- A Bloomberg Businessweek analysis of 34,000 transactions flagged as potential insider trades by analytics platform Polysights found that approximately $200 million worth of flagged trades occurred on Polymarket from January to June 2026, with geopolitical and war markets driving the majority of the surge; flagged trading volume spiked sharply in late February tied to Iran-related bets, and Iran markets have disproportionate overlap with US-regulated cryptocurrency funding — 71% of Iran war market volume on Polymarket came from wallets funded through US-regulated crypto exchanges like Coinbase, versus 49% for all markets combined; 57% of profitable flagged wallets were created less than 24 hours before the trades were made, a pattern that Polysights analyst Tre Upshaw says is characteristic of insider activity.
- The most serious cases have moved from regulatory concern to criminal prosecution: the Commodity Futures Trading Commission and federal prosecutors brought charges against US Army soldier Gannon Ken Van Dyke, alleging he used classified information about a military operation in Venezuela to earn more than $400,000 on Polymarket; in Israel, authorities filed charges against a military reservist and a civilian for allegedly using secret military intelligence to place Polymarket wagers on Israeli security operations in Iran; and Kalshi flagged to the CFTC a White House teleprompter operator who allegedly profited from bets on President Trump’s speeches; a set of 38 connected Polymarket wallets separately turned $1.6 million in profit by betting on Iran and Venezuela military actions with a near-perfect win rate, with all 38 wallets cashing out through the same Coinbase deposit account — a pattern that on-chain researchers traced and flagged to investigators.
- The enforcement challenge is structural and likely permanent: prediction markets reward those who get information first, and the universe of “insiders” in event-based markets is fundamentally different from insider trading in stocks — it includes government officials, military personnel, campaign aides, athletes, award show producers, and anyone else with non-public knowledge of an event outcome; the rules governing what constitutes illegal trading in prediction markets are still developing, with Polymarket updating its terms in March 2026 to prohibit trading on stolen confidential information and illegal tips, and Kalshi screening to prevent politicians from betting on their own campaigns and athletes from betting on their sports; but enforcement is complicated by anonymous blockchain wallets, VPN circumvention of geographic restrictions, and the difficulty of tracing tipping networks across pseudonymous accounts.
- Industry proponents argue the insider trading problem is a feature, not a bug: economist Robin Hanson, regarded as the godfather of prediction market theory, argues that insiders trading on privileged knowledge actually improves price accuracy and makes prediction markets more valuable as forecasting tools — insiders are simply people with the best information, and their trading moves prices toward truth faster than public information alone would; against this view, regulators and platforms are tightening surveillance, with Goldman Sachs banning employee prediction market trading for finance and politics, the US Senate unanimously banning member and staff trading on prediction markets, and the White House sending a staffwide email warning against using confidential information on event betting platforms; the top 1% of profitable wallets in flagged Polymarket trades captured more than half of all winnings, indicating that any “democratizing” effect of prediction markets is heavily offset by concentrated informational advantage.
What Happened?
Bloomberg Businessweek published an investigation finding approximately $200 million in flagged potential insider trades on Polymarket in the first half of 2026, heavily concentrated in geopolitical markets tied to the US-Iran conflict. The analysis reveals that 71% of Iran war betting volume came from US-regulated crypto exchanges, that 57% of profitable flagged wallets were created less than 24 hours before trades, and that a cluster of 38 connected wallets made $1.6 million betting on Iran and Venezuela military events with near-perfect accuracy before cashing out through the same Coinbase account.
Why It Matters?
Prediction markets have grown from niche curiosities into high-volume financial instruments with hundreds of millions of dollars at stake on geopolitical and military events. The Bloomberg investigation reveals that this growth has created a new class of insider traders — government officials, military personnel, and others with privileged access to information about real-world events — who can profit directly from classified or non-public knowledge. Unlike traditional securities insider trading, the jurisdictional, definitional, and enforcement challenges are substantially harder: anonymous blockchain wallets, pseudonymous identities, VPN circumvention, and the absence of clear legal frameworks make systematic enforcement nearly impossible at scale. The national security dimension — US soldiers and intelligence-adjacent personnel potentially trading on classified information — makes this a problem that goes well beyond financial market integrity.
What’s Next?
Watch the Van Dyke criminal case, which is the first major test of whether trading on classified military information in prediction markets constitutes a prosecutable offense — the outcome will signal how aggressively the US government will pursue future cases; watch whether the CFTC issues formal guidance on insider trading in prediction markets, which would give platforms clearer enforcement mandates and give users clearer rules; watch whether major financial institutions follow Goldman Sachs in restricting employee prediction market activity, which would reduce one category of potential insider trading from the finance sector; and watch Polymarket’s and Kalshi’s enforcement actions, particularly their referrals to law enforcement — the companies have referred nearly 100 wallets to law enforcement to date, a number that will grow significantly as the platforms scale and the sophistication of detection improves.
Source: Bloomberg Businessweek
