A Google Engineer Made $1.2M Insider Trading on Polymarket

Michelle Spagnuolo, a 36-year-old Italian citizen living in Switzerland and employed as a Google engineer, used confidential internal Google data to place bets on the prediction market Polymarket under the username Alpha Raccoon, generating over $1.2 million in profit. He bet on David being named Google's most searched person of the year, entering positions when the market assigned David roughly a 0.1% probability on November 28, before odds climbed to 4%, then 8%, and eventually around 10% just before resolution. Spagnuolo faces one count each of commodities fraud, wire fraud, and money laundering under a federal complaint unsealed in New York City, with prosecutors alleging he took deliberate steps to conceal the source and ownership of his proceeds.

A separate Polymarket account named Betwick, with no known connection to Spagnuolo, turned approximately $3,000 into $125,000 by betting yes on David when the market implied a 2% probability, raising unresolved questions about whether additional individuals had advance knowledge of the same internal Google data.

This case is described as the second significant criminal matter tied to Polymarket trades. A US soldier was charged in April for betting that American forces would capture Nicolas Maduro using information about a planned military operation, making more than $400,000 in profit. Israeli authorities separately charged a reservist and a civilian with using classified military information to profit from Polymarket bets tied to the timing of Israeli strikes against Iran, allegedly netting $150,000. A CBS 60 Minutes segment working with Bubble Maps also identified certain accounts on US-related military action markets on Polymarket with a 98% win rate, suggesting the insider trading problem may be broader than the charged cases alone.

Insider trading has surfaced on competing prediction market Kalshi as well. A Mr. Beast editor was caught trading on a video outcome he already knew and made approximately $5,000. Former California gubernatorial candidate Carl Langford was banned for five years and fined $2,000 after betting on his own race and promoting it on the platform. These cases are smaller in scale but indicate the issue is not confined to a single platform or category of market.

The episode surfaces a genuine tension in how prediction markets should handle insider participation. Byron Gilliam argued that prediction markets can only deliver better information to the world if insiders are free to trade, and that allowing company insiders to bet on prediction markets could make odds a real-time barometer of a company's fortunes, potentially replacing quarterly earnings reports as a source of market-moving information. David Nellis countered that Spagnuolo's trades involved buying cheap tokens from counterparties who lacked the same information, which raises direct questions about market fairness regardless of whatever informational benefits insiders might theoretically provide to the broader market.

The cases collectively point to a structural challenge for prediction markets as they grow in scale and legitimacy. The same feature that makes these markets potentially valuable, namely that people with superior knowledge will express it through prices, also creates conditions where uninformed participants are systematically exploited by those with privileged access. Whether regulators treat prediction market insider trading as equivalent to securities fraud, or develop a distinct framework, remains an open question given that the legal charges against Spagnuolo rely on commodities fraud and wire fraud rather than traditional insider trading statutes.