Federal authorities have charged a Google software engineer with insider trading for allegedly using confidential company data to generate more than $1.2 million in profits trading prediction market contracts on Polymarket. The case marks one of the most significant enforcement actions ever brought against a prediction market trader — and the first to involve a corporate insider allegedly exploiting proprietary employer data.
The Commodity Futures Trading Commission filed a civil complaint on Wednesday against Michele Spagnuolo, an Italian citizen. Simultaneously, the US Attorney’s Office for the Southern District of New York unsealed criminal charges of commodities fraud, wire fraud and money laundering. US Attorney Jay Clayton was direct in his statement: “Today’s charges reinforce a decades-old message: corporate insiders cannot use confidential business information to turn a profit in our markets.”
Table of Contents
The AlphaRaccoon account
Spagnuolo allegedly operated under the alias “AlphaRaccoon” on Polymarket, purchasing shares across at least 23 Google-related contracts with what the CFTC described as “near-perfect accuracy.” Federal authorities allege the trading generated approximately $1.2 million in profits — with Spagnuolo risking approximately $2.75 million across Google-related contracts between October and December 2025.
The alleged edge came from Spagnuolo’s access, through his role as a Google software engineer, to internal tools containing nonpublic data from Google’s 2025 Year in Search list — a compilation of the year’s most-searched terms and people, marked “Google Confidential.” Prosecutors alleged the company treats this information as commercially sensitive given its impact on advertising engagement, media attention and user traffic.
The contracts at the centre of the case
Among the markets referenced in the complaints were contracts involving Kendrick Lamar, Pope Leo XIV, Bianca Censori, Jimmy Kimmel and d4vd. FBI investigators concluded that the market assigning d4vd as Google’s most-searched person of the year had been at “near-zero probability” when Spagnuolo allegedly took his position — yet it resolved in his favour.
The largest individual profits included approximately $474,360 tied to a market on whether Pope Leo XIV would be Google’s most-searched person of the year, around $193,127 from d4vd-related contracts, and approximately $165,943 from markets related to Bianca Censori. Investigators traced cryptocurrency transfers connected to the AlphaRaccoon account across multiple wallets and services before linking the activity back to Spagnuolo.
The community had already noticed
What makes the case particularly notable is that the AlphaRaccoon account had already attracted significant attention within prediction market communities long before charges were filed. Polywhaler, an account that tracks large Polymarket wallets and trading activity, stated it had flagged the account approximately six months earlier due to its unusually consistent success on Google-related markets — posting a reported 100% win rate across resolved trades and generating more than $1.3 million in profit.
“We flagged the recently arrested Google insider trader 6 months ago, while everyone else missed it,” the account wrote following the charges.
Polymarket itself claimed credit for the detection, stating on X: “Proud to announce Polymarket’s market integrity infrastructure flagged another trader who was arrested this morning in New York for insider trading.”
The broader implications
CFTC Chairman Michael Selig used the case to send a broader message to the prediction market industry. “The Commission will not tolerate fraud, manipulation, or insider trading, regardless of the technology or platform that is used,” he said.
The AlphaRaccoon case is the clearest demonstration yet that the insider trading risks long associated with financial markets are fully present in prediction markets — and that the stakes are now large enough to attract federal criminal enforcement. For Polymarket, the case is a double-edged moment: it demonstrates the platform’s market integrity infrastructure is capable of flagging suspicious activity, but it also confirms that the platform has been used to profit from confidential corporate information at scale.
For the prediction market industry as a whole, the case adds significant weight to the regulatory argument that event-based contracts require the same insider trading protections as conventional financial instruments — and that the current frameworks, in many jurisdictions, are not yet equipped to provide them.
