A senior engineer who has worked at Google for more than 12 years used the company’s undisclosed “Year in Search” internal data to accurately place bets on the prediction market platform Polymarket and won more than US$1.2 million (approximately NT$38 million). The U.S. Federal Prosecutor for the Southern District of New York announced on May 27, 2026 that 36-year-old Google software engineer Michele Spagnuolo was charged with three counts of commodity fraud, telecommunications fraud and money laundering. If convicted, he will face decades in prison. The case is the second recent prosecution by the U.S. Department of Justice targeting insider trading in prediction markets, and highlights the long-standing regulatory gray area issues at prediction trading platforms such as Polymarket.
Google Engineer Wins $1.2 Million on Polymarket Using Internal Data, Arrested on Insider Trading Charges
According to the indictment, Spagnuolo is an Italian citizen who lives in Switzerland and works as a security engineer at Google. He opened an account on Polymarket in 2024 under the username “AlphaRaccoon” and used cryptocurrency to place bets from multiple accounts.
Between October and December 2025, Spagnuolo invested a cumulative approximately $2.75 million in bets related to Google search trends. He used internal Google tools to access unpublished Yearly Search data to predict which celebrities and public figures would appear on Google’s annual trending searches.
One of the most notable bets: Spagnuolo accurately predicted that indie musician D4vd would be the most searched person globally on Google in 2025. On Polymarket, the odds of D4vd appearing at the top of the search list were almost zero, but Spagnuolo was still betting heavily because he knew D4vd had already become the number one search on Google of the year. But the reason why D4vd has become the most popular search topic on Google in 2025 is lamentable: because ofSuspected of murdering 14-year-old girl Celeste Rivas-Hernandez was arrested and her body was found in the back of her towed Tesla. This horrific case attracted global attention and caused a surge in searches for D4vd.
A Google spokesperson said Spagnuolo was using a “marketing data tool that all employees have access to” but that “using this confidential information to place bets is a serious violation of company policy.” Google has suspended Spagnuolo and is cooperating with law enforcement agencies in their investigation.
During the FBI investigation, although Spagnuolo used cryptocurrency to conduct transactions from multiple accounts, agents were able to link all transactions to the same person through one of the accounts opened with an Italian ID card. Spagnuolo was arrested in New York and later released on $2.25 million bail.
Jay Clayton, U.S. Attorney for the Southern District of New York, said: “Spagnuolo violated his obligations to his employer by using Google’s confidential business information to obtain more than $1.2 million in trading profits on Polymarket. Insider trading undermines the integrity of our markets, and the American people demand that this greed-driven conduct be investigated and prosecuted.”
Polymarket’s Insider Trading Dilemma
Polymarket is a cryptocurrency-based prediction market platform that allows users to place bets on political events, world events, sporting events and even online trends. Unlike traditional stock trading, prediction markets have long existed in a legal gray area. Supporters view them as effective predictive tools, while critics view them as gambling sites that lack insider trading protections.
A Polymarket spokesperson responded to the case: “Polymarket is working closely with the U.S. Attorney’s Office for the Southern District of New York and the Commodity Futures Trading Commission (CFTC). We are currently the only
Prediction platform that cooperated with investigations leading to insider trading prosecutions. Blockchain transactions are transparent and traceable, and bad actors leave footprints. “
This is not the first time Polymarket has faced insider trading controversy. Earlier in 2026, a U.S. Army soldier was indicted for using his inside knowledge of military operations involving the capture of Venezuelan President Maduro to win $400,000 on Polymarket. In March 2026, CBS News reported that a number of questionable bets related to geopolitical events had prompted calls from members of Congress for greater regulation.
Blockchain transparency: a double-edged sword
Ironically, although Spagnuolo used cryptocurrency and multiple accounts to try to hide his identity, the transparency of the blockchain ended up being a key clue for investigators to track him down. All transactions on Polymarket are recorded on the blockchain and can be reviewed by anyone. As early as December 2025, Polymarket users noticed the abnormal trading behavior of the “AlphaRaccoon” account on contracts related to the “Most Searched Person of the Year” and marked it as suspicious.
FBI agents eventually linked all the scattered cryptocurrency transactions to the same person through one of the accounts Spagnuolo opened with an Italian ID. This proves one thing: trying to be completely anonymous in the on-chain world is far more difficult than many people imagine. Every transaction is a permanent record, and every wallet address may become a clue to your true identity.
Regulatory outlook for prediction markets
The successive cases of insider trading have once again brought the regulatory issues of the prediction market to the forefront. An analysis published by Stanford Law School in April 2026 noted that the rapid growth of prediction market participants is accompanied by concerns about regulatory manipulation and insider trading. Some legal scholars believe that prediction market contracts create new ways for people with privileged information to profit quietly before the public becomes aware of it.
For Google, this incident also sounded the alarm about internal information security. As an information security engineer, Spagnuolo was responsible for protecting the company’s data, but he instead used the access rights for personal gain. Google emphasizes that the tool is “open to all employees,” but this also raises a question: In the era of AI, when more and more internal company data have market value, how to balance employees’ information access rights and prevent abuse will become a governance problem that all technology companies must face.