ESPN Tests AI That Can Detect When Someone Is Bluffing

An artificial intelligence (AI) poker-tells detector debuted during ESPN’s coverage of the World Series of Poker (WSOP) Main Event. The system flags likely bluffs using body language alone.

Independent AI engineer Luke Geel spent six months building the tool. Omaha Productions, the Peyton Manning-owned company behind the broadcast, applies it only to players who have already been eliminated.

How the AI Reads Poker Tells

The computer vision system tracks blink rate, eye gaze, posture, and how players handle their chips. It compares those patterns with results from earlier hands to estimate whether someone holds a strong hand.

Geel, who also develops AI for the US Air Force, announced the debut on LinkedIn. He described the project as harder than expected.

“It was significantly more difficult than I had initially hoped. I can’t just, like, upload a YouTube URL and say, ‘find their tells.'”

Geel said in an interview with Sportico.

Final table coverage airs on ESPN during the first week of August. The debut caps a busy month of AI headlines, days after Musk and Altman clashed over Apple’s lawsuit against OpenAI, as the global AI model race intensifies.

Players Split Between Cheating Claims and Curiosity

Reaction ran hot on Instagram and other platforms. Several players called the feature cheating and bad for the game, while others said they would run it on their own footage to remove personal tells.

Skeptics also question its purpose, since broadcasts already show every player’s hole cards. Meanwhile, Geel said multiple players asked about using the model to study future opponents.

The same approach could eventually read negotiators, job candidates, or sales prospects. That prospect revives earlier warnings about the dark side of AI, even as systems that let AI agents trade crypto enter other high-stakes settings.

Whether audiences treat the feature as insight or intrusion may decide if it returns next season.