Trump speech bets have landed longtime teleprompter operator Gabriel Perez on unpaid administrative leave after he allegedly used advance knowledge of prepared remarks to win more than $100,000.
According to ABC News, White House Press Secretary Karoline Leavitt announced the disciplinary action Thursday afternoon after the outlet published its investigation into Perez’s prediction market activity. Leavitt said she spoke directly with Trump about the allegations and that he considered the situation a “disgrace.” In the audio, she appears to say that Perez was placed on paid leave. However, later reports clarified that the dismissal is actually unpaid. She said Trump personally decided to place Perez on leave.
Leavitt said she was not aware of any other White House employees who had made similar trades. White House spokesperson Davis Ingle also addressed the controversy while responding to the outlet’s questions.
“The White House has strict ethics guidelines that we expect all staffers and officials to follow,” Ingle said, according to ABC News.
Now, the details of what Perez allegedly bet on are what make this story especially wild. According to reports, Kalshi operates a category of prediction contracts called “Mentions” markets. Those markets allow users to put money on whether a person will say a particular word, repeat a certain phrase or discuss a specific subject during a public speech.
Perez was not allegedly betting on whether Trump would give a good speech or whether an event would attract a large audience. The Trump speech bets reportedly focused on the exact language that might come out of Trump’s mouth.
Federal investigators believe Perez placed wagers connected to more than a dozen Trump speeches during a three month period. Those appearances reportedly included a December primetime address, remarks at the World Economic Forum in Davos, Switzerland, the February State of the Union address and a March Medal of Honor ceremony.
Perez has operated Trump’s teleprompter since the 2016 campaign. He was often one of the last people to review prepared remarks before Trump delivered them publicly. Perez also reportedly received last minute edits directly from Trump, giving him access to words, phrases and talking points before viewers, journalists and other prediction market traders could hear them.
That kind of access could provide a serious advantage in a market where the entire wager depends on whether one specific word or subject appears.
Investigators discovered instances in which Perez allegedly began backing out of bets while Trump was speaking. Those exits reportedly occurred when Trump skipped sections of prepared remarks containing words that Perez had predicted would be mentioned. In other words, investigators believe he may have watched the speech unfold, noticed Trump had gone off script and attempted to limit his losses before the market closed.
Trump has openly acknowledged that he does not always stick to the teleprompter.
“You know, when you go up here, you take a big chance, especially me because I go off teleprompter about 80% of the time,” Trump said during January remarks to the Detroit Economic Club.
Investigators believe that Detroit appearance was also among the speeches connected to Perez’s alleged wagering activity.
Kalshi reportedly identified the unusual Trump speech bets through its internal monitoring system and contacted the Commodity Futures Trading Commission, which regulates the platform.
“Our surveillance team promptly flagged and referred these trades to the CFTC, and we are cooperating and assisting regulators,” Kalshi head of enforcement Bobby DeNault said in a statement.
Perez participated in an interview with federal regulators and acknowledged making some of the trades. The outlet reported that the CFTC contacted federal prosecutors in Manhattan during the investigation, but prosecutors declined to open a criminal case. Perez is reportedly discussing a possible civil settlement with regulators.
Under the terms being discussed, Perez could be required to return his alleged profits and agree not to engage in similar trading activity, according to ABC News. The CFTC declined to comment publicly on the investigation.
Kalshi prohibits users from placing wagers with confidential information obtained through their jobs. The company recently began requiring users to disclose their employers as part of an effort to detect suspicious trading.
“If you have information by virtue of your job or your employment, something that you have a legal duty surrounding, and you have an obligation not to take that, misappropriate it for yourself,” DeNault said in May.
The CFTC has been making it clear that prediction markets are not an insider trading free zone. According to a February CFTC advisory, Kalshi previously investigated an editor who allegedly traded on the expected contents of YouTube videos before those videos were published. Kalshi ordered that trader to surrender profits, pay a penalty and accept a two year suspension.
In March, CFTC Enforcement Director David Miller said the agency considers insider trading in prediction markets illegal when someone uses material nonpublic information in violation of a duty owed to the source of that information. According to Miller’s official remarks, the agency is particularly concerned about government employees using information obtained through their positions to place wagers.
The Trump speech bets investigation also arrives as federal prosecutors pursue their first major criminal cases involving alleged prediction market insider trading.
According to the Department of Justice, Army soldier Gannon Ken Van Dyke was charged in April with allegedly using classified information about a military operation involving Nicolás Maduro to place Polymarket wagers. Prosecutors allege that Van Dyke earned approximately $409,881. He has pleaded not guilty, and the charges remain allegations.
According to a separate Department of Justice announcement, Google software engineer Michele Spagnuolo was charged in May with allegedly using confidential company information to earn approximately $1.2 million through wagers involving Google search data. Spagnuolo has pleaded not guilty, and the allegations have not been proven in court.
Trump has offered mixed opinions about the booming prediction market industry.
“Well the whole world, unfortunately, has become somewhat of a casino, and you look at what’s going on all over the world in Europe and every place they’re doing these betting things. I was never much in favor of it. I don’t like it conceptually, but it is what it is,” Trump told reporters.
The White House’s decision to remove Perez from duty without pay gives the controversy immediate consequences, even as the regulatory investigation remains unresolved. Perez allegedly knew which words were loaded into the teleprompter, placed money on whether Trump would say them and earned more than $100,000 before the system finally flagged the pattern.
