Prediction Markets as “Truth Machines”

Prediction markets have rapidly emerged as prominent tools for estimating the likelihood of uncertain events, ranging from elections and Super Bowl results to daily weather extremes in specific cities. Instead of relying primarily on formal mathematical models, these platforms harness the dispersed, partial information and heterogeneous beliefs of individuals, who anonymously express their views by risking real money. In a typical market, participants can buy “Yes” or “No” contracts on questions such as “Will Democrats win the 2028 U.S. presidential election?” Each contract sells for some fraction of a dollar and pays out one dollar if the chosen outcome occurs and zero otherwise. Under this design, the going price of the “Yes” contract for a given event can be interpreted as a market‑based estimate of the probability that the event will happen.

In the United States, Kalshi and Polymarket currently stand out as the two largest and most visible prediction market platforms. Trading activity on real‑world events has expanded dramatically as these markets have gained public attention; for instance, Polymarket’s contracts related to the 2024 presidential election reportedly reached peak trading volumes in the billions of dollars. Industry research from Eilers & Krejcik projects that total annual trading volume in prediction markets could climb to roughly one trillion dollars worldwide by the end of the decade, underscoring how quickly the sector may scale. Yet, as of February 2026, around half of Polymarket’s trading volume and about 90% of Kalshi’s volume were concentrated in sports‑related markets, even as both firms faced lawsuits in multiple U.S. states alleging that such sports contracts violate local gambling laws. Sports will likely remain a major driver of revenue growth, but the distinctive informational value of prediction markets arguably lies in domains where people with different knowledge sets and beliefs can collectively generate meaningful signals about economic, political, and social outcomes. As prediction markets continue to expand, both investors and regulators will need to grapple with their unusual position at the intersection of the financial system and the media ecosystem.

A central feature of platforms like Kalshi and Polymarket is that they allow traders to act on possibly confidential or otherwise non‑public information while remaining anonymous. This has fueled accusations that these markets inadequately deter insider trading. A striking example came shortly before the January 2026 capture of Venezuelan president Nicolás Maduro, when an anonymous Polymarket account placed a series of sizable bets on his capture and ultimately profited more than four hundred thousand dollars once the operation succeeded. Critics seized on this episode as evidence that anonymity effectively shields insiders from accountability for trading on privileged information. At the same time, the incident can be read as a demonstration of the epistemic power of prediction markets: observers who noticed the sudden spike in the “Yes” price after the large wager received an early, albeit noisy, signal that Maduro’s capture might be imminent. Thus, even if these platforms perform poorly by the standards of traditional financial markets in policing insider trading, they may still function as powerful aggregators of information that is not fully public. This conclusion, however, comes with an important caveat: when insiders are not merely better informed about an outcome but can influence or determine it, the boundary between forecasting events and helping to bring them about becomes dangerously blurry.

That concern is closely tied to the problem of moral hazard created by prediction markets when some participants have direct or indirect control over the outcomes on which they are betting. In economics, “moral hazard” refers to situations where an agent has incentives to take on greater risk or behave differently because they are partially insulated from the consequences of their actions. A classic insurance example is a driver who becomes less careful after purchasing comprehensive coverage. In the context of prediction markets, imagine a member of the Federal Open Market Committee (FOMC), which sets the target range for the federal funds rate in the United States, secretly placing bets on the magnitude of an upcoming rate cut. Because that policymaker has influence over the decision, the market does not merely allow them to profit from advance knowledge; it also creates incentives to tilt policy itself for personal gain. Even if such extreme scenarios are rare, the mere possibility undercuts claims that prediction markets are neutral, trustworthy mechanisms for aggregating information, especially when the participants include actors who can shape the very outcomes being priced.

Beyond insider trading and moral hazard, there is also a deeper, more diffuse worry: many people feel moral discomfort at seeing probabilities assigned—and financial profits attached—to grave political and human events. The controversy surrounding the Policy Analysis Market (PAM) illustrates this unease. Proposed in 2001 by the Defense Advanced Research Projects Agency (DARPA), then housed within the U.S. Department of Defense, PAM was designed as a government‑run prediction market in which traders could buy and sell futures contracts tied to geopolitical events in the Middle East. The idea was that contract prices, reflecting the aggregated beliefs of informed traders, would help U.S. officials better assess and track geopolitical risk over time. Yet in July 2003, the initiative was denounced in Congress as a “terrorism futures market,” and public backlash was swift. Critics argued, among other concerns, that such a market could effectively reward terrorists or their collaborators for betting on political violence before carrying out attacks. In response, defenders of PAM noted that similar profit opportunities already exist indirectly, because the prices of stocks and commodities like oil often react sharply to terrorist attacks and other geopolitical shocks.

Although PAM’s failure involved a government‑run market, whereas Kalshi and Polymarket are privately operated, the underlying anxieties it exposed are likely to resurface as contemporary prediction markets seek a more central role in the information environment. Major media outlets such as CNBC, CNN, and Dow Jones have already begun incorporating data from Kalshi and Polymarket into their coverage, both on air and online, thereby treating prediction market prices as indicators of the likelihood of various newsworthy events. In an era marked by AI‑driven misinformation, fragmentation of traditional news media, and declining trust in legacy institutions, the ability of prediction markets to serve as credible sources of public information will depend on whether society can reconcile their hybrid identity: they are at once epistemic institutions designed to aggregate beliefs about future events and venues for speculative trading where participants seek financial gain, sometimes under conditions of anonymity and asymmetric power.