CFTC Chairman Michael Selig Says AI Will Catch Insider Trading on Polymarket

The Commodity Futures Trading Commission is using artificial intelligence to hunt for insider trading on crypto prediction markets like Polymarket, with CFTC Chairman Michael Selig telling WIRED, "We’re going to find them, and we’re going to bring actions."

“For most of the past year, it looked like prediction markets had kicked off a new golden age of fraud”

Ars Technica

Selig said the agency is staffing up and leaning into automation, adding that "When we feed it into AI, we get really great information."

Ars Technica

Ars Technica reported that the CFTC wants to catch traders within the United States who are sneaking onto offshore markets, including Polymarket, by using virtual private networks.

The CFTC’s approach combines tools that analyze trading patterns with third-party blockchain tracing tools like Chainalysis and market abuse detection software including Nasdaq Smarts, according to Ars Technica.

In parallel, Startup Fortune described the Van Dyke case as a catalyst for enforcement, saying the Justice Department unsealed an indictment on April 23 alleging classified information about Operation Absolute Resolve was used to place wagers on Polymarket markets tied to Venezuela and Maduro.

The CFTC’s AI push is framed as part of a broader shift toward rapid detection, with TheLines saying the CFTC confirmed it is actively using AI tools to monitor prediction markets and crypto trading activity.

In that same account, TheLines quoted CFTC Chairman Mike Selig discussing how the agency uses artificial intelligence systems alongside blockchain analytics tools to scan markets for suspicious activity or insider trading patterns.

MEXC

MEXC reported that on Polymarket automated bots now run more than 30% of active accounts and that 14 of the top 20 accounts are controlled by bots.

MEXC also said the platform’s trading activity fell 8.9% in April for the first time since August, while Dune Analytics data showed the platform and its US operation registered $10.2 billion in bets in April.

As enforcement expands, MEXC said Selig stated the agency will take action against U.S. users who attempt to mask their location by utilizing VPNs to access prohibited services, and that the CFTC has received allegations of "hundreds to thousands" of potential cases.

The CFTC’s AI surveillance is also tied to specific enforcement and compliance mechanics, including the Van Dyke insider-trading case described by Startup Fortune, where prosecutors alleged he placed about 13 bets totaling roughly $33,034 and allegedly made about $409,881 in profit.

Startup Fortune said the Commodity Futures Trading Commission described the matter as its first insider trading case involving event contracts and alleged Van Dyke bought more than 436,000 Yes shares in a Maduro-related contract using classified nonpublic information.

To operationalize market integrity, Startup Fortune reported that on April 30 Polymarket selected Chainalysis to help deploy an on-chain market integrity system across its DeFi platform to monitor trading activity, enforce market integrity rules, support investigations, and build detection models for patterns suggesting insider knowledge.

Ars Technica added that the CFTC’s headquarters in Washington, DC is where Selig spoke to WIRED, and it described the agency’s tools as including both in-house systems and third-party tracing and detection software.

Meanwhile, Startup Fortune framed the compliance stakes as a test for prediction markets to prove they can police privileged information, emphasizing that the case forces the market to explain why it should be trusted with contracts tied to politics, war, companies and public decisions.