Federal authorities on Thursday charged a U.S. Army soldier who took part in the operation to capture Nicolás Maduro with using classified information to reap more than $400,000 from bets on the former Venezuelan leader’s ouster.

Gannon Ken Van Dyke was charged in Manhattan federal court with commodity and wire fraud, according to the Justice Department. Prosecutors alleged that Van Dyke, a master sergeant with the U.S. Army Special Forces, misused information about the operation to make well-timed bets on Polymarket that Maduro would lose power by the end of January. Maduro was captured by the U.S. in early January.

Manhattan U.S. Attorney Jay Clayton said that Van Dyke tried to cover his tracks after making the trades by sending most of his gains to a foreign cryptocurrency account and asking Polymarket to delete his account. Van Dyke, 38 years old, is an active-duty soldier who had been stationed at Fort Bragg in North Carolina.

Traders have used prediction markets such as Polymarket to profit from a volatile policy environment under President Trump —from sudden shifts in tariffs to military strikes. Clayton said earlier this year that he expected to bring enforcement actions over trades that appeared to benefit from inside information.

NEWSLETTER TABLE TALK

Never miss a story.
Subscribe now.

The most important news & topics every week in your inbox.

Van Dyke took part in the planning and execution of the mission, dubbed Operation Absolute Resolve, prosecutors said. On Dec. 8, he received a classified briefing about the plan to arrest Maduro and signed a nondisclosure agreement providing that he would never reveal the information “by writing, word, conduct, or otherwise,” according to court records.

“The defendant allegedly violated the trust placed in him by the United States government by using classified information about a sensitive military operation to place bets on the timing and outcome of that very operation, all to turn a profit,” Clayton said. “That is clear insider trading and is illegal under federal law.”

Prediction markets offer traders an opportunity to bet on a range of financial, political and sports-related outcomes. They are regulated under U.S. laws that govern how derivatives are traded, but Polymarket also operates an offshore platform that allows traders to bet on a wider range of contracts.

Polymarket said it has policies designed to combat insider trading. “When we identified a user trading on classified government information, we referred the matter to the DOJ & cooperated with their investigation,” the company said on X.

Polymarket has a data partnership with Dow Jones, the publisher of The Wall Street Journal.

The case against Van Dyke comes one day after Kalshi, a competitor to Polymarket, announced that it had fined and suspended three congressional candidates for betting on their own races.

Maduro and his wife were captured during a nighttime operation that was the culmination of months of secretive planning. The mission was led by elite special-operations forces and was supported by every branch of the U.S. military, as well as intelligence and law-enforcement agencies.

While Trump had railed for months against the Maduro regime, there was little in the immediate run-up to suggest to the general public that such drastic action was imminent.

Van Dyke allegedly made 13 bets between Dec. 27 and Jan. 2 on the U.S. going after Maduro. The concentrated nature of the bets, coming from a brand-new account, fueled suspicions that someone with knowledge of the operation’s planning had made them.

Hours after Van Dyke’s last wager, prosecutors said, the U.S. Special Forces entered Caracas and apprehended Maduro and his wife at their residence. Van Dyke spent about $33,000 in long-shot contracts that netted him a profit of $409,000, according to the Justice Department.

The Commodity Futures Trading Commission, a regulatory agency that under the Trump administration has championed the value of prediction markets, also sued Van Dyke in civil court.

Write to Dave Michaels at dave.michaels@wsj.com and Sadie Gurman at sadie.gurman@wsj.com