Prediction market insider trading sits at the centre of a fast-growing debate about what these platforms actually are: financial exchanges, glorified sportsbooks, or something stranger still. Carrie Sun, author of Private Equity: A Memoir, spent several months trading on Kalshi, one of the two leading prediction market platforms, and found herself wrestling with all three possibilities at once.
The premise is simple enough. Prediction markets let you trade on the outcome of real-world events. A binary option on whether J.D. Vance tops the Republican ticket in 2028 costs 38 cents on Kalshi; if he wins the nomination, you collect a dollar. These are called event contracts, and they now cover everything from Federal Reserve press conferences to Grammy categories to whether Costco raises the price of its hot dog combo.
From Grammy Night to a Question About Prediction Market Insider Trading
Sun’s entry point was personal. Her husband, Chris, is a classical composer whose album Don’t Look Down received three Grammy nominations. Five days before the ceremony, the Kalshi odds sat tight across three categories. Then, in one category, a single large bet pushed a little-known album called Seven Seasons to 97 percent probability. The total volume in that market was only $7,916. Someone, Sun concluded, must have known the outcome in advance.
She hedged. She placed $20 on no for Chris in each of the other two categories, reasoning that shorting her spouse was a rational hedge against heartbreak. The market, it turned out, was wrong in all three categories. Seven Seasons did not win. The presumed insider was, apparently, just a fan willing to burn money on obscure art. And Sun found herself watching her husband accept an award from behind a screen full of numbers that had already hollowed out the moment.
That experience raised the question that kept nagging at her: how easy is it to trade on information that nobody else is supposed to have?
At awards shows, the universe of potential insiders is wide. Vote-tallying firms, printers, publicists, their friends’ friends. Sun called Robert DeNault, head of enforcement at Kalshi and a former white-collar defence attorney, whose mandate is to build the rules of the road for insider trading on the platform. Kalshi’s rulebook is clear in principle: you cannot trade on a market if you have material nonpublic information about the outcome, or any ability to influence it. DeNault told Sun that the company runs internal surveillance systems and external security vendors, with thresholds that trigger flags for anomalous activity.
One such flag led Kalshi to fine and temporarily ban a MrBeast video editor who had been betting that specific words would appear in MrBeast videos before they dropped. A Discord group of traders, including a full-time sharp named Foster, had spotted the pattern themselves before Kalshi acted. They saw suspicious buy pressure the night before each video, watched every flagged word appear on cue twelve hours later, and quietly stopped trading those markets.
The Regulatory Picture Around Prediction Markets
The grey area DeNault described is vast, and regulators are only now beginning to map it. According to analysis by Cohen Milstein, sports-related trading now accounts for 85 percent of trading volume on Kalshi, a figure that underscores just how far these platforms have moved from their academic origins as information-aggregation tools into something closer to a mainstream betting product. The same analysis notes that new CFTC Chairman Michael Selig withdrew a 2024 proposed rule that would have banned sports and politics-related wagers on prediction markets entirely, leaving the regulatory framework in a state of flux.
That flux may not last. According to Latham & Watkins, in March the CFTC signalled it intends to take additional steps regarding prediction markets, including potential rulemaking on the application of insider trading principles and prohibitions on arranged trades and disruptive trading practices. The CFTC’s Division of Enforcement already holds authority to police illegal trading practices on any Designated Contract Market, including those related to event contracts.
The real-world consequences are not hypothetical. In late April, federal prosecutors charged Master Sergeant Gannon Ken Van Dyke, a US Army Special Forces soldier, with using classified information to trade on Polymarket, the crypto-based platform. Prosecutors allege he had helped plan and execute an operation to capture Venezuelan President Nicolás Maduro, then placed bets on Polymarket totalling nearly $410,000 in profit before asking the platform to delete his account. Van Dyke has pleaded not guilty.
On Polymarket, allegations of insider trading have themselves become the subject of markets. Earlier this year, an independent blockchain investigator known as ZachXBT alleged that an insider trading ring was operating at Axiom, a trading platform. Someone made over $400,000 betting that Axiom would be the next company ZachXBT exposed. Axiom said it was ‘shocked and disappointed’ and removed access to the internal tools allegedly used.
What the Markets Do to the Moment
Sun’s broader concern is subtler than the legal question. During the Oscars, she watched the Kalshi odds for each category spike to 99 percent roughly 20 seconds before her livestream showed the winner announced on stage. Someone at the Dolby Theatre, or receiving texts from inside, was profiting off the gap between their reality and hers. It looked, to her, like latency arbitrage in its purest form.
By the end of her experiment, Sun’s Kalshi portfolio was up around 25 percent. Her biggest single win was a $95.82 positive return on a Federal Reserve press conference mention market, after she correctly anticipated that a Bloomberg News reporter would ask Jerome Powell about stagflation. The stagflation market had $3.7 million wagered across various words and phrases at that single press conference alone.
What Sun found herself describing was not a new form of investing or even a new form of gambling, but something more corrosive: a machine that converts anticipation into data, replaces the experience of an event with a number, and leaves you staring at your portfolio at the moment your husband walks offstage with an award. The event becomes irrelevant. What matters is the position.
With the CFTC now signalling that formal rulemaking on prediction market insider trading is on the horizon, the question of who is allowed to know what, and when, is about to get considerably harder to avoid.
