Google Engineer Charged With Using Confidential Year in Search Data to Win $1.2 Million on Polymarket Contracts
Federal authorities charged a Google software engineer with insider trading after he allegedly used nonpublic Year in Search data to generate more than $1.2 million in profits on Polymarket event contracts. The Commodity Futures Trading Commission filed a civil complaint against Michele Spagnuolo on Wednesday. The U.S. Attorney’s Office for the Southern District of New York brought separate criminal charges of commodities fraud, wire fraud and money laundering.
The case centers on at least 23 Google-related prediction market contracts. Authorities say the trading showed near-perfect accuracy. This is the latest test of whether prediction market platforms can police themselves when real-world information asymmetry collides with liquid event contracts.
Access to Nonpublic Data and the Trading Pattern
Michele Spagnuolo worked as a software engineer at Google. He had access to internal tools that contained the nonpublic 2025 Year in Search List data marked Google Confidential. Prosecutors say Google viewed the information as commercially sensitive because the annual rankings drive advertising engagement, media attention and user traffic.
The AlphaRaccoon account purchased shares across contracts tied to Kendrick Lamar, Pope Leo XIV, Bianca Censori, Jimmy Kimmel and d4vd. Between October and December 2025 the account risked approximately $2.75 million on Google-related contracts. Some of those positions sat on outcomes the broader market viewed as highly unlikely.
FBI investigators noted that the market assigned to d4vd becoming Google’s most-searched person carried a near-zero probability. The account still won. That level of calibration does not happen by chance.
How the AlphaRaccoon Activity Looked to Outside Observers
The AlphaRaccoon account had drawn attention inside prediction market communities months before the charges surfaced. Polywhaler, an account that tracks large Polymarket wallets, flagged the trader roughly six months earlier because of unusually successful Google-related trades.
Polywhaler flagged the recently arrested Google insider trader six months ago while everyone else missed it. Screenshots shared by the account showed a reported 100 percent win rate across resolved trades and more than $1.3 million in profit.
The largest reported positions included approximately $474,360 tied to Pope Leo XIV, roughly $193,127 tied to d4vd-related contracts and approximately $165,943 linked to Bianca Censori. Profitable positions also sat on contracts involving Donald Trump, Kendrick Lamar, Jimmy Kimmel, Luigi Mangione and Tyler Robinson.
After eighteen years across iGaming and sportsbook operations I have seen sharp money move lines for less visible edges than this. When one trader prints at that clip on a narrow theme the pattern eventually surfaces.
Regulatory and Platform Response
Polymarket positioned the outcome as validation of its own monitoring rather than a failure. U.S. Attorney Jay Clayton reinforced a decades-old message that corporate insiders cannot use confidential business information to turn a profit in our markets. CFTC Chairman Michael Selig stated that the Commission will not tolerate fraud, manipulation or insider trading regardless of the technology or platform that is used.
This marks the second major insider trading-related case connected to Polymarket. Earlier in 2026 federal prosecutors charged U.S. Army Special Forces soldier Gannon Ken Van Dyke with using nonpublic information related to Venezuelan President Nicolás Maduro to profit more than $400,000 on related contracts.
Risk, Precedent and the Integrity Debate
Kalshi has faced similar issues. In February the platform disclosed that it had closed two insider trading cases and referred both matters to the CFTC. One involved former California gubernatorial candidate Kyle Langford. Another involved Artem Kaptur, an editor working on content for YouTube creator MrBeast.
In April Kalshi published three additional enforcement notices tied to political candidates who traded on markets linked to their own elections. One candidate, Mark Moran, said he wanted to get caught to expose systemic conflicts of interest and manipulation across prediction market platforms.
The incidents form part of a wider integrity debate. Prediction markets trade on real-world outcomes that insiders can sometimes see first. That creates structural risk no amount of post-trade flagging fully removes.
A counterargument holds that these cases remain rare and that platforms are improving detection. The limitation is clear. Once an insider decides to act the informational edge can be decisive before any algorithm or community watcher catches up. The allegations against Spagnuolo remain accusations and he has not been convicted of any offense.
The Bottom Line
Prediction markets now sit at meaningful scale and draw real money. That scale inevitably attracts sophisticated actors who possess nonpublic data. For industry executives the takeaway is operational. Platforms must invest in pre-trade pattern detection, stricter information barriers with corporate partners and faster information sharing with regulators. Operators and data partners should treat these incidents as early signals rather than isolated anomalies. The next twelve months will show whether the integrity infrastructure scales with the liquidity or whether repeated cases erode trust before the category reaches its full commercial potential.
