In mid-March, Emanuel Fabian, a military correspondent for the Times of Israel, contacted Israeli police over threats and harassment he was facing because of his reporting on an Iranian missile strike. However, the threats didn’t come from political or religious radicals. They came from bettors on Polymarket who demanded Fabian alter his report to state the missile was intercepted, a distinction that would have changed the financial outcome for those wagering on the popular prediction market. High stakes were involved, as the specific contract for an Iranian strike on that date had exceeded $15 million in trading volume.
The bettors menacing Fabian were one part of a much larger wager market around the US-Israeli war with Iran that has illustrated just how big prediction markets have become — and dramatically underscored some of the broader questions raised about the markets. On the Polymarket platform alone, more than $529 million was staked on the timing of the attacks. Some of that activity was likely from legitimate economic actors looking to hedge against the fallout from the war, but a surge of well-timed bets just before the first strikes raised suspicions of insider trading and the prospect of speculators profiting off death and misery.
Prediction markets, once a niche market oddity and an academic curiosity, have entered the mainstream over the last few years and are now increasingly making headlines, often for all the wrong reasons. They went from crude experimental tools used by economists and political scientists to forecast election results or macroeconomic outcomes, to giant platforms like Kalshi and Polymarket with billions of dollars in trading volume and growing public visibility.
Today, prediction markets function as real-time aggregators of expectations about politics, economics, geopolitics and technology, but they also offer a space where users can bet on literally any event with a measurable outcome one can think of, from the Iranian missile strike on Israel to more trivial or esoteric questions: whether it will rain in Los Angeles tomorrow, if Jesus Christ will return this year or if the US government will confirm that aliens exist by 2027.
The growth of these platforms has been meteoric. Kalshi, currently the largest regulated prediction market exchange in the US, has seen total trading volume surge 12-fold in a single year to reach $24 billion in 2025. On this year’s Super Bowl Sunday alone, single-day trading on Kalshi exceeded $1 billion. At the same time, decentralized or predominantly offshore platforms like Polymarket have attracted global participation, particularly during major political events such as US elections. These platforms often rely on cryptocurrency infrastructure and are accessible to a broad international user base.
A better way to forecast and hedge
Advocates of prediction markets often argue that they help close information gaps by aggregating dispersed knowledge into prices. According to this line of argument, the fact that participants are putting money behind their views — unlike an opinion poll or a survey — gives them added value. The structure of these markets transforms personal opinions or predictions into actual financial positions (or, to be precise, binary “event contracts” with possible outcomes of “yes” or “no”). The opinions of participants should be more reliable data points, since they are actually putting money behind their beliefs. If true, prediction markets can play an important social and economic role by providing an unbiased “truth serum” for public policy and global risk assessment. In an era of fragmented media and extreme algorithmic bias skewing the information each of us consumes, these markets offer a neutral, consensus-driven view on the actual likelihood of geopolitical events, policy developments or the trajectory of the economy.
The intellectual foundation of prediction markets lies in a well-established concept from economics and social science: collective forecasting largely outperforms individual experts. When many participants independently assess information and trade based on their expectations, the resulting price aggregates this broad range of different perspectives. This phenomenon is often described as the “wisdom of the crowds.” The concept dates at least as far back as 1907, when Francis Galton — a British statistician who was the cousin of Charles Darwin — analyzed around 800 guesses of an ox’s weight. Galton found that, although individual guesses varied wildly, the median estimate was almost exactly accurate. The experiment has been replicated countless times since then, using different scenarios (one common variant being guessing at the number of marbles in a large jar) with similar results.
Prediction markets provide a mechanism to convert that collective knowledge into measurable probabilities, as traders may possess different pieces of relevant information through local knowledge, specific industry insight or expert interpretation of public data. When they trade against each other, prices adjust until the market reflects the best estimate based on all available information.
‘Prediction markets expand the set of tools available for managing uncertainty and risk in areas where conventional hedging tools do not exist or do not suffice.’
Like in financial markets, algorithmic traders looking for arbitrage can, interestingly, help aggregate information across multiple prediction market platforms as well. Bots constantly scan for prices, looking for discrepancies between prediction markets as opportunities for low-risk profit, a possibility that pops up especially for more niche bets. If a niche market on Polymarket puts the likelihood of “Yes” on some possible event at 60¢, but a similar market on Kalshi or a sports betting site implies it should be 70¢, a bot will instantly buy the “Yes” shares on Polymarket and sell the other to “arb” the difference. Much like currency arbitrage, this kind of trading activity also helps close the gap of information and improve market efficiency.
Event contracts also serve another, very practical purpose: Because they pay out when a specific outcome occurs, they can function as a form of insurance against political, regulatory or economic risks that are otherwise difficult to hedge through traditional financial instruments. Oil futures or shipping insurance might have provided a straightforward way for some companies to protect against the fallout from war in Iran. But plenty of other economic actors are paying steep costs from the conflict in ways that would not have been as easy to mitigate in the insurance or futures markets.
There are plenty of other scenarios where prediction markets provide a very useful vehicle for hedging. Importers concerned over the possibility of government tariffs could buy contracts that pay out if such tariffs are implemented. If their fears are well-founded and the tariffs come to pass, the financial payout from the prediction market could help offset the losses. Similarly, businesses dependent on regulatory approvals, election outcomes, or court rulings could hedge the financial impact of potential unfavorable decisions by taking positions that anticipate them. In this light, prediction markets expand the set of tools available for managing uncertainty and risk in areas where conventional hedging tools do not exist or do not suffice.
…or an insider’s casino…
One major concern is the risk of insider trading — that the markets provide an alluring and tough-to-trace way for well-informed bettors to use non-public information for personal gain. There have been numerous well-documented suspicious cases already. In January, Portugal banned Polymarket after suspicious betting patterns saw over 4 million euros wagered during the presidential election as odds shifted dramatically two hours before official results — but just as non-public exit polls began circulating confidentially, raising serious suspicions over leaked results. In another high-profile example, a recently created account bought a large contract predicting that Venezuelan President Nicolás Maduro would be removed from power just hours before his capture by US special forces, leading to a payout of almost $440,000.
Traditionally, insider trading refers to transacting in a security while in possession of material non-public information (“MNPI”). For example, if an employee trades his company’s stock while being privy to MNPI, such as knowledge of an upcoming merger or earnings information, he could possibly be guilty of insider trading. The law governing these scenarios is relatively straightforward and has been clearly defined over decades of jurisprudence. Prediction markets, however, do not fall under securities law in the United States; they are so far treated as futures contracts by regulators.
For example, in the US, Kalshi is regulated by the US Commodity Futures Trading Commission (CFTC) and not the Securities and Exchange Commission (SEC), and it is considered a Designated Contract Market (DCM), which is a financial exchange designated to trade futures, swaps and/or options on commodities. Polymarket is a bit more complex, as it has a hybrid regulatory status: its main product operates offshore, while it has been building a separate regulated pathway for US operations. Regulators argued it required a license as a derivatives exchange and CFTC supervision, which it did not have, so at one point, US access was restricted and Americans were geoblocked. To work around this, the platform then acquired QCX (a CFTC-licensed derivatives exchange) to obtain the necessary regulatory framework. After that acquisition, the CFTC granted approvals allowing a regulated US version of Polymarket, though most of its user base remains international.
‘The insider’s action thus transmits private information indirectly into the public market.’
The problems that prediction markets pose to regulators are novel, and when it comes to establishing harm or unfair advantage from insider trading or using sensitive information in trades, the lines are particularly blurry. For example, even if insider trading rules were to apply, what happens if an executive assistant comes across the script for her company’s upcoming earnings call and uses that to purchase a highly specific contract that the company’s earnings script will contain a particular number of words, or that it will repeat the word “future” or “quarter” a certain number of times? The employee is enriching herself with privileged information and participating in a market on the basis of information asymmetry. But is that directly harming the company and its shareholders, since her bet has nothing to do with the actual substance of the script or non-public financial results?
The implications become more concerning when it comes to bets involving military actions or classified government decisions, not least because patterns of insider trading could expose confidential information to market observers. Contracts predicting whether a country will launch a military strike, sign a ceasefire or escalate a conflict may attract traders who have access to privileged information within governments or security institutions. When these traders act on that information and place bets that are publicly visible, prediction markets can become indirect channels that not just allow them to profit from confidential knowledge but also disseminate said knowledge.
The bet on Maduro’s capture stood out. So did a similar well-timed bet from someone using the account name “Magamyman” who won $553,000 predicting when Iran’s supreme leader, Ali Khamenei, would be killed in an airstrike. Similar big trades that are likely based on insider knowledge usually stand out. Large positions against the prevailing consensus that previously had very low probabilities, often from newly created accounts and betting on highly specific events and timing, all draw the attention of other traders. And scrutiny of prediction markets to spot insider trades seems likely to increase, given the apparent frequency of such cases. Other participants may start interpreting such unusual positions as a signal and start buying the same outcome — while discerning some of the underlying top-secret information. The insider’s action thus transmits private information indirectly into the public market.
…or a force for corruption?
This becomes more troubling when we consider the incentives that prediction markets provide not only to trade on outcomes — with or without insider information — but to actually influence the outcomes themselves. The Israeli journalist who found himself threatened over his report on an Iranian missile strike wrote later that some of the traders offered to share their earnings with him in exchange for changing his story. He was not tempted by the offer, but others may be less scrupulous — or even place bets themselves when they are in a position to trigger either “Yes” or “No” on the contracts. In a more lighthearted example from the 2026 Grammy Awards, host Trevor Noah referenced prediction markets in his opening monologue. The comedian deliberately said the word “potato” on stage, then jokingly added: “If you had me saying ‘potato’ on Polymarket, you just made a ton of money — so congratulations Noah_22.”
It’s not funny, though, when the bets involve precise military actions, terrorist attacks, political instability or even assassinations — and especially when the bettors hold the power over actually carrying out those actions, and not just reporting on them. This can obviously create deeply uncomfortable ethical incentives, as the potential for large financial gains tied to such outcomes raises concerns about the possibility of attempting to influence these events, or perhaps even cause them. In the wake of the US and Israel strikes on Iran, Polymarket removed a market that allowed traders to bet on the likelihood of a nuclear detonation around the world. The bet was first listed late last year, but drew little attention until the current conflict broke out, with daily trading volume reaching almost $244,000 before it was taken down amid widespread criticism.
However, it doesn’t have to be about something as difficult to engineer as a nuclear strike for a bet to be dangerous. It could be about the timing of a workers’ strike, where a trader with meaningful influence over union scheduling — or a trade union official themselves — could stand to profit by nudging leaders toward certain strike dates. A market predicting unrest in a particular city on a specific day could incentivize organizers or provocateurs to manipulate the timing of events. Bets on high-profile legal outcomes could incentivize bribery, intimidation of witnesses or other illegal interventions to influence court decisions. Of course, there are already laws in place that prohibit this type of activity — but this doesn’t change the fact that prediction markets can create highly lucrative new incentives to commit these crimes.
Balancing bad headlines and useful outcomes
Prediction markets are complicated, with lots of tricky open questions and very legitimate concerns that need to be resolved. Still, they do provide meaningful benefits to legitimate economic actors and society more broadly, which regulators need to keep in mind as they decide how to intervene further. Regulators too often tend to throw the proverbial baby out with the bathwater when regulating new technologies and industries. With prediction markets, however, hopefully they will find ways to reduce genuine risks without mutilating the beneficial functions of these platforms.
Useful cautionary tales can be found in Europe’s track record with overregulation in emerging and strategic sectors. In the energy sector, a combination of aggressive decarbonization mandates, underinvestment in domestic production and (in Germany in particular) a hostility toward nuclear power all conspired to sharply increase dependence on imported natural gas from Russia right on the eve of the 2022 invasion of Ukraine, leaving Europe acutely exposed. In technology, the European Union’s General Data Protection Regulation (GDPR) imposed significant compliance costs that disproportionately burdened startups and smaller firms, contributing to the persistent gap between Europe and the US in tech sector competitiveness.
Similar regulatory “overkill” ensured the digital asset sector in the EU would also be left behind, as the bloc was early and all too eager to implement comprehensive crypto regulation through the Markets in Crypto-Assets Regulation (MiCA). There, too, the heavy compliance burdens have prompted firms to reconsider operating in the region or delay expansion. More recently, the EU seems to be repeating the same mistake yet again, this time with AI. The EU AI Act introduces extensive ex-ante compliance requirements that many industry participants argue will hinder innovation compared to less restrictive jurisdictions.
While regulatory clarity often is a benefit for any emerging industry or technology, there is a serious risk that excessive precaution and knee-jerk regulation may come at the cost of competitiveness. When regulation moves faster than the underlying technology and market structure, it has the capacity to suppress the development of entire sectors and to waste the potential benefits they would offer to the economy and society had they been allowed to flourish. In the case of prediction markets, the danger is that they will be treated as illegal gambling platforms and banned altogether — as a number of EU countries have already done, including the Netherlands, France and Spain.
