Will there be a peace agreement between Israel and Hezbollah before April 30? Who will win the 2028 US presidential elections? Will the price of Bitcoin go up or down in the next five minutes? These are some examples of bets available today on Polymarket or Kalshi, leading prediction markets. In the past, bets have also been placed on a US attack on Iran or whether the downed fighter pilot in that conflict would be rescued, causing controversy over lack of ethics. But it also generates another greater distrust: these bets and some winnings raise suspicions about scenarios of market manipulation and use of classified information.
Read more Brussels will ease mergers to create large European companies that boost investment
These platforms, where users can gamble money on almost anything, are very popular in the US, where they continue to grow. But they should not be taken lightly, according to an article published today in the journal Science, which calls for stricter regulation. There, federal law prohibits this type of betting on crimes or murders, where Kalshi is headquartered, but Polymarket mainly operates abroad and any internet user can access it through a VPN (a virtual private network that allows you to connect as if you were in another country).
“Prediction markets pose threats to democratic integrity, from electoral manipulation to insider trading on classified government measures,” the text states. This is one of the main criticisms made by jurists from the University of Haifa, Nizan Geslevich and Sharon Rabinovitz, who authored the study: scientific literature, they argue, has shown that operators’ beliefs determine market probabilities, which in turn determine those beliefs again.
Furthermore, “prediction markets create financial incentives for citizens to leak or exploit classified information, eroding the national security on which democratic governance depends.” A few days before the US attack on Venezuela, an anonymous user suspiciously bet on Polymarket that the military operation was imminent. He won more than 400,000 euros with the investment made a few hours before Trump gave the green light.
How to regulate this market? In the US, Biden tried to regulate the sector. The Commodity Futures Trading Commission (CFTC) opened a case against Polymarket, which was closed with a fine of 1.4 million and a ban on US citizens accessing the platform. But with Trump’s arrival, all legal problems disappeared. The two major companies in the sector have the president’s son, Donald Trump Jr., as a board member or strategic advisor. In the EU, the Digital Services Act (DSA) applies only to platforms with more than 45 million users, numbers that prediction markets are still far from generating.
“Policy measures should address platform design, market transparency, and regulatory architecture,” say Geslevich and Rabinovitz, who call for the creation of interinstitutional working groups bringing together experts in finance, cybersecurity, and public health to monitor market manipulation and compulsive use patterns. “Prediction markets are at a crossroads: if designed ethically, they could improve decision-making; as currently used, they pose risks of causing behavioral and democratic harm,” they conclude.
Addictive design
But there is also a warning about the danger of these markets for another reason: their effects on public health can be severe, perhaps even greater than those of conventional betting houses.
“The time that passed between the popularization of tobacco and the emergence of scientific consensus on its harmful effects allowed millions of deaths that could have been avoided. The delay between the proliferation of social networks and the recognition of their health consequences may have turned entire generations into unwitting experimental subjects,” describe the article’s authors. “We are facing a similar moment. Many features of prediction markets are common to trading, video games, gambling, social media platforms, and various apps. However, prediction markets intensify pioneering techniques from those industries, apply them to socially and politically relevant content, and wrap them in the epistemic authority of ‘prediction’.”
For the experts, prediction houses offer a product that, like tobacco, gambling, or social networks, can be very harmful to health and society. But it does so with an aura of respectability that may cause potential users to ignore those possible negative consequences.
Part of the blame, say the authors, lies with the scientific community itself, where many see this phenomenon as a harmless social experiment. “Their silence could help legitimize prediction market systems that seek to appropriate scientific credibility while violating its principles,” they emphasize.
Read more Opposition unions head to the United States embassy to demand democracy in Venezuela
The growth of these websites is meteoric. By the end of 2025, prediction markets were moving about 2 billion dollars a week in bets. Just Kalshi and Polymarket, two of the main operators, traded nearly 12 billion in deals in December, a 400% increase over the previous year, according to investment bank Piper Sandler. Although they have been operating for years, the boom of these platforms came in the 2024 US presidential elections: when polls still showed a tie between Trump and Harris, prediction markets clearly favored the Republican magnate. Their success put them on the radar.
The article identifies three main problems around prediction markets, which are interrelated: their effects on public health, their design in the style of gambling platforms, and their potential as a tool for democratic manipulation.
Traditional casinos impose a series of controls, such as age verification or delays in cash withdrawals, to try to make players more aware of what they are doing. In prediction markets, their structure, such as the gamified interface, algorithmic segmentation, or cryptocurrency integration, makes these systems “function as mechanisms for spreading harm on a population scale, systematically evading safety measures.”
The authors also emphasize that prediction markets disproportionately affect young and financially inexperienced users, as well as economically marginalized groups, “for whom participating in this market is a way to reaffirm their civic commitment or even identity expression.” The fact that some media, such as The Wall Street Journal, CNN, or Reuters agency, integrate these systems, or that financial influencers talk about them, gives them social validation that can be attractive and normalizes their use.
Without the stigma of being labeled a gambler, it is easier to get carried away, even though the consequences may be similar. “Participants perceive themselves as analysts or engaged citizens rather than gamblers, a distortion that hinders seeking help even when suffering economic losses or psychological harm,” they state.
Public health
46.2% of adults and 17.9% of adolescents worldwide have gambled in the past year, according to data presented by the study. Gambling-related disorders affect between 2.7% and 15.8% of online gamblers, far exceeding rates recorded in traditional gambling. Harms that do not reach clinical thresholds also generate considerable economic costs: more than 100 billion euros across the EU, the researchers reflect. Access to treatment is low (between 4% and 20%), with dropout rates between 39% and 70%.
The World Health Organization classifies gambling-related disorders as mental health disorders. It also recognizes the specific risks of digital platforms. This came after decades of research showing that behavioral addictions activate similar neural pathways, produce comparable alterations, and respond to interventions similar to substance addictions. “Despite similar risk markers, prediction markets remain insufficiently studied, although evidence points to their similarity,” argue Geslevich and Rabinovitz.
“Although only 2% of those who use prediction markets develop behavioral disorders, widespread dysfunction among millions of occasional users could generate public health costs comparable to the documented burden of gambling,” warn the authors. “Prediction markets risk reproducing this burden on a large scale through addictive mechanisms similar to online casinos, operating outside regulatory frameworks, public health infrastructure, and social awareness that have taken decades to develop for traditional gambling,” the study concludes.
But the researchers acknowledge that key differences remain between online casinos and prediction markets: “[The latter] are based on verifiable real-world events, can reward user skill, and usually involve sporadic use. Therefore, gambling rates cannot simply be applied to them.”