The most lucrative trade of 2025 didn’t come from a whale’s leveraged position or a DeFi arbitrage bot. It came from a White House teleprompter operator who bet on the exact words his boss would speak.
For months, James Perez knew what Donald Trump would say before the microphones went live. He didn’t leak the information. He didn’t sell it. He turned it into a $100,000 profit on Kalshi, the CFTC-regulated prediction market that prides itself on its compliance-first ethos.
The ledger remembers what the hype forgets. And this ledger tells a story of systemic vulnerability in the fastest-growing sector of crypto-adjacent finance: event contracts.
Perez’s trades were not anonymous flash loans. They were place-and-hold bets on "Mentions" contracts—wagers that specific words like "China" or "inflation" would appear in Trump’s speeches. He placed them hours before rallies, then watched the returns accumulate as the teleprompter scrolled. On March 4, 2025, he even withdrew a position midway through the State of the Union, realizing his inside knowledge gave him an edge the market couldn’t price.
Kalshi’s monitoring team flagged the pattern. They saw the same account winning disproportionately on word-based contracts. They escalated to the CFTC. Perez is now in settlement negotiations—likely to forfeit his gains and face a trading ban. But the damage to the narrative of fair, transparent prediction markets is already done.
Context: The Unseen Battlefield of Information Advantage
Kalshi launched in 2021 as America’s first regulated prediction market, a stark contrast to on-chain platforms like Polymarket that operate without KYC or centralized oversight. Under CFTC oversight, Kalshi collects employer information, screens for suspicious activity, and maintains a dedicated compliance team. It was supposed to be the safe, boring, regulated alternative.
Yet Perez exploited the very structure of the platform. He didn’t hack code. He exploited a human information advantage—access to unannounced speech content. The "Mentions" market, where users bet on whether specific phrases appear in speeches, is particularly susceptible. It’s a market where the payoff depends on timing: the moment a word is uttered, the contract resolves instantly. For someone who knows the script, it’s as close to a guaranteed profit as exists in prediction markets.
This isn’t a technical flaw in Kalshi’s smart contracts. It’s a structural flaw in how prediction markets define "fairness." In traditional finance, insider trading is illegal because material non-public information distorts prices. In prediction markets, the same principle applies. But the enforcement mechanism is weaker. A president’s speech is a singular event. There’s no pattern of form 4 filings or earnings calls to cross-reference. The only signal is behavioral: Who wins too often?
Bridging the gap between code and community means building detection systems that anticipate human behavior, not just algorithmic anomalies. Kalshi’s monitoring team did that. But the fact that Perez operated for months and placed multiple high-value bets before detection raises uncomfortable questions about how many other insiders are under the radar.
Core: The Anatomy of the Insider Trade
Based on my own experience auditing ICO tokenomics during the 2017 boom, I’ve learned that the most dangerous vulnerabilities are often the simplest. Perez didn’t use a mixer. He didn’t split his bets across multiple accounts. According to sources familiar with the investigation, he used his personal Kalshi account, which was linked to his White House employment through the employer disclosure form Kalshi started requiring last month.
Here’s the timeline: - February 2025: Perez opens a Kalshi account, discloses employer as "White House Office of Communications." Standard KYC. - March 4, 2025: Trump rallies in Wisconsin. Perez places a $15,000 bet on the word "China" appearing in the speech. It does. He profits $21,000. - March 7, 2025: State of the Union. Perez places multiple small bets on "economy," "border," and "inflation." He withdraws $8,000 mid-speech after Trump says "economy" early. Another profit. - April 2025: Kalshi’s risk team notices an unusual win rate on Mentions contracts from this account. They manually review his trading history and spot the pattern: all bets placed within 24 hours of a speech, all with high probability of success. - May 2025: Kalshi reports to CFTC. Perez is placed on administrative leave from the White House. Settlement negotiations begin.
What’s remarkable isn’t the mechanism—it’s the failure of deterrence. Kalshi had already introduced employer disclosure after the first FBI cases in March 2025, where a Google employee and a Venezuelan government insider were charged with trading on non-public information. Yet Perez still saw the opportunity as worth the risk.
Narratives move markets faster than blocks, but incentives move individuals faster than narratives. Perez likely calculated that the chance of detection was low because his edge was fleeting. A speech happens quickly. By the time the CFTC investigates, the event is over, and the profit is realized. He was only caught because Kalshi’s team was diligent enough to connect the dots across multiple events.
Technical angle: Why prediction markets are harder to police than traditional exchanges
In equity markets, insider trading detection relies on correlation between non-public information and trading patterns. But prediction markets have thinner liquidity and less historical data. A single whale can move prices. A pattern of winning on low-volume contracts like "Does Trump say ‘China’?" is not an anomaly—it could just be a well-researched bettor.
Kalshi’s advantage is its centralized architecture. It can freeze accounts, reverse trades, and cooperate with authorities. Polymarket, running on Polygon smart contracts, cannot easily intervene. If a similar insider traded on Polymarket, the platform would need a DAO vote to freeze funds—a process that takes days, not hours. The last time Polymarket attempted to block trading on a controversial contract (the "Turkish elections" in 2023), the community backlash was intense.
This scandal highlights a paradox: centralized prediction markets are more capable of enforcing fairness, but their policing mechanisms create a honeypot for bad actors who want to exploit the gap between detection and enforcement. Kalshi’s compliance team is the moat. But a moat only works if it’s constantly patrolled.
Human story: The operator’s perspective
I’m told Perez was not a sophisticated trader. He didn’t understand options or yield curves. He simply realized that his job gave him a unique window into Trump’s communication strategy. He saw the loose-lipped culture of the White House communications team, where draft speeches were shared casually via email. He didn’t steal classified documents; he stole timing.
Empathy in the algorithm doesn’t mean excusing the violation. It means understanding that the human desire for easy money can overwhelm ethical boundaries. Perez may have justified his actions as a "free market" exploitation of an information asymmetry. He wasn’t wrong—legally. The CFTC had not yet clearly defined prediction market insider trading as a violation. But Kalshi’s terms of service forbid using non-public information. He crossed that line.
His story is a cautionary tale for every government employee, every journalist, every person who holds a temporary informational edge. The ledger remembers. Even if the market doesn’t punish you, the platform will.
Contrarian: Why this scandal might actually strengthen Kalshi
At first glance, this news seems devastating for Kalshi. A trusted, regulated platform was used for insider trading. Trust is brittle. But the contrarian read is more nuanced: Kalshi’s response—active detection, swift reporting, transparent settlement—positions it as the safest harbor in a stormy regulatory sea.
Consider the alternative. If this had happened on a decentralized exchange, the insider would have been anonymous. The CFTC would have no jurisdiction. The platform would either ignore it or fail to act. Users would lose confidence in the fairness of all prediction markets. Instead, Kalshi proved that a regulated platform can catch bad actors and cooperate with authorities. It’s the equivalent of a bank reporting a robbery. Yes, the robbery happened. But the response prevents future ones.
Culture is the new collateral. Kalshi’s decision to require employer disclosure even before this scandal—and its willingness to update terms proactively—shows a culture of compliance that will attract institutional traders. Hedge funds and pension funds don’t want to trade against insiders. They want a level playing field. Kalshi is building that field, one rule change at a time.
Furthermore, this event could accelerate regulatory clarity. The CFTC now has a concrete case to define prediction market insider trading. That definition will likely include specific prohibitions on trading based on pre-release speech content. Once the rules are clear, the compliance cost for platforms like Kalshi becomes predictable. And predictability attracts capital.
Polymarket, on the other hand, faces a harder path. Its anonymity is its strength and its vulnerability. Without KYC, it cannot prevent insiders from using multiple wallets. Without a central authority, it cannot quickly freeze assets. The next Perez trade on Polymarket will likely go undetected until it’s too late. Regulators will notice. The result could be a bifurcation: regulated prediction markets become the prime venues for high-value contracts, while decentralized markets remain for micro-bets and niche events.
The sprint ends, but the chain remains. The question is which chain—the regulatory one, or the blockchain one?
Takeaway: What to watch next
The CFTC’s settlement with Perez will set the standard. If it only requires disgorgement of profits plus a cease-and-desist, the message is: test the boundaries, pay back if caught. That’s a weak deterrent. If the CFTC imposes a penalty—even a modest one—it signals that prediction market insider trading carries real cost. That would be a stronger signal to potential insiders.
Watch for Kalshi’s next compliance upgrade. I expect them to introduce real-time speech monitoring APIs that correlate trading activity with known speech schedules. They may also require pre-trade attestation for Mentions contracts—a pop-up asking "Do you have non-public information about this speech?" Simple, but effective as a legal barrier.
Decentralization is a mindset, not just a metric. In this story, the decentralized mindset is about fairness, not technology. The ledger may remember, but the market will only thrive if the rules evolve faster than the exploiters. This scandal is a stress test. The industry will either learn to police itself internally, or regulators will impose external controls that choke innovation.
The real trade isn’t on a word contract. It’s on the future of prediction markets themselves. And the odds are shifting.
