You can bet on almost anything online these days, but trying to outsmart Donald Trump’s notoriously unpredictable speaking style using his own teleprompter script takes a special kind of audacity.
That is exactly what Gabriel Perez, Trump’s longtime teleprompter operator, allegedly did.
Perez, who has worked with Trump since 2016 and pulled a comfortable $175,000 annual salary as a deputy assistant and technical adviser, now finds himself on unpaid administrative leave. The White House has made it clear he won’t be returning.
His crime? Allegedly leveraging his front-row access to the president's prepared remarks to rake in over $100,000 on the prediction market Kalshi.
Specifically, Perez targeted "mention markets"—highly speculative pools where users wager on whether the president will utter specific words or phrases during a speech.
But this is not just a story about a rogue staffer trying to make a quick buck. It exposes a massive, systemic vulnerability in the booming world of prediction markets, where the line between "shrewd analyst" and "insider trader" is rapidly blurring.
The Perfect Inside Job
To understand how Perez allegedly pulled this off, you have to understand how prediction markets operate.
Unlike traditional sportsbooks, platforms like Kalshi allow users to buy "yes" or "no" contracts on real-world events. Ahead of major addresses—like the State of the Union or speeches at the World Economic Forum in Davos—Kalshi routinely lists contracts on whether the president will say words like "Hormuz," "rigged election," or "fake news".
The odds fluctuate wildly in real-time as the speech approaches. If you have the actual text of the speech sitting on a screen in front of you before the president even takes the stage, you don't need to guess. You already know the answers.
According to investigators, Perez placed bets on more than a dozen of Trump's speeches.
But betting on Trump isn't as simple as reading a script. Trump is famous for tossing his prepared remarks out the window and riffing off-the-cuff.
This is where the scheme gets wild. Sources close to the investigation revealed that when Trump began to veer off-script mid-speech, Perez would frantically cancel his active bets in real-time to mitigate his losses.
Ultimately, Kalshi's internal surveillance team noticed highly irregular trading patterns. When they dug into the account, they discovered the holder was a federal employee working directly on the teleprompters. Kalshi immediately froze the account—locking down roughly $90,000 in profits—and flagged the activity to the Commodity Futures Trading Commission (CFTC).
Prediction Markets Have an Inside Information Problem
For years, prediction markets have defended their existence by arguing they act as "truth machines," aggregating public intelligence to forecast events more accurately than pundits or polls.
But as these platforms scale to hundreds of millions of dollars in volume, they are attracting a different crowd: people who don't want to forecast the future because they already control it.
The Perez scandal is not an isolated incident. It is part of a broader, deeply concerning trend of insiders treating prediction markets like a personal piggy bank:
- The Special Forces Soldier: In April, federal authorities charged a Special Forces soldier involved in the raid targeting Venezuelan President Nicolás Maduro. He allegedly used confidential military intelligence to net $400,000 on prediction markets.
- The Tech Engineer: A Google software engineer was recently accused of using internal data on search trends to clean up on Polymarket.
- The Politicians: In May, Kalshi had to suspend and fine three congressional candidates who were caught wagering on their very own elections.
When the people making the news are also the ones betting on it, the entire premise of an open, fair marketplace collapses.
What Happens Next
Perez is currently in settlement talks with the CFTC. While the White House issued strict warnings to staff in March against using non-public information to trade on prediction markets, the legal system is still catching up to this new frontier of financial misconduct.
The Department of Justice is actively pursuing its first-ever insider trading prosecutions tied to these platforms.
If you are thinking of dipping your toes into prediction markets, understand the reality of the game you are playing. While platforms like Kalshi and Polymarket are rapidly deploying stricter surveillance tools, you are still trading in unregulated or newly regulated waters.
If you do decide to trade, stick strictly to macroeconomic data or broad geopolitical events where a single rogue actor cannot easily manipulate the outcome. Avoid hyper-specific "mention" or event contracts where a single staffer behind a curtain can instantly render your analysis worthless.
The house always wins, especially when the house is the one typing the script.
