banner flagged the exception. The market page said “leaves office” and left it at that. As Kalshi itself would later concede in an email to users: “Many users did not have a full understanding of the rules for this market.”

Fourteen hours of trading after the bombs fell

On Friday evening, February 27, the “Before March 1” contract traded at $0.02 — a 2% implied probability. The “Before April 1” contract sat at $0.28. Prediction markets, like intelligence agencies, did not see what was coming.

At approximately 1:14 AM ET on Saturday, February 28, U.S. and Israeli forces launched coordinated strikes against Iran — Operation Epic Fury. The first bombs hit. Prices barely moved. By 2:30 AM, Donald Trump posted a video announcing “major combat operations.” The March 1 contract ticked up to $0.06. The April 1 contract climbed to $0.41. Traders were waking up.

By mid-morning, unconfirmed reports that Khamenei had been killed began circulating on social media and news outlets. Kalshi’s market became the center of the information universe. The March 1 contract hit $0.34. The April 1 contract surged to $0.66. Trading volume exploded — roughly $20 million changed hands on Saturday alone, more than doubling the market’s prior lifetime volume.

Then Kalshi did something extraordinary. Its official account posted on X: “BREAKING: The odds Ali Khamenei is out as Supreme Leader have surged to 68%.” CEO Mansour reposted it. The platform was actively promoting a market it would later refuse to pay out. Underneath the promotional post, Kalshi added a reminder about the death carveout — the first time many traders saw it. A green box warning was retroactively added to the market page’s UI, which Kalshi described as a “visual highlight of existing terms” rather than a rule change. But for the thousands of traders who had entered positions before that moment, the disclosure came after the bet, not before.

Trading continued for approximately 14 hours after strikes began. Kalshi did not halt the market until roughly 2:59 PM ET — nearly twelve hours after Trump’s video announcement, and long after multiple outlets reported Khamenei’s likely death. The last trade before the halt printed at $0.09 on the March contract and $0.42 on the April contract. At 4:37 PM, Trump formally announced Khamenei’s death. Iranian state television confirmed it hours later. By 10:06 PM, Kalshi formally closed the markets. By 10:19 PM, settlement was executed.

The March 1 contract settled at $0.02 per share. The April 1 contract settled at $0.29. These were the prices from before 1:14 AM — before the first bomb fell. Traders who had bought “Yes” at $0.50 expecting a $1.00 payout received pennies. One Israeli-American executive profiled by the Washington Post had bet $3,460 across multiple contracts, briefly saw $63,000 in apparent winnings on his screen, and received a fraction of that.

What $259.84 might cost the prediction market industry

The class action complaint — Risch v. KalshiEX LLC, Case No. 2:26-cv-02390 — landed on March 5, 2026, in the U.S. District Court for the Central District of California. The 25-page filing, brought by plaintiffs Adam Risch and Yonatan Gliksman through Los Angeles firm Novian & Novian LLP, represents what may be the most consequential consumer protection case in prediction market history.

The irony is in the numbers. Risch and Gliksman held combined positions worth approximately $259.84 in a market that processed more than $54 million in volume. Risch had purchased 255 “Yes” contracts on the April 1 expiration at $0.28 each back on February 18 — a $75 investment. Gliksman bought 670 contracts on the March 1 expiration at roughly $0.07 on the morning of February 28, and another 781 contracts that afternoon. Their personal stakes are modest. The proposed class they represent is not.

The complaint defines the class as all U.S. persons who held “Yes” positions on any expiration date when Kalshi halted trading on February 28 — potentially thousands of traders across millions of dollars in positions. Two subclasses divide the group: Subclass A covers those who acquired positions before February 28 (represented by Risch), and Subclass B encompasses those who entered on the day of the strikes (represented by Gliksman). The legal theory targets five Kalshi entities: KalshiEX LLC, Kalshi Inc., Kalshi Klear LLC, Kalshi Klear Inc., and Kalshi Trading LLC.

“The operative contract terms, as presented to and understood by reasonable consumers, provided that ‘if Ali Khamenei leaves office before March 1, 2026, then the market resolves to yes.’ This language was clear, unambiguous, and binary.”
— Risch v. KalshiEX LLC, Class Action Complaint

Six causes of action stack the complaint: breach of contract, fraud, negligent misrepresentation, unjust enrichment, violation of California’s Unfair Competition Law (Bus. & Prof. Code §§ 17200), and violation of the California Consumers Legal Remedies Act (Civ. Code §§ 1750). The plaintiffs seek compensatory damages equal to $1 per “Yes” share, punitive damages sufficient to deter similar conduct, disgorgement of all profits, and injunctive relief requiring Kalshi to prominently and conspicuously disclose all material settlement terms before trades are executed.

The complaint cites Kalshi’s own post-crisis email as evidence of bad faith — the admission that “many users did not have a full understanding of the rules” — and frames the entire market as the “poster child of unfair competition, deceptive corporate behavior, and consumer fraud: a company that lures consumers in with clear promises, then pulls the rug out from under them when those promises become inconvenient.”

Risch v. KalshiEX is not the only legal action. Law firm Lieff Cabraser Heimann & Bernstein announced its own investigation on March 3. At least four other lawsuits targeting Kalshi are pending in federal courts across New York, Illinois, and Alabama, though these primarily address separate allegations about unlicensed sports gambling.

Kalshi’s defense collapses under its own admissions

Mansour’s defense crystallized into three pillars, articulated most clearly in a March 6 post responding to the lawsuit:

The third point is the most verifiable. Kalshi spent approximately $2.2 million reimbursing all trading fees from the Khamenei market, fully refunding cost-of-entry for anyone who purchased positions after the strikes began at 1:14 AM, and covering net losses so that no trader ended up with less money than they started with. Spokesperson Elisabeth Diana told CNN: “We even reimbursed all fees and net losses out of pocket — to the tune of millions of dollars — to make sure not a single person lost money on this market.”

But Kalshi’s definition of “lost money” is the crux of the dispute. Traders who bought “Yes” at $0.50 and received $0.29 in settlement plus a refund of their $0.50 cost basis didn’t lose money in an accounting sense — they broke even. They did, however, lose the $0.50 in profit they would have received had the contract paid $1.00. Kalshi frames this as making traders whole. Traders frame it as having their winnings confiscated. The distinction between “no one lost money” and “no one got paid” is the fault line of the entire case.

The first two pillars are harder to sustain simultaneously. Mansour insisted the rules “were clear from the outset” and “did not change.” But on the same day, Kalshi posted two contradictory clarifications on February 28. The first stated the market would settle based on “the last traded price prior to confirmed reporting of death.” The second, posted later, corrected this as “grammatically ambiguous” and restated it as “the last traded price prior to death” — a materially different standard that yields different settlement prices (death occurred around 1:14 AM; confirmed reporting came hours later). Kalshi reimbursed traders who transacted between the two contradictory notices.

The deeper contradiction lives in Kalshi’s email to users. You cannot simultaneously claim that your rules were always clear and also concede that “many users did not have a full understanding” of those rules. One of these statements is true; they cannot both be. The class action complaint treats this admission as a confession. Bloomberg columnist Matt Levine offered perhaps the sharpest analytical frame: if traders had genuinely understood the death carveout, prices should have dropped when Khamenei’s death was reported — not surged to 68%. The price action itself is evidence that the market was misleading.

There is also the Jimmy Carter precedent. In late 2024, Kalshi ran a market asking “Who will be at Trump’s inauguration?” with Carter as one option. After Carter died in December 2024, Kalshi settled that contract to “No” — effectively resolving a market based on someone’s death. As one trader noted on Kalshi’s Discord: “You settle on death, just not when it makes you money.”

A rulebook amendment that confirms the problem

Three days after the crisis and two days before the lawsuit, Kalshi filed a “Draft Death Caveat Rulebook Amendment” with the CFTC on March 2, 2026. Signed by Chief Regulatory Officer Richard Heaslip, the filing amended Rule 6.3 of the exchange rulebook by inserting a new subsection (e) that codified death settlement procedures across all contracts involving living persons. The rule took effect March 17.

“The amendment further codifies and memorializes the exchange’s practice for markets in which the market’s primary subject is a natural person that has passed away.”
— Richard Heaslip, Kalshi Chief Regulatory Officer, CFTC filing cover letter

The new Rule 6.3(e) grants Kalshi sweeping discretion. If a contract’s subject dies before expiration, Kalshi may settle at the last traded price before death. If trading was “materially affected by the circumstances giving rise to the death,” Kalshi may instead roll the price back to before those circumstances became “known or reasonably anticipated by market participants.” The exchange may rely on approximate timestamps. If no price is deemed fair, the Outcome Review Committee determines settlement. All determinations are “final and not subject to review.”

The timing tells a story. If the death carveout was always part of Kalshi’s rules and always clearly disclosed, why did the exchange need to rush a formal rulebook amendment to the CFTC within 48 hours of the controversy? The answer, which industry analysts were first to articulate, is that the death carveout was never part of Kalshi’s main rulebook. It appeared only in the technical specifications of individual contract PDFs. The March 2 filing moved it into the general exchange rules — an implicit admission that the prior disclosure framework was inadequate.

For traders, Rule 6.3(e) creates a new reality. Every Kalshi contract involving a living person now carries an explicit death-settlement mechanism where the exchange has sole discretion over the settlement price, the timestamp used, and whether the Outcome Review Committee intervenes. Those determinations cannot be appealed. For the class action plaintiffs, the amendment is a gift — it concedes that formalization was needed, which undermines the claim that prior rules were sufficient.

The same event, four different outcomes

The Khamenei crisis exposed a natural experiment: the same real-world event resolved differently across every major prediction market platform, revealing how contract design determines who gets paid. Kalshi is not the only platform to face payout disputes with traders — but its handling of this one was uniquely damaging.

Polymarket’s Iran-related markets also became ground zero for insider trading allegations that raised national security concerns. Blockchain analytics firm Bubblemaps flagged six newly created accounts that collectively earned $1.2 million by betting on the February 28 strikes hours before they began. All six wallets were created in February 2026, most funded within 24 hours of the attack, and all exclusively traded Iran strike-date contracts. The largest single wallet turned roughly $61,000 into $494,000. A separate account called “Magamyman” made $553,000 wagering $32,000 just hours before strikes when odds sat at 17%. PANews conducted a deeper on-chain analysis identifying 521 suspicious addresses trading during specific windows, with 62 having no activity outside Iranian markets.

The comparison is damning for Kalshi. PredictIt used six additional words — “or otherwise vacate” — and avoided a crisis. Polymarket used “consensus of credible reporting” and ultimately paid out, despite a messy dispute process. Only Kalshi, the platform that positioned itself as the most institutional, most regulated, most trustworthy venue, refused to pay.

A regulatory firestorm with no clear resolution

The political response was immediate and bipartisan in its alarm, if not in its proposed solutions. Senator Chris Murphy (D-Conn.) had been raising concerns about prediction markets since January 2026. On March 1, he posted: “It’s insane this is legal. People around Trump are profiting off war and death. I’m introducing legislation ASAP to ban this.” In subsequent statements, he alleged that “there were potentially people inside the Situation Room last week cheerleading America into war because they had made secret bets that were going to pay off.” His proposed bill would ban prediction market trades related to government actions, including military strikes, with carveouts for traditional financial instruments like bond futures.

On February 24 — before the strikes — six Democratic senators led by Adam Schiff (D-CA) had already sent a letter to CFTC Chair Michael Selig demanding action. Co-signed by Senators Cortez Masto, Blumenthal, Booker, Kaine, and Rosen, the letter cited CFTC Regulation 40.11 and argued the Commission must “categorically prohibit any contract that resolves upon or closely correlates to an individual’s death.” The senators set a March 9 deadline for response.

“Prediction markets are promoting opportunities to bet on events that can only be seen as a proxy for war or assassination. This betting market shouldn’t exist in the first place.”
— Amanda Fischer, Policy Director, Better Markets

Amanda Fischer, Policy Director at Better Markets and former SEC Chief of Staff, offered the sharpest institutional critique. She told the Washington Post she was “not convinced” by Mansour’s explanation: “How is an 86-year-old theocratic leader supposed to lose his power other than through death?” On the CFTC’s capacity to regulate the industry, Fischer was blunt: “The CFTC is not well equipped to police these markets. They do not have nearly enough staff or expertise.” She noted the agency’s Chicago office “recently lost its last enforcement attorney out of a team that once numbered 20.”

CFTC Chair Selig — notably the only sitting commissioner on a body designed for five — addressed prediction markets at the Milken Institute Future of Finance conference on March 3 but did not directly confront the death-contract question. He promised “guidance in the very near future” and said the agency would “set very clear standards as to what can be self-certified in our markets and what cannot.” But Selig’s track record gives reformers little comfort: he withdrew the Biden-era proposed rulemaking on event contracts, stacked his 35-member Innovation Advisory Committee with industry CEOs from Polymarket, Kalshi, Coinbase, Robinhood, FanDuel, and DraftKings — and included zero consumer advocates.

Separately, Rep. Ritchie Torres (D-NY) introduced the Public Integrity in Financial Prediction Markets Act of 2026 with 30-plus Democratic cosponsors, including Nancy Pelosi. The bill would prohibit federal officials from trading prediction contracts tied to government policy when they possess material nonpublic information. Senators Jeff Merkley and Amy Klobuchar introduced a companion Senate bill, the End Prediction Market Corruption Act, which goes further by completely banning elected officials from all event contract trading. Reps. Blake Moore (R-Utah) and Salud Carbajal (D-Calif.) introduced additional legislation seeking to block platforms like Kalshi and Polymarket from offering markets on war and sports topics.

Meanwhile, the state versus federal jurisdiction war grinds on. Nevada’s Gaming Control Board sued Kalshi on February 18. A federal judge sent the case to state court on March 2, ruling the Commodity Exchange Act does not “completely preempt” state gaming laws. Both Kalshi and Polymarket have appealed. Massachusetts obtained an injunction. Michigan filed civil enforcement. Kalshi faces approximately 20 lawsuits from states and tribal authorities. The company’s argument that CFTC regulation preempts all state gambling oversight has been rejected by courts in Nevada, Maryland, and Massachusetts, with only Tennessee ruling in Kalshi’s favor — creating a circuit split that may eventually reach the Supreme Court.

What traders need to understand about settlement risk

The Khamenei debacle introduced a category of risk that most prediction market traders had never considered: settlement risk — the possibility that you correctly predict an outcome but don’t get paid because of how the contract defines “outcome.”

The “last traded price” mechanism deserves particular scrutiny. When Kalshi invokes this provision, the settlement price is the fill price of the last matched order on the exchange’s central limit order book before a timestamp the exchange selects. In the Khamenei case, Kalshi chose 1:14 AM ET — the moment of first strike impact. But the lawsuit alleges Kalshi never disclosed how it selected this timestamp, and the contract terms give the exchange “sole discretion” over the determination. Under the new Rule 6.3(e), Kalshi can also roll the price back to before circumstances became “known or reasonably anticipated by market participants” — a standard so elastic it could apply to news reports, social media rumors, or military build-ups days before an event.

This creates an asymmetric payoff structure that sophisticated traders must internalize. In a standard binary contract, “Yes” holders face two outcomes: $1.00 if correct, $0.00 if wrong. With a death carveout, a third outcome exists: the event happens, you’re correct, and you receive some arbitrary amount between $0 and $1 determined by the exchange. The expected value of a “Yes” position is therefore lower than the market price implies, because the highest-probability resolution scenario (death of an elderly leader) triggers the worst possible payout. Traders who priced the April 1 contract at $0.28 were implicitly assigning a probability to Khamenei leaving office — but the contract couldn’t pay them for the most likely mechanism of departure.

The broader implication is structural. Every geopolitical leadership contract on a CFTC-regulated platform now carries embedded optionality that benefits the exchange. The death carveout functions as a free put option for the platform: if the leader leaves office alive, the exchange operates normally. If the leader dies, the exchange settles at a lower price and keeps the difference between what it collected from “No” holders and what it pays “Yes” holders in its clearinghouse. Kalshi maintains it made no money and reimbursed losses. But the structural incentive exists regardless of whether it was exercised in this instance.

For traders evaluating future contracts, the lesson is procedural. Read the CFTC-filed contract specification — not just the market page summary. Look for settlement exceptions, Outcome Review Committee provisions, and discretionary authority clauses. Compare language across platforms: PredictIt’s “otherwise vacate” and Polymarket’s “removed from power” both encompassed death without carveouts. Kalshi’s “out as Supreme Leader” did not. Three words of contract language determined whether traders received $1.00 or $0.02 for the same correct prediction.

The paradox at the center of a $44 billion industry

The prediction market industry processed roughly $44 billion in volume in 2025 — a figure that may reach $1 trillion by decade’s end, according to Eilers & Krejcik. Robinhood CEO Vlad Tenev calls prediction markets the “fastest-growing business we’ve ever had,” with over 12 billion contracts traded in 2025 and a $300 million revenue run rate. Kalshi alone raised $1 billion at its $11 billion valuation in December 2025, and both Kalshi and Polymarket are now targeting $20 billion valuations in new funding rounds. DraftKings acquired a CFTC-registered exchange. FanDuel partnered with CME Group. Coinbase acquired a prediction market startup. Donald Trump Jr. serves as a paid advisor to both Kalshi and Polymarket simultaneously. The prediction market wars have never been more intense.

This gold rush faces an existential question that the Khamenei market crystallized: can you build a market that everyone understands is about mortality, promote it as mortality becomes imminent, then claim it was never about mortality when the bill comes due?

Kalshi’s position is that legal compliance required the carveout. The Commodity Exchange Act prohibits contracts involving assassination, terrorism, and war. Without the death exclusion, Kalshi argues, the market couldn’t exist. But this defense contains its own contradiction. If the market’s primary resolution mechanism (death) is legally prohibited, perhaps the market itself shouldn’t have been listed. Amanda Fischer put it most directly: “How is an 86-year-old theocratic leader supposed to lose his power other than through death?”

The MrBeast insider trading fine from February 25, 2026 — video editor Artem Kaptur fined $15,000 plus disgorgement for trading $4,000 on YouTube streaming markets with inside information — showed Kalshi’s surveillance systems work when they want to. The platform disclosed 200 insider trading investigations in the past year. It can detect anomalous trading patterns. It can freeze accounts. It chose to keep the Khamenei market open for 14 hours after bombs started falling.

Polymarket’s insider trading problem is arguably worse in scale — the six mystery wallets, the 521 suspicious addresses flagged by PANews, the Israeli military personnel indicted for using classified intelligence to trade — but Polymarket at least paid out. Kalshi’s wound is self-inflicted: a regulated exchange, seeking institutional credibility, fumbled the highest-profile event in its history through contract design that prioritized legal hedging over customer trust.

The class action will take years to resolve. Kalshi will argue the rules were disclosed, the settlement followed those rules, and no trader suffered a net loss after reimbursements. The plaintiffs will argue the disclosure was inadequate, the rules were unconscionable as applied, and “not losing money” is not the same as receiving promised payouts. Both sides have ammunition. But regardless of the legal outcome, the reputational damage is done.

Prediction markets promised to be the purest expression of information efficiency — a mechanism where the crowd’s wisdom is distilled into prices that outperform pundits, polls, and intelligence agencies. The Khamenei market delivered on that promise right up until the moment it mattered. Prices surged because traders correctly identified what was happening. Then the platform told them their correct prediction didn’t count. If $54 million in trading volume on the correct side of a binary outcome doesn’t trigger a payout, the question isn’t whether prediction markets work. It’s whether the people running them will let them.

---

Sources

• Commodity Exchange Act, Section 5c(c)(5)(C)(i) — CFTC
• 17 CFR 40.11 — Special certification procedures for event contracts — Electronic Code of Federal Regulations
• H.R. 7004 — Public Integrity in Financial Prediction Markets Act of 2026 — Congress.gov
• Senator Schiff letter to CFTC Chair Selig on death contracts — Office of Senator Adam Schiff
• CFTC Should Investigate Potential Insider Trading Related to Iran Attack — Public Citizen
• Kalshi Rule 6.3(e) Death Caveat Rulebook Amendment — Kalshi