Diamond Hands, War Plans

In the predawn hours of Jan. 3, U.S. Special Operations Forces captured Venezuelan President Nicolás Maduro in Caracas. Hours later, Gannon Ken Van Dyke withdrew $409,881 from his Polymarket account, converting a $33,034 marker placed over the preceding week into a 12-fold return. He had repeatedly bet that Maduro would be out of power by Jan. 31. 

According to a federal indictment unsealed in April 2026, Van Dyke knew why. He was a Special Forces Master Sgt. assigned to U.S. Army Special Operations Command at Fort Bragg, involved in planning and executing Operation Absolute Resolve—the very operation he spent the week of Dec. 27, 2025, betting on. His last trades came the evening of Jan. 2. The raid began before dawn the next morning. He is reportedly the first U.S. service member prosecuted for using classified information to trade on a prediction market.

“Diamond hands” (💎🙌) —a trader’s shorthand derived from Reddit trading forums for holding a position through volatility in expectation of eventual payoff—captures something important about the alleged conduct. Van Dyke’s conviction was not belief. It was foreknowledge.

The case also exposes a broader structural problem. Prediction markets have expanded into geopolitical and national security domains—precisely the domains where relevant information is often classified, compartmented, and unevenly distributed. In those contexts, prediction markets do not merely aggregate public sentiment. They convert private information into publicly observable signals.

That creates two related problems. First, trading activity tied to military operations or covert action may generate observable indicators available to any adversary monitoring a public price feed. Second, the legal framework is fragmented. Criminal statutes, commodities regulations, ethics rules, and insider trading doctrines each reach portions of the conduct, but none cleanly fits the problem—a patchwork of overlapping regimes designed for different purposes, none adapted to prediction markets tied to national security events.

Van Dyke’s prosecution demonstrates that existing authorities are sufficient to bring charges after the fact. But the prosecution also shows that regulatory gaps remain unique to the national security context. 

What Is a Prediction Market

Prediction markets translate expectations into tradable positions. Participants buy and sell binary contracts—paying $1 if an event occurs, $0 if it does not. Prices function as market-implied probabilities: A contract trading at $0.70 reflects a roughly 70 percent market estimate that the event will occur.

The theory is that markets aggregate dispersed information efficiently. In low-stakes contexts, that information is usually public, diffuse, or speculative. Bets exist on everything from elections to the date of Taylor Swift’s wedding. 

Unlike traditional gambling, there is no house. Participants trade against each other; platforms collect fees. That structure concentrates winnings sharply among the most informed. A Wall Street Journal analysis found that more than 70 percent of Polymarket users lose money, while fewer than 2,000 accounts capture 67 percent of all profits. The market rewards information advantages systematically and at scale.

Over the past two years, prediction markets have grown from a niche product into a significant financial phenomenon. More than $12 billion reportedly flowed through prediction market platforms in December 2025 alone. As these markets expand into geopolitical and military events, they increasingly intersect with the national security system. Cryptocurrency-related investments are especially popular within the military community. 

Two exchanges currently dominate the market. Kalshi operates as a U.S. Commodity Futures Trading Commission (CFTC)-regulated contract market subject to customer identification requirements, surveillance obligations, and reporting rules. Polymarket uses an offshore structure with a specific U.S. subsidiary, restricts U.S. users through geolocation controls, and relies heavily on cryptocurrency infrastructure. Those restrictions, however, are readily circumvented using virtual private networks.

The platforms differ in attribution, not visibility. Kalshi transactions are tied to identified users. Polymarket transactions are publicly visible on a blockchain but can be pseudonymous without forensic attribution. But a trader’s identity is not the only thing that matters. The trade itself is a market signal.

The Commodity Exchange Act’s “Special Rule” authorizes the CFTC to prohibit event contracts involving terrorism, assassination, war, or similar activity contrary to the public interest. But this authority has structural limits that foreclose any application to Van Dyke. The statutory default is permissive: Even contracts involving terrorism and assassination are allowed unless and until the commission affirmatively acts. In practice, the CFTC has administered the Special Rule through case-by-case public interest review rather than categorical prohibition—and it has not issued a rule or order covering political prediction markets or assassination contracts for prediction markets. The Special Rule has been invoked once—against Kalshi's proposed election contracts in 2023—and was overturned by the U.S. Court of Appeals for the D.C. Circuit in September 2024 on the grounds that election contracts did not constitute "gaming." It has never been invoked for the national security categories most directly at issue here: war, terrorism, or assassination.  Even if it had been, it would not reach Van Dyke’s trades: The rule is a prospective listing mechanism directed at exchanges, not a retroactive enforcement tool against trader conduct. It controls what CFTC-regulated platforms can list, not who may trade or on what informational basis. And its jurisdictional reach extends only to domestic, CFTC-registered entities. Van Dyke traded on Polymarket's offshore exchange—beyond the Special Rule’s reach entirely.

From Prices to Signals: Prediction Markets as an Intelligence Surface

When trades are signals, that creates distinct national security vulnerabilities. Military operations, covert activities, and geopolitical crises involve highly asymmetric information environments—a small number of people possess relevant information as the broader public relies on open-source reporting or inference. Open-source intelligence (OSINT) analysts have long tried to close that gap through indirect indicators: the so-called Pentagon Pizza Tracker holds that late-night delivery surges signal imminent military action, a theory complicated by the fact that the building contains more than twenty internal restaurants. The Strava episode illustrated the real vulnerability more precisely: In 2018, a fitness app’s global heatmap inadvertently exposed U.S. military base locations in Syria, Iraq, and Afghanistan because soldiers were jogging with consumer devices, and the aggregate movement data was publicly visible. No classified access was required. A prediction market price feed is the same problem, stripped of ambiguity—no aggregation, no inference, and a signal financially incentivized to be accurate. In those environments, trading activity becomes informative.

An adversary intelligence service need not know who placed the trade—only what it suggests. A large, concentrated, precisely timed position tied to a military event can be an observable indicator of privileged access. That distinguishes this context from traditional insider trading: In securities markets, insider trading primarily implicates counterparties and market integrity. In national security prediction markets, the trading activity itself may generate operationally relevant signals observable in real time.

According to the indictment, Van Dyke created his Polymarket account on Dec. 26, 2025.  Between Dec. 27 and Jan. 2—the eve of the Caracas raid—he allegedly placed 13 bets tied to Maduro losing power by Jan. 31. This timeline lines up with media reports that operators trained on a mock-up of the Maduro compound and postured for days to execute the mission, waiting for good weather conditions. After the operation became public and scrutiny increased, prosecutors allege he attempted to conceal the activity by asking Polymarket to delete his account, changing associated email addresses, and routing proceeds through cryptocurrency infrastructure and foreign-hosted accounts.

Van Dyke was reportedly identified not through domestic financial surveillance or insider threat monitoring, but because Polymarket itself flagged the activity and referred it to the Department of Justice. The Justice Department and the CFTC brought parallel criminal and civil actions.

Van Dyke’s case is the clearest public example of the problem, but not the only one. Six individuals allegedly placed concentrated bets predicting U.S. strikes against Iran shortly before explosions in Tehran were reported. In another instance, newly created accounts placed significant wagers on a U.S.-Iran ceasefire just before President Trump announced it. Israeli authorities separately arrested reservists accused of using classified information to trade on military operations tied to the Iran-Israel conflict.

Not every anomalous trade reflects insider access—force buildups, diplomatic signaling, and herding behavior can all produce concentrated positions without classified information. Each anomalous trade is a public, time-stamped indicator, observable to any actor monitoring the market. What once required costly human intelligence may now be inferred from a public price feed. 

A large, concentrated, precisely timed position is itself an observable event—one that can influence adversary decision making, media narratives, and intelligence assessments regardless of whether it reflects genuine foreknowledge or ultimately proves correct. If prediction markets come to be perceived as reflecting insider access, that perception amplifies the effect. Prediction markets have acquired enough institutional credibility that their movements drive media coverage and shape analytical assumptions.

That credibility is the attack surface. Hostile actors could exploit prediction markets as a mechanism for strategic deception. By placing conspicuous or strategically timed positions designed to mimic insider trading, an adversary could manufacture the appearance of foreknowledge, trigger speculative reporting or intelligence scrutiny, and distort how observers interpret unfolding events. Prediction markets therefore may create not only opportunities for intelligence collection but also a novel channel for military deception and influence operations.

Fragmented Legal Regimes and the Limits of Existing Law

The legal framework is fragmented but not absent. In Van Dyke’s case, federal prosecutors have assembled overlapping theories drawn from national security law, fraud doctrine, commodities regulation, and ethics restrictions. Each partially reaches the conduct, but none fully resolves it. 

The Espionage Act

The most intuitive response is that trading on classified information should fall within national security law, but it’s not a clean fit. The Espionage Act is designed to prohibit unauthorized retention, transmission, or disclosure of national defense information. A cleared insider can place a profitable trade, move a market price, and generate an observable public signal without communicating the underlying information to anyone. The statutes do not clearly map onto: The provisions are disclosure oriented, and prediction market trading exposes information through market activity, not transmission.

Courts have not meaningfully addressed whether generating a public market signal through trading behavior constitutes “communication” of protected information within the meaning of those statutes. 

Fraud and Theft

The central doctrinal question is whether classified government information is government “property” for purposes of federal fraud and theft statutes, or instead reflects a regulatory interest protected through classification controls. Fraud provides a plausible framework, but it raises the question of whether classified information obtained through federal employment is “property” under the meaning of fraud statutes. Wire fraud (18 U.S.C. § 1343) requires a scheme to defraud executed through interstate wire communications. The precedent in Carpenter v. United States suggests that confidential business information can be property within the statute’s meaning, bringing misappropriation within the reach of wire fraud. Cleveland v. United States, however, excludes purely regulatory governmental interests. If the classification system is deemed regulatory rather than proprietary, the misappropriation of information theory from Carpenter could not be used to prosecute the use of classified national security information to game prediction markets.

Wire fraud requires a scheme to defraud someone of money or property through misrepresentation—not just wrongdoing, and not just harm. The Supreme Court’s 2025 decision in Kousisis v. United States eliminated the requirement that a victim suffer a net financial loss, but preserved the property requirement and declined to extend liability to intangible harms or regulatory violations. Whether classified information qualifies as “property” — the central question of Cleveland—therefore remains dispositive. Kousisis may offer a cleaner path: Where classified insight induces a transaction through material misrepresentation, wire fraud liability may attach without resolving the property question. But anonymous, exchange-mediated trading presents a threshold problem—no representation is made to any counterparty, and no identifiable person is deceived into transacting. The inducement element fails before reaching the property question.

The federal theft statute (18 U.S.C. § 641), which prohibits theft or conversion of government property, offers an alternative vehicle. Some courts have construed § 641 to reach unauthorized use of confidential information absent physical transfer. The conversion prong—covering whoever “knowingly converts to his use” any “thing of value of the United States:—is the most plausible hook, requiring no copying, transmission, or dissemination.  Profitable use of the information is the conversion. Two uncertainties remain: Whether internalized classified knowledge qualifies as a “thing of value,” and whether use without deprivation satisfies conversion where the government retains access to the underlying information. Van Dyke was charged under § 641; whether courts will sustain the theory is untested.

Insider Trading and Commodities Law

This is insider trading. Not in a loose or analogical sense—a participant with access to classified information, trading on that information ahead of the market for personal gain, has done precisely what insider trading law was designed to prohibit. But the rules that give insider trading law its teeth—Rule 10b-5, the duty-of-trust doctrine developed through Chiarella v. U.S., Dirks v. SEC, and U.S. v. O'Hagan, the disclosure-or-abstain obligation, the settled body of doctrine that makes prosecution predictable—were developed for securities markets. Prediction markets, however, are not securities, as defined by 15 U.S.C. § 78(c). Under SEC v. W.J. Howey Co., an investment contract qualifies as a security only if it involves an investment of money in a common enterprise with profits expected from the managerial efforts of others. Prediction market contracts satisfy the first two elements—participants invest money in a pooled market structure—but fail the third and most important: The payout on a prediction market contract is determined not by the business decisions of any issuer, promoter, or manager, but by the occurrence of an event outside of anyone’s control. When a trader buys a contract tied to an election outcome or a policy decision, profit and loss are functions of reality, not of anyone’s managerial choices. That is precisely the distinction Howey drew. Prediction market contracts fall outside the statutory definition of a security—and therefore outside the entire apparatus of securities enforcement.

Prediction market contracts are CFTC-regulated derivatives. The Commodity Exchange Act (CEA) (7 U.S.C. § 6b) has anti-fraud laws, which are modeled on Rule 10b-5, but there are key differences. For derivatives, the CFTC’s primary anti-fraud Rule 180.1 does not impose a general affirmative duty to disclose nonpublic information. The operative gap is not structural but doctrinal—decades of duty-of-trust analysis refined in securities law have no direct analogue in CEA enforcement.

Both frameworks also require some form of deception or breach of duty to the counterparty—mere possession of material nonpublic information is not enough. A cleared insider trading on internalized classified knowledge may fall entirely outside traditional doctrine if the conduct involves no disclosure, no tipping, and no misrepresentation.

The STOCK Act 

The STOCK Act provides perhaps the most direct statutory hook. The act amended the Commodity Exchange Act (CEA) (7 U.S.C. § 6c)  to prohibit federal employees (and not just Congress members or their staff) from using material nonpublic information obtained through official positions to trade in commodities-related markets. Unlike traditional fraud frameworks, the provision does not depend primarily on deception. It turns on the status, source of information, and use.

The statute contains a notable ambiguity. The CEA amendment expressly covers congressional and judicial personnel, but does not expressly identify service members. Executive branch personnel are separately covered by a STOCK Act duty-of-trust provision (15 U.S.C. § 78u-1)—but whether uniformed service members qualify as “executive branch employees” under that provision is unresolved on the statute’s face. Federal law frequently distinguishes civilian employees from uniformed service members, and the STOCK Act does not address that distinction. The act’s financial disclosure provisions offer a partial textual signal: they cover uniformed members at O-7 and above but exclude O-6 and below—suggesting Congress considered uniformed personnel but made selective coverage choices. That signal cuts both ways, and no court has resolved the question. A uniformed service member trading on classified operational information occupies uncertain statutory ground. The Van Dyke indictment suggests that the Department of Justice concluded that the STOCK Act applies to enlisted military personnel—the indictment charged him with violations of the STOCK Act. But a charging decision is not a judicial holding, and the question of whether the provision reaches uniformed personnel remains untested in court.

Executive branch ethics regulations (5 C.F.R. § 2635.703) separately prohibit federal employees from using nonpublic information for financial gain, including classified information. But this operates as an internal ethics constraint, not a market-facing enforcement regime—violations are addressed through administrative discipline, clearance adjudication, or removal. For military personnel, the Uniform Code of Military Justice provides parallel accountability through failure to obey a general order (10 U.S.C. 892), failure to obey general order or regulation (also 10 U.S.C. § 892), conduct unbecoming an officer for commissioned or warrant officers (10 U.S.C. § 933), or, failing any other more specific article, the general article for crimes prejudicial to good order and discipline in the armed forces or of a nature to bring discredit upon the armed forces (10 U.S.C. § 934). 

Taken together, and using the Van Dyke case as an example, these overlapping frameworks address the emerging issue of prediction markets unevenly and with gaps. The Department of Justice charged Van Dyke with three CEA counts—including unlawful use of confidential government information under (Counts One and Two) and commodities fraud under (Count Three)—plus wire fraud and money laundering. Prosecutors did not choose to charge him with theft of government property or violations of the Espionage Act. Charging decisions involve many considerations, and the omission does not conclusively establish that the statute is unavailable. The Van Dyke indictment demonstrates the government is not bereft of prosecutorial tools. What it lacks is a settled, purpose-built regime.

How Can Misuse of Prediction Markets by Military or Intelligence Personnel Be Detected?

Compounding the problem is that enforcement requires detection, and while the trades themselves are public, there is no mechanism to detect national security insiders who abuse their access. The existing oversight mechanisms are siloed, with different authorities and data streams. 

Kalshi, as a CFTC-regulated exchange, is required to conduct surveillance to detect anomalies suggesting manipulation or wash trading (which is defined as the purchase and sale of the same instrument; in the same or substantially similar quantity; at the same or similar time; with no change in beneficial ownership; and with intent to create artificial activity).

Polymarket presents a categorically different problem. As a foreign corporation, it largely falls outside the CFTC regulatory framework. Its blockchain architecture makes transactions publicly visible but not inherently attributable—wallet identification requires forensic analysis that is typically investigative rather than proactive.

The Van Dyke indictment documents his alleged concealment methods. Van Dyke allegedly used VPNs to access the offshore Polymarket exchange and place bets on the Venezuela operation. He then allegedly withdrew the proceeds from his Polymarket account through a foreign cryptocurrency “vault” that “generates interest for depositors by lending cryptocurrency and tangible assets to others” (i.e., a “DeFi lending vault” designed to evade anti-money laundering and counterterrorism finance regulations). Van Dyke also reportedly made operational missteps: He asked Polymarket to delete his account—falsely claiming he had lost access—and attempted to change his associated email, which was originally registered under his name. Polymarket, despite being a foreign corporation with only a U.S. subsidiary subject to U.S. jurisdiction, voluntarily disclosed its findings and cooperated with the Department of Justice. Better operational security on his part could have required blockchain forensics rather than platform records to identify him. 

Polymarket’s cooperation may not have been entirely altruistic. In the wake of the indictment, both platforms publicly touted their “comprehensive” market integrity frameworks—a posture calibrated as much toward forestalling congressional action as toward genuine compliance. Kalshi went further, using the indictment as a competitive weapon to distinguish itself from Polymarket's offshore operations—a distinction with some factual basis: Van Dyke had been rejected by Kalshi’s compliance system before migrating to Polymarket’s unregulated DeFi interface. The irony is pointed: The platform that enabled the conduct was the one that cooperated; the platform that blocked him received the credit.

But poor operational security and voluntary cooperation are unreliable detection mechanisms. Affirmative reporting requirements must be in place. Two reporting avenues exist. Domestic exchanges can be required to make reports under the Bank Secrecy Act (31 U.S.C. § 5318). But a trader who deposits and withdraws cryptocurrency through a foreign exchange, without ever trying to convert his crypto holdings into cash may generate no domestic regulatory footprint—not because U.S. law does not apply, but because no domestic surveillance mechanism observes the transaction until funds enter the U.S. financial system. 

The most direct detection mechanism is the insider threat infrastructure. Executive Order (EO) 13587 and the Office of the Director of National Intelligence National Insider Threat Policy require agencies to integrate information bearing on potential insider threats; for the Department of Defense, that mandate runs through the Defense Insider Threat Management and Analysis Center (DITMAC). Anomalous financial activity is an established behavioral indicator, and clearance holders bear a continuing obligation to self report significant financial changes. But that obligation depends on voluntary compliance. The Continuous Vetting process is not calibrated to detect exploitation of classified access for market gain. Current collection focuses on credit bureau reports, which provide some insight into distress indicators (bankruptcies, debt accumulation) or purchases indicating unexplained affluence. A cleared insider who trades successfully on classified foreknowledge may, by standard indicators, appear more financially stable than before. 

Even where detection is possible, information sharing limits response. CFTC surveillance data is not routinely shared with the intelligence community or counterintelligence at the Department of Defense. Existing coordination mechanisms are reactive and predicate-dependent—under 31 C.F.R. § 1010.520, the Department of Treasury’s Financial Crime Enforcement Network (FinCEN) can be tasked only upon certification of reasonable suspicion of terrorist activity or money laundering—neither of which prediction market trading facially satisfies. No existing framework performs the correlation this conduct requires: mapping anomalous trading activity against an individual’s access to relevant classified information at the time of the trade.

Legislative and Regulatory Responses

In the aftermath of the Maduro raid, multiple proposals have emerged in Congress intended to address insider trading on military and national security activities. These include the following:

• The Public Integrity in Financial Prediction Markets Act of 2026 (HR 7004)  would prohibit federal elected officials, political appointees, executive branch employees, and congressional staff from buying or selling prediction market contracts tied to government policy, government action, or political outcomes when they possess material, nonpublic information, or could reasonably obtain such information through their official duties.
• The Preventing Real-Time Exploitation and Deceptive Insider Congressional Trading Act (PREDICT Act) (HR 8076) would prohibit members of Congress, their family members, their fiduciaries, the president, the vice president, any political appointee, any federal employee GS-15 or above or O-7 or above, judicial officers, or judicial employees from entering into prediction market contracts dependent on the occurrence, nonoccurrence, or the extent of the occurrence of a specific political act.
• The DEATH BETS Act (S. 4035) would categorically prohibit all regulated exchanges from listing for trading or accepting for clearing any contracts that involve, relate to, or reference terrorism, assassination, war, or any similar activity, as determined by the CFTC, or any contract that involves, relates to, or references any individual’s death. This would enhance the current restrictions in the CEA, removing CFTC discretion to allow such contracts if not contrary to the public interest. 
• The BETS OFF Act (HR 7955), similarly to the DEATH BETS, would categorically prohibit all regulated exchanges from listing any event about terrorism, assassination, war, and government actions broadly where “an individual knows or controls the outcome.” Unlike the DEATH BETS Act, the BETS OFF Act would also amend existing laws against illegal gambling to shut down payment systems used by illegal online platforms and impose criminal penalties for U.S. persons who promote, manage, own, or supervise these businesses domestically.

The more sweeping proposals are unlikely to be enacted in their current form, given the Trump administration’s broader deregulatory posture toward digital asset and prediction markets.  Given, however, that the CFTC has sought comment on the application of the CEA and the scope of contracts that may be contrary to public interest (including those pertaining to war and terrorism) and how to regulate insider trading, there may be an emerging bipartisan path for a narrow statute or regulation that prohibits such contracts on exchanges subject to U.S. jurisdiction. Congressional support is substantial but stratified. The broader proposals—prohibiting categories of contracts tied to war, terrorism, or government action—have drawn heavy Democratic backing, with the Torres bill alone accumulating 42 co-sponsors, but limited Republican enthusiasm in a deregulatory administration.

The narrower insider trading prohibition has a more credible bipartisan path: the PREDICT Act (Rep. Adrian Smith, R-Neb.; Rep. Nikki Budzinski, D-Ill.) in the House, and in the Senate the Public Integrity in Financial Prediction Markets Act of 2026 (Sen. Todd Young, R-Ind; Sen. Alissa Slotkin, D-Mich.; Sen. John Curtis, R-Utah; Sen. Adam Schiff, D-Calif.)—both target government officials without disturbing the broader prediction market structure, a frame more consistent with the administration's posture and more likely to survive it. That political will appears to be consolidating: On 22 May 2026, the House Oversight Committee launched a formal investigation under Republican chairmanship, and the Senate Commerce Committee held a hearing at which lawmakers from both parties scrutinized the platforms. Congressional attention appears increasingly likely, though the scope of any response remains uncertain.

Recommendations

The legislative proposals address the supply side—platforms and contracts—but largely leave the demand side intact. Van Dyke traded on an offshore platform that happened to cooperate with law enforcement. A motivated insider could use a platform less inclined to do so. Payment restrictions create friction but cannot eliminate access for an actor who avoids the domestic financial system entirely. 

A more comprehensive approach would focus on the trader, not the platform. Cryptocurrency-funded offshore platforms will continue to exist, and military personnel will continue to invest in crypto, including through prediction market platforms that use cryptocurrency for settlement, making permissible speculation and insider trading financially indistinguishable from the outside. This will require expanding the financial data that the Department of Defense’s insider threat program (DITMAC, mentioned above) currently collects. That data must reflect how military members and national security professionals actually invest today.

Security managers, insider threat program personnel, and force protection officers responsible for signature management should be explicitly authorized—through statutory Congressional mandate—to collect, retain, and analyze prediction market activity about pending military operations as an anomalous behavioral indicator. That mandate would need to address the authority gap directly: Existing DITMAC collection authorities under EO 13587 and SEAD 3 do not clearly apply to financial market participation, and any expansion could implicate Privacy Act obligations and Fourth Amendment considerations that Congress, not the executive branch, should resolve.

Three reforms should proceed in parallel. First, amend the STOCK Act to expressly include uniformed service members in the CEA’s prohibition, closing any gap created by the Act’s silence on military personnel. Second, the director of national intelligence should update Security Executive Agent guidance and the SF-312 nondisclosure agreement specifically to prohibit using classified information to inform trading decisions—not merely to prohibit disclosure. Third, the executive branch should require affirmative reporting of significant prediction market positions as a condition of access under existing security clearance regulations, creating a contemporaneous record for anomaly detection. These reforms will address the problem at its source—the cleared insider— rather than trying to regulate offshore platforms. 

Conclusion

The trades preceding the Maduro raid and the Iran strikes were publicly visible, observable to any actor monitoring the price feed. Van Dyke’s indictment alleges that at least some of his prediction-market trades involved classified access rather than ordinary speculation.

His prosecution also illustrates the deeper structural problem. The government assembled overlapping theories—fraud, commodities, theft, ethics—from frameworks designed for different purposes. Significantly, the Department of Justice led with § 6c(a)(3)—the STOCK Act’s CEA provision—applied here to an active-duty service member, testing an interpretive question the statute does not resolve on its face. The insider threat system did not reportedly identify the conduct. The domestic financial surveillance system did not reportedly identify it. Polymarket did.

The legislative response addresses portions of the problem but leaves core structural gaps unresolved. Categorical bans on domestic contracts will not stop a cleared insider from accessing offshore platforms through VPNs and cryptocurrency. Existing insider trading doctrine does not map neatly onto prediction markets tied to military operations. The current surveillance architecture is not designed to correlate trading behavior with classified access. As prediction markets become more popular, cleared insiders will continue to face temptations to capitalize on their knowledge.  

Prediction markets will continue expanding into national security events because there is a market for it. The relevant policy question is therefore not whether these markets should exist at all. It is whether the United States intends to develop a coherent framework capable of addressing what happens when classified access becomes a tradable advantage in a public market.