On the afternoon of Friday, February 27th, as Donald Trump was flying to Corpus Christi, Texas, to deliver a speech on energy, he signed an order authorizing Operation Epic Fury, the military assault on Iran that is now in its second week. At 2:30 A.M. the following morning, Trump announced, on Truth Social, that the attack had begun. Hours later, Bubblemaps, a Paris-based company that tracks data on crypto transactions and presents it in visual form, posted a message on X: “Six suspected insiders made $1.2M betting on a US strike on Iran.”

The data firm pointed to bets placed in six accounts on the prediction platform Polymarket, which enables people to wager anonymously using crypto on practically anything, including sports events, election outcomes, which movie will win Best Picture at the Oscars, whether Attorney General Pam Bondi will be next to leave Trump’s Cabinet, and whether the U.S. government will confirm that aliens exist before the end of this year. Since December, Polymarket has been taking bets on whether the U.S. would strike Iran by a certain date: the “suspected insiders” identified by Bubblemaps had predicted that it would happen by February 28th. Most of these accounts were funded in the final twenty-four hours before the bombing began, and they hadn’t placed bets on any other events, the firm pointed out. The biggest winner staked $60,812.42 and won $499,863.81.

Not surprisingly, these revelations stirred controversy, especially after it emerged that another Polymarket account, pseudonymously named Magamyman, had placed even bigger bets on a U.S. strike and on whether Ayatollah Ali Khamenei, who was killed in the U.S.-Israeli attacks, would no longer be Supreme Leader at the end of March. “It’s insane this is legal. People around Trump are profiting off war and death,” the Democratic senator Chris Murphy said on X.

The White House denied that anybody connected to the President had placed the bets, and Murphy didn’t provide any new information to substantiate his claims. But the timing of the wagers added to suspicions that certain individuals are using prediction markets to profit from advance knowledge of U.S. military actions. At the start of January, another anonymous account on Polymarket gambled more than thirty-two thousand dollars on the possibility that Venezuelan President Nicolás Maduro would be out of office by the end of the month. The very next day after the bet was placed, U.S. commandos seized Maduro and flew him to the United States. Whoever was behind the account made more than four hundred thousand dollars.

Given the rampant corruption surrounding the Trump Administration, it’s reasonable to wonder whether the well-timed wagers are coming from people directly connected to the White House. But there are other possibilities, too. An analysis of Polymarket data by the Times showed that on Friday, February 27th, more than a hundred and fifty accounts each bet more than a thousand dollars that a U.S. strike on Iran would take place by the next day. Preparing to launch such an attack surely involved informing a large number of people, including members of the U.S. and Israeli military chain of command, well-connected diplomats, and ordinary soldiers or airmen who got called into action. Earlier this year, Israeli authorities charged two people, a civilian and a military reservist, with using classified information to place bets on Polymarket. “Part of the problem is that the number of people who are insiders is very large,” Dennis Kelleher, the president and co-founder of Better Markets, a Washington-based public-interest group, told me. “We are incentivizing lots of people to use their knowledge of classified information for their own benefit.”

The basic problem, Kelleher said, is the way in which prediction sites are regulated and policed—or, rather, not regulated and not policed. Despite Murphy’s statement about the bets on the Iran strikes being legal, using classified information to make a profit is potentially a serious crime in the U.S. Moreover, in 2011, the Commodity Futures Trading Commission (C.F.T.C.), which has jurisdiction over “events contracts” under the Commodity Exchange Act, issued a ruling that contracts involving terrorism, assassination, or war were contrary to the public interest and prohibited on a U.S. trading exchange.

In practice, however, it’s relatively simple to skirt these laws and regulations by exploiting the anonymity of prediction platforms, particularly Polymarket. And there isn’t much reason to fear being caught. Although Polymarket is based in New York—it was founded in 2020 by Shayne Coplan, a young entrepreneur who dropped out of N.Y.U. to pursue crypto ventures—its main prediction platform operates offshore. This structure means that it isn’t subject to the know-your-customer laws, which require most U.S.-based financial institutions, including crypto exchanges, to ascertain the real identities of their customers and to monitor their transactions. The fact that bets on Polymarket are placed on the blockchain, which is extremely difficult to penetrate, provides another layer of protection for would-be profiteers.

Technically, it is still a violation of American securities laws for a U.S. resident to place a bet on Polymarket’s offshore market, but many people get around this by installing a virtual private network, or V.P.N., which masks their I.P. address, and using it to connect their crypto wallet to the site. Can you blame them when the President has made a fortune by promoting his family’s crypto companies and has shown few qualms about crypto’s criminal underbelly? Last year, he pardoned Ross Ulbricht, who was serving life in prison for running Silk Road, a dark-web marketplace frequented by drug dealers, and Changpeng Zhao, the billionaire founder of the Binance crypto exchange, who had pleaded guilty to failing to maintain effective anti-money-laundering programs. In the current Administration, U.S. regulators and law-enforcement agencies seem to largely turn a blind eye to crypto violations. During the Biden Administration, the Justice Department and the C.F.T.C. launched investigations into whether Polymarket was illegally accepting bets from people in the U.S. But last summer the two agencies informed the company that they had closed their investigations and would take no further action.

A couple of months after this news emerged, Polymarket announced that Donald Trump, Jr., was joining the company’s advisory board, and that 1789 Capital, the venture-capital outfit at which he is a partner, had made a “strategic investment” in the firm. Donald, Jr., also acts as a “strategic adviser” to Kalshi, which is another prediction market, but one that operates in the U.S. Earlier this year, the C.F.T.C. withdrew a Biden-era effort to prevent Kalshi from accepting bets on the outcome of sports events and political elections. The agency’s Trump-appointed chair, Mike Selig, said last week that it is preparing new rules for prediction markets, but most observers expect the agency to stick with its light-handed approach.

Some economists look fondly on prediction markets, which they view as a tool for aggregating dispersed information and making better predictions. Last year, Rajiv Sethi, an economist at Barnard, co-authored a paper which found that prediction markets did a better job of predicting the outcome of the 2024 Presidential election than statistical models did, although they were less successful in projecting down-ballot races. When I called up Sethi last week, he pointed out that prediction markets also indicated that Joe Biden would drop out of the race weeks before it happened. “Markets have a capacity to forecast continuously when you are in uncharted waters and history is not a good guide,” he said. “In some circumstances, that can be socially useful.”

But even as Sethi defended the existence of prediction markets, he emphasized that crypto-based platforms need proper oversight, including effective know-your-customer rules that are properly enforced. If such a system was in place, he said, “many of the problems could be mitigated.” One option could be to ban companies like Polymarket from operating in the U.S. if their offshore markets refuse to abide by U.S. regulations. (Last year, Polymarket bought a U.S.-based trading platform, and it is now operating a beta version of a U.S. site, which is focussed on sports betting.) Another possibility, Sethi said, is having a third party carry out the identity-verification process. Speaking to the over-all challenge of taming the prediction market, he said, “You need to have the political will, and you need to be clever. But you can do it.”

Kelleher agrees. Citing efforts to counter money laundering and other financial crimes, he pointed out how the U.S. authorities had used their capacity to bar financial institutions from having access to U.S. banks, and other U.S. financial services, as leverage to change behavior. “If you have large quantities of money crossing borders, it’s very hard for it not to encounter the U.S. financial system somewhere in the world,” he said. But for any crackdown to have a decent chance of succeeding, regulators and law enforcers would need the enthusiastic backing of the Administration, and right now that is nowhere in sight. “When the President says anything goes, and his son is on the advisory boards of these companies, it’s the ultimate conflict of interest,” Kelleher said. “The regulators are effectively prohibited from acting in the public interest.”

In the absence of any guardrails, Polymarket is still offering bets connected to the war on Iran, including wagers on who the next Supreme Leader will be, whether the Iranian regime will fall by the end of this month, or a later one, and whether U.S. forces will enter Iran by certain dates. As of Sunday afternoon, the prices quoted on the site implied that the probability of U.S. troops going in by the end of March is fifty-two per cent, and this figure rises to sixty-eight per cent for entry by the end of this year. In the tech-powered and unregulated financial capitalism that Polymarket represents, war and destruction is just another tradable commodity. ♦