ESPN's AI Tool Claims to Read Poker Players' Tells—Pros Aren't Convinced

ESPN introduced a new artificial intelligence system during the 2026 World Series of Poker Main Event broadcast designed to detect what poker players call "tells"—the behavioral cues, gestures, and physical reactions that might hint at a player's hand strength or intentions. The tool sparked considerable discussion within the professional poker community about whether AI can meaningfully replicate one of the game's most prized human skills.

How the System Works#

Created by Luke Geel, an AI engineer in the U.S. Air Force, the tool analyzes camera footage from the tournament to build detailed profiles of individual players' behavioral patterns. The system tracks numerous visual metrics: eye movement, blink rate, posture, the way players handle chips, hand movements, and other physical indicators. By correlating these observations with the actual outcomes of hands, the AI generates predictions about what type of hand a player likely holds—whether it's a strong made hand, a drawing hand, or a bluff.

The tool appeared as an on-screen overlay during the early days of the broadcast in July, displaying both the tracked metrics and a "hand strength model" chart that offered probabilistic assessments of different hand categories.

The Small Sample Problem#

Poker professionals who spoke with WIRED expressed significant skepticism about the system's reliability, particularly given its training constraints. Although the 2026 Main Event attracted more than 9,000 entries, only a handful of tables were captured by the broadcast cameras—the same footage used to train the AI. Most players never appeared on screen at all, and even those who reached the final table didn't have enough accumulated camera time for the system to develop a comprehensive dataset covering the full range of poker situations.

Michael Gagliano, a 17-year poker professional who reached this year's final table competing for a $10 million prize, spent considerable time reviewing ESPN's archived broadcasts hunting for tells on his opponents. He concluded that the limited screen time for any single player severely restricted what could actually be learned. "I don't know how much actual information I'm going to be able to act on from what I saw," Gagliano noted.

Why Tells Are More Complex Than They Appear#

Tell detection involves far more subtlety than most casual fans appreciate, according to top-ranked pros. The 1998 film Rounders—where Matt Damon's character spots a tell based on Oreo cookie-eating habits—presents a misleading oversimplification of the skill.