Google Engineer’s $1.2M Polymarket Win Exposes Prediction Market Enforcement Gap
Federal charges against Michele Spagnuolo prove existing fraud laws can reach crypto-native platforms, but reveal why regulators lack surveillance infrastructure to detect insider trading before trades settle.
A Google information security engineer netted $1.2 million on Polymarket by trading on confidential search trend data before public release, according to charges unsealed May 28, marking the second prediction market insider trading case in six weeks and exposing critical gaps in how regulators police crypto-native finance.
Michele Spagnuolo, a 13-year Google veteran, allegedly placed $2.75 million in bets between October 15 and December 4, 2025, using proprietary ‘Year in Search’ data to predict which topics would trend before Google’s public announcement. According to NBC News, the complaint states Spagnuolo transferred $3.8 million in USDC to a Polymarket address, settling trades hours before information went public and realising $1.2 million in profit.
The charges — commodities fraud, wire fraud, and money laundering — demonstrate that existing statutes can be retrofitted to cover Prediction Markets, yet the case arrived only after settlement and platform cooperation. “Unlike the counterparties to his trades, Spagnuolo knew the outcome of these wagers before the trading public did because he had accessed Google’s confidential, commercially valuable internal data,” FBI Special Agent Brandon Racz wrote in the complaint, as cited by CoinDesk.
Second Case in Six Weeks Signals Pattern
The Spagnuolo prosecution follows charges filed April 23 against U.S. Army soldier Gannon Van Dyke, who allegedly wagered $33,000 on Polymarket using classified military intelligence about a Venezuela operation and realised roughly $400,000 in profit, according to Sidley Austin legal analysis. Both cases apply the ‘Eddie Murphy Rule’ — a 2010 amendment extending commodities fraud prohibitions to cover misappropriated nonpublic information — to crypto-native platforms that were not contemplated when the statute passed.
The rapid succession of cases prompted House Oversight Committee Chair James Comer to launch a formal investigation on May 22 into whether government employees are exploiting insider knowledge on prediction platforms. “There’s a concern now that members of Congress, members of the president’s administration, any type of government employee, can use basic insider knowledge and make huge profits on anything government-related,” Comer told CNBC.
Regulatory Framework Lags Market Growth
Prediction market volumes reached $51 billion in 2025 and could hit $240 billion in 2026, according to CoinDesk reporting on a Bernstein analysis. Yet the CFTC’s surveillance infrastructure remains built for traditional commodity markets, not for policing pseudonymous on-chain trading where participants can transfer stablecoins across wallets without triggering know-your-customer checks until settlement.
“Prediction markets are not a haven for using misappropriated confidential or classified information for personal gain. This is clear insider trading and it is illegal under federal law.”
— Jay Clayton, U.S. Attorney for Southern District of New York
CFTC Chairman Michael Selig announced January 29 that the agency would withdraw its 2024 proposed rule prohibiting event contracts and instead promulgate ‘clear rules’ to regulate prediction markets, according to Paul Weiss analysis. The CFTC issued a Prediction Markets Advisory on February 25 stating that exchanges have an ‘independent duty’ to maintain audit trails and enforce rules against prohibited practices, but stopped short of mandating pre-trade surveillance systems comparable to those required of stock exchanges.
The CFTC’s authority over prediction markets derives from classifying event contracts as swaps under the Commodity Exchange Act. Unlike the SEC’s comprehensive insider trading framework for securities, commodities law historically focused on manipulation and false reporting rather than information asymmetries. The 2010 Dodd-Frank Act added prohibitions on trading based on misappropriated nonpublic information, but enforcement relies on post-trade detection rather than real-time surveillance mandates imposed on stock exchanges.
Data Access as New Insider Vector
The Spagnuolo case highlights a structural vulnerability: tech companies grant employees access to proprietary data signals — search trends, user engagement metrics, model training outcomes — that function as leading indicators for public events. Spagnuolo’s role as an information security engineer gave him ‘Google Confidential’ access to Year in Search data weeks before publication, creating an exploitable window unavailable to external traders.
This pattern differs from traditional insider trading, where material nonpublic information typically concerns a single issuer’s financial performance. A Google employee with access to search trend data holds signals relevant to dozens of potential prediction market outcomes — trending public figures, viral topics, geopolitical events — without triggering the relationship-based disclosure requirements that govern corporate insiders under securities law.
- No pre-trade surveillance mandate: CFTC advisory delegates monitoring to platforms without specifying technical requirements or algorithmic detection thresholds.
- Stablecoin transfer opacity: USDC and USDT transfers across wallets occur without triggering AML checks until fiat conversion or settlement.
- Cross-platform arbitrage: Traders can place correlated bets on Polymarket, Kalshi, and offshore platforms simultaneously, fragmenting audit trails.
- Data compartmentalisation failures: Tech firms lack standardised frameworks for flagging employees with access to prediction-market-relevant proprietary signals.
Google responded to the charges with a statement emphasising that the company ‘has strict policies and technical controls to prevent the misuse of confidential information,’ but declined to specify whether Spagnuolo’s data access underwent regular audits or whether the company maintains automated monitoring for employee trading activity on prediction platforms.
What to Watch
The CFTC faces pressure to publish specific rulemaking proposals by Q3 2026, following Chairman Selig’s January commitment to establish ‘clear rules.’ Congressional scrutiny will likely accelerate this timeline — the House Oversight investigation could produce testimony from Polymarket and Kalshi executives as early as July, forcing platforms to disclose the extent of their internal surveillance capabilities.
For prediction markets, the immediate question is whether platforms will implement voluntary pre-trade monitoring systems before regulatory mandates arrive. Polymarket’s cooperation with federal prosecutors suggests a strategic shift toward building compliance infrastructure that could become an industry standard. The company stated it is ‘the only prediction platform to date whose cooperation has led to insider trading charges,’ according to TechCrunch.
Technology companies now confront a data governance problem: as AI systems and proprietary algorithms generate increasingly granular signals about future events, employee access to these systems creates exploitable information asymmetries. The enforcement gap persists because regulators lack both the technical infrastructure to detect pseudonymous on-chain trading in real time and the statutory authority to mandate pre-emptive surveillance. Until the CFTC establishes specific audit trail requirements and cross-platform data-sharing protocols, prediction market enforcement will remain reactive — dependent on platform cooperation and post-settlement forensics rather than algorithmic detection systems that could flag suspicious trading patterns before profits are realised.