ESPN's poker AI claims to spot bluffs — but pros aren't buying it

espn deployed an ai tool during the 2026 world series of poker main event that uses computer vision to analyze player biometrics, such as eye movements, posture, and fidgeting, to predict the likelihood of bluffs or strong hands. designed by air force engineer luke geel, the system aimed to digitize the human skill of spotting tells by processing footage from the broadcast's featured tables. however, professional poker players, including final table competitor michael gagliano, are skeptical of the tool's utility because the ai trained on a small sample size, as most of the 9,000 entrants never appeared on camera. this highlights the challenge of applying ai to complex, human-centric behaviors where data scarcity is an issue. the industry reaction suggests that while ai can entertain viewers, it currently lacks the data depth to replace human intuition in high-stakes environments.