ESPN has introduced an artificial intelligence system that flags likely bluffs by analyzing poker players' body language during its World Series of Poker Main Event coverage. The tool, built by independent AI engineer Luke Geel over six months, is being applied only to players who have already been eliminated from the tournament, under an arrangement with Omaha Productions, the company owned by Peyton Manning.
How the AI reads players
The computer vision system tracks blink rate, eye gaze, posture, and chip handling, then compares those patterns with earlier hands to estimate hand strength. Because the AI is used only on eliminated players, it doesn't affect the outcome of the tournament — but it does give viewers a new window into the psychology of the game. Geel, who also develops AI for the US Air Force, built the model using publicly available footage and training data from past WSOP events.
Mixed reactions from the poker world
Players have responded with a split of opinions. Some called the technology a form of cheating, arguing that it undermines the skill of reading opponents. Others expressed curiosity about using it to study their own tells and improve their game. Skeptics question the AI's purpose since broadcasts already show every player's hole cards — making the tells less relevant for viewers who already know the hands. Geel noted that multiple players asked about using the model to study future opponents, suggesting a demand for the tool beyond entertainment.
Broader implications beyond poker
The same approach could eventually be used to read negotiators, job candidates, or sales prospects, Geel said. That possibility revives warnings about the dark side of AI in high-stakes settings, where automated body-language analysis could be used to gain an unfair advantage or invade privacy. The final table coverage, which will include the AI detector, airs on ESPN during the first week of August.
