Polymath Poker: Why AI Strategy Fails: Lessons from High-Stakes Poker

In mid-2026, artificial intelligence presents a paradox. AI models and agents have outrun every forecast and timeline, while enterprise deployments have disappointed. Individuals using AI report doubling their productivity; most enterprises are fighting for a few percent.Understanding that paradox — and rethinking strategy around it — is the most important task facing executives today.Polymath Poker translates the language of high-stakes poker into a practical way of thinking about artificial intelligence, markets, leadership, and human capital. It deploys ideas missing from standard AI strategy discussions: expected value, Bayesian updating, range thinking, accelerated learning, risk of ruin, cognitive biases, peak mental performance, incentives, moats, and decision quality under uncertainty.The starting premise is that intelligence is only the ante: edge belongs to the people who build a harness around intelligence — whether that intelligence be human or machine. The harness, in this view, isn’t a restrictive cage; it is what lets force become work.Later, the book dispels the notion that you can rent innovation through SaaS AI, dispels the “myth of the moat,” and shows how to call “consultants’ bluffs.”Polymath Poker also examines poker’s staking economy — how circular capital flows may be risk concentration masquerading as diversification. Separately, it argues that leaders in volatile conditions should trade short-term optimality for longer-term optionality.Why poker. High-stakes poker used to be sold as bravado: colourful characters, trash talk, soul reads, and televised theatre. That game is mostly gone. The best players now train their minds the way athletes train their bodies. Their decisions are grounded in game theory, probability, risk psychology, bankroll discipline, emotional control, and ruthless review of their own mistakes — the habits AI strategy now requires.This is not a book about playing cards. It is an AI strategy book for CEOs, founders, investors, operators, and serious competitors who need better language for why AI programmes fail, why smart people still make poor bets, and why the difference between a good decision and a good outcome matters more than ever.This WSOP Pre-Release Edition is a limited first run, released live at the 2026 World Series of Poker — early, a little raw by design, and closest to the table. It’s the version tested in the same environment that inspired it. The formal edition comes later; this is the one that was actually at the felt.Paul Gibbons is an AI strategist, former IBM partner, former derivatives trader, and author of eight books on change, leadership, and human capital. He also competes at the highest levels of tournament poker, with over $1 million in live earnings, including a 37th-place finish in the 2025 WSOP Main Event.Inside the book:Introduction — The Hand Everyone RemembersChapter 1 — The Poker Laboratory: Luck, Samples, and the Amateur’s DelusionChapter 2 — Smart Is the Ante: Why Intelligence Needs a HarnessChapter 3 — Stop Using McKinsey’s Vocabulary: Poker Words That Help StrategistsChapter 4 — The Game Has Changed: From Solvers to Cognitive AthleticsChapter 5 — You Have No MoatChapter 6 — The Staking Economy and the Capital Structure of AIChapter 7 — Calling Consultants’ BluffsClosing — The Adaptive FoldAppendices: Adaptive Adoption and the Corpus · The Polymath at the Table · Poker Writers, Thinkers, and Fellow Travellers Read more