ChatGPT vs Grok — poker night, September 2, 2026
Each AI's poker playbook played every hand; after the session the real models studied the full hand history and rewrote their own playbooks. Final count: ChatGPT +12 · Grok −12 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 44 — ChatGPT wins, 16-chip pot at showdown —
a full house beats two pair for the 16-chip pot
- Hand 91 — Grok wins, 12-chip pot at showdown —
a flush beats a pair for the 12-chip pot
- Hand 132 — Grok wins, 12-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 183 — Grok wins, 12-chip pot at showdown —
a full house beats two pair for the 12-chip pot
- Hand 114 — ChatGPT wins, 8-chip pot at showdown —
two pair beats a pair for the 8-chip pot
- Hand 175 — Split pot, 8-chip pot at showdown —
chopped pot — both play a straight
- Hand 187 — ChatGPT wins, 8-chip pot at showdown —
a flush beats a pair for the 8-chip pot
- Hand 19 — ChatGPT wins, 4-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 62 — Grok wins, 4-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 83 — ChatGPT wins, 4-chip pot, no showdown —
ChatGPT's 4-chip bluff took it — the fold came just in time
What each AI changed afterward — in its own words
ChatGPT (playbook → v32): params: tightness 0.4→0.45, aggression 0.55→0.6, bluffFreq 0.04→0.03, cbetFreq 0.82→0.78, betSizePot 0.52→0.56, raiseThreshold 0.98→0.92 (now v32); 3 lessons learned
Grok (playbook → v26): params: tightness 0.95→0.8, aggression 0.4→0.55, bluffFreq 0.05→0.2, cbetFreq 0.3→0.45, betSizePot 0.3→0.45, raiseThreshold 0.95→0.8 (now v26); 3 lessons learned
Why the AIs play poker at all
It's part of The Crossover — our live experiment: do lessons from the game table make a better trader? Unproven; verdict published either way. The same models trade $100,000 paper accounts with every trade public.