Polymarket Users Pocket $37K After Paris Weather Feed Error Sparks Integrity Concerns

Two Polymarket accounts have come under scrutiny after pocketing more than $37,000 from prediction markets tied to two unusual temperature readings at a weather station located inside France's largest airport. The markets tracked the highest temperature in Paris on April 6 and April 15, using data from the Charles de Gaulle Airport station, measured in degrees Celsius.

French outlet BFMTV reported on Monday that the April 6 reading briefly climbed above 21 degrees Celsius before dropping back down. That market resolved with a single winner collecting more than $16,000.

Blockchain analytics platform Bubblemaps flagged a near-identical pattern for the April 15 contract. The sensor held at roughly 18 degrees Celsius for most of the day before spiking to 22 degrees Celsius and sliding back shortly afterward.

"That spike didn't show on nearby stations," Bubblemaps analysts wrote in a post on X, comparing the Charles de Gaulle reading with outputs from other French weather sensors.

The firm noted that moments before the jump, a single trader began accumulating "NO" shares on the 18-degree outcome, later cashing out with more than $21,000. The episode arrives as prediction markets face heightened scrutiny worldwide over alleged insider trading and possible violations of gambling laws.

Ruben Hallali, a meteorologist interviewed by BFMTV, said the timing and magnitude of the readings made a natural explanation difficult to accept.

He described the two-degree swings on those specific dates as highly improbable over such short windows, adding that the pattern suggested someone familiar with the sensors may have intervened to validate the bets.

Météo France, the country's official government weather agency, has reportedly filed a formal complaint with the Roissy Air Transport Gendarmerie Brigade. The complaint alleges tampering with the operation of its automated data processing systems.

The case has reignited a long-running industry debate about the reliability of real-world data feeds, known as oracles, that underpin blockchain-based prediction markets. A single compromised input can unilaterally determine the outcome of thousands of dollars in wagers, leaving platforms exposed to reputational and potential legal risk.

Operators typically rely on authoritative feeds or multiple providers to reduce the risk of a single point of failure. Polymarket has not publicly addressed whether it will refund wagers or revise how weather data is sourced. The platform has previously adjusted market outcomes when external information proved inaccurate, though such interventions remain rare.

For now, French investigators will determine whether the two spikes were genuine instrument faults or the result of coordinated manipulation. The outcome could have implications well beyond the $37,000 payout, potentially shaping how prediction markets verify information from physical sensors as the sector pursues broader regulatory acceptance across Europe and the United States.