Authored by TBI’s Science & Technology policy team
Chapter 1
The ability to anticipate what happens next has always been a source of power for governments. This is why bureaucracies have invested heavily in capabilities such as intelligence gathering and economic modelling – they improve the ability to understand potential futures and act or react accordingly. Today the tools available to do this are changing. Advances in artificial intelligence have ushered in a new era of “superforecasting”, with AI generating predictions with greater accuracy than some of their human counterparts and at a fraction of the cost.[_]
Prediction markets are playing a key role in this transformation. These markets allow participants to trade contracts based on whether they think future events will occur. In a relatively short period, prediction markets have become an integral part of the emerging digital economy, as demonstrated by growing trading volumes and their integration into mainstream financial and information platforms.
Whether governments choose to use prediction markets or not, their growing influence means leaders need to understand them. They are already being used to forecast political developments and the information they generate is increasingly accessible to other actors such as businesses, investors and the public. Yet the rise of prediction markets also raises significant questions around regulation and the potential implications for economic and national security.
This paper provides a short introduction to prediction markets. It explains what they are and how they work, examines how they are being used, and looks at the opportunities and risks they create, it also offers leaders several options for how to improve their knowledge and understanding of these markets. The appropriate approach to adoption and regulation will differ between countries, but making the right choices starts with understanding the technology and developing a strategy that reflects respective countries’ circumstances. Governments must build familiarity with the technology, examine whether and how prediction markets could add value to existing workflows, and ensure they understand the regulatory and national-security implications.
Editor’s note: We are grateful to ADI Foundation for supporting the research that informed this publication. The analysis and conclusions are solely the responsibility of TBI, in line with our intellectual-independence policy.
Chapter 2
Prediction markets provide a way for market participants to put money behind a prediction about whether a particular outcome will occur – for example, whether a candidate will win an election or a specific event will happen before a given date. Participants trade contracts tied to these outcomes, with the price reflecting the probability collectively assigned to an event. For example, a contract trading at $0.70 suggests that market participants collectively assign a probability of about 70 per cent to the event occurring.
The underlying logic is that individuals possess different pieces of information about future developments and that, by creating a financial incentive for participants, market prices can reveal collective expectations that might otherwise remain dispersed. A 2026 study analysing about 1 million resolved forecasts across four platforms and 12 domains found that prediction markets were generally well calibrated: events assigned a given probability tended to occur at roughly that rate. Accuracy, however, varied across domains.[_]
Three characteristics distinguish prediction markets from their conventional forecasting counterparts. First, they use financial incentives to aggregate dispersed information and continuously update expectations. This gives greater influence and rewards to participants who are better informed and more confident in their assessments. Second, unlike conventional opinion surveys, expressing confidence in a prediction generally involves accepting financial risk. As a result, market prices should be more accurate because traders are economically incentivised to process all available information.[_] Thirdly, they offer high-frequency and near-instantaneous updates in response to new information, providing rapid feedback on emerging events.
The concept of prediction markets is not new: they have existed in some form since the 16th century.[_] However, their scale and commercial significance have grown considerably, and the industry is projected to reach $96 billion by 2035.[_] This growth is driven by competing platforms such as Polymarket and Kalshi, each accounting for about 50 per cent of the market,[_] alongside increased visibility and low barriers to entry. The growth of these platforms has transformed the prediction-market landscape, with trackers reporting a 53 per cent increase between August and September 2026, with $77 billion traded in September across platforms.[_]
The scale of expansion suggests prediction markets are moving beyond their origins in experimental forecasting to become an important feature of the digital economy, increasingly embedded in major exchanges, crypto wallets and financial media as a source of probability data. Google, for example, has integrated prediction-market information from Kalshi and Polymarket into Google Finance.[_]
Chapter 3
Be it politics, economics or geopolitics, prediction markets are playing an increasingly important role. As a result, leaders and decision-makers should develop an understanding of the different ways in which the data from these markets are or can be used.[_] Here are three examples,
Politics and elections: Political forecasting remains one of the most visible uses of prediction markets. The 2024 presidential elections in the United States demonstrated more than $3.6 billion in trading volume on Polymarket alone, reinforcing the growing influence of contract markets.[_] Research also suggests that prediction markets in that election performed better than traditional polling.[_] The 2026 midterm elections will provide another important test of how these markets perform in a complex and closely watched electoral environment.[_] However, highly confident market forecasts can still be wrong, particularly when participation or information quality is uneven.[_] Even so, research suggests that prediction markets are beginning to have an impact on voter behaviour.[_] So, while they should not be relied on as a digital oracle, they do provide strong and important signals about sentiment for political leaders.
Economic forecasting and financial decision-making: Macroeconomic forecasting is one of the most promising applications of prediction markets, allowing participants to trade on outcomes such as interest rates, inflation figures, employment data and economic growth. In February 2026, an article by the US Federal Reserve concluded that prediction markets provide a valuable, high-frequency source of information, including the distribution of possible outcomes.[_]
Geopolitics and strategic foresight: Prediction markets also cover diplomatic negotiations, political instability, major government decisions and international conflicts. For governments they provide an additional source of information that can be leveraged for understanding changing geopolitical dynamics. However, compared with other markets, such as sports, these markets see significantly less volume. This weakens the value of the signal that is provided but also opens up opportunities for potential adversarial operations to influence reactions in a particular direction.
Chapter 4
The growing influence of prediction markets has raised significant questions and concerns about the technology and how it is regulated. Leaders seeking to harness the potential of prediction markets will need to understand the risks and adapt to an evolving regulatory environment.
Many concerns revolve around the potential for market manipulation, insider trading and risks to information integrity and national security. This is a side effect of prediction markets’ core feature: creating financial incentives to influence market prices. In areas where trading activity is limited, participants with substantial financial resources may be able to move prices significantly.[_] If those prices are subsequently reported as reliable forecasts, market manipulation could influence public perceptions of political or economic developments.
These risks are heightened when markets concern national security, government decisions or diplomatic developments. Individuals with access to sensitive or non-public information may be financially incentivised to trade before the broader market can react, while actors with sufficient resources may seek to manipulate the signals on which others rely. Recent cases involving government and military personnel demonstrate that these risks are not merely theoretical.[_],[_]
In October 2026, US Secretary of War Pete Hegseth warned in a memo that when an American service member, intelligence professional or other department civilian engages in a financial transaction tied to a military outcome, they create a signal that hostile foreign intelligence services can exploit to anticipate movements, targets and timing.[_] The memo followed the arrest of a US Army soldier who leveraged confidential information about the timing of the capture of Venezuelan President Nicolás Maduro to make $410,000, as well as reports of highly specific, well-timed bets on whether the US and Iran would reach a ceasefire on 7 April.[_],[_]
A second question, and one that is especially pertinent for leaders, is around the reliability of the information produced by prediction markets. They have demonstrated considerable forecasting potential. Their prices can provide useful real-time measures of the likeliness of an event and have, on some occasions, performed better than conventional surveys and forecasting methods. Across US presidential elections between 1988 and 2008, for example, prediction markets proved more accurate than polling at comparable points in the election cycle.[_] However, prediction markets are not consistently accurate and tend to perform best when outcomes are already relatively predictable. Their effectiveness depends on market conditions – and prices, while broadly reliable, can still be affected by biases, inconsistences and manipulation.
Different Approaches to Regulation
These questions of accuracy, manipulation and information integrity feed directly into a third challenge: how prediction markets should be regulated. Prediction markets operate at the intersection of financial regulation, gambling regulation and digital-platform governance, with approaches varying significantly across jurisdictions.
In June 2026, the Commodity Futures Trading Commission (CFTC) in the US proposed a framework for contracts involving activities such as terrorism, assassination, war and gambling.[_] However, disputes over how to regulate these markets continue to grow. For example, 39 states have backed New Jersey’s petition seeking to establish state authority to regulate prediction markets, rather than leaving oversight exclusively to the federal CFTC.[_],[_] Businesses have also called for greater regulatory clarity, with the NFL urging the Supreme Court to clarify its position in October 2026.[_]
The United Kingdom has adopted a different approach. Its Gambling Commission has stated that current commercial prediction markets fall within the definition of betting intermediaries and would generally require “betting intermediary” licences.[_]
These divergent approaches reflect disagreement over whether prediction-market contracts constitute financial instruments, betting products or a distinct category of information markets. Gibraltar recently created a separate category altogether, offering a bespoke regulatory framework for prediction markets with its own licensing and compliance requirements, thereby acknowledging they differ from traditional betting or financial mechanisms and providing legal certainty for operators.[_] For governments, the challenge is to protect consumers and maintain market integrity without restricting potentially valuable forecasting and information services. As prediction markets increasingly operate across borders, fragmented oversight could lead to inconsistent safeguards and regulatory gaps.
Chapter 5
Prediction markets are becoming an important part of the digital economy and wider information environment. Governments will therefore need to make choices about how they engage with them: whether and where their data can improve existing forecasting and decision-making, how their use should be regulated and what safeguards are needed to manage emerging risks. There is no single approach that will be appropriate for every country. But making good choices requires governments to understand how prediction markets work, what they can and cannot do, and how they interact with existing institutions and information sources.
First, build greater understanding of prediction markets across institutions that may encounter or use the information they generate, not just those responsible for regulating them. Relevant economic, security, intelligence and strategic-policy institutions should understand how prediction markets operate, what information their prices convey and where their limitations lie. This should include targeted training and education to build institutional familiarity with prediction markets and ensure officials are equipped to interpret their outputs appropriately.
Second, understand and gain familiarity with potential applications by running small, targeted trials to determine where prediction markets could add value to existing forecasting and decision-making processes. Trials could test applications in areas such as macroeconomic forecasting, geopolitical analysis, strategic foresight or policy implementation, comparing prediction-market outputs with established analytical approaches. The objective should be to generate evidence about where these tools improve the speed, accuracy or quality of existing assessments (and where they do not), before considering wider adoption.
Third, craft regulatory regimes and national-security safeguards that align with a country’s interests and ambitions. Policymakers should assess whether existing frameworks adequately address insider trading, market manipulation, consumer protection and contracts linked to sensitive government or security-related events. Particular attention should be given to markets that create financial incentives to exploit classified information or influence real-world outcomes. Governments should also consider how prediction-market information may affect wider information integrity and financial stability as its distribution expands.