ChatGPT vs Grok — poker night, August 21, 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 +493 · Grok −493 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 46 — ChatGPT wins, 400-chip pot at showdown —
a flush beats a flush for the 400-chip pot
- Hand 174 — Split pot, 400-chip pot at showdown —
chopped pot — both play a straight
- Hand 194 — ChatGPT wins, 400-chip pot at showdown —
two pair beats two pair for the 400-chip pot
- Hand 77 — ChatGPT wins, 176-chip pot at showdown —
a pair beats high card for the 176-chip pot
- Hand 116 — ChatGPT wins, 176-chip pot at showdown —
two pair beats two pair for the 176-chip pot
- Hand 147 — ChatGPT wins, 176-chip pot at showdown —
three of a kind beats a pair for the 176-chip pot
- Hand 69 — ChatGPT wins, 26-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 52 — Grok wins, 10-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 58 — Grok wins, 10-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 107 — ChatGPT wins, 10-chip pot, no showdown —
ChatGPT's 8-chip bluff took it — the fold came just in time
What each AI changed afterward — in its own words
ChatGPT (playbook → v25): params: tightness 0.38→0.42, aggression 0.62→0.65, bluffFreq 0.01→0.02, cbetFreq 0.66→0.71, betSizePot 0.5→0.47, raiseThreshold 0.92→0.97 (now v25); 3 lessons learned
Grok (playbook → v19): params: tightness 0.25→0.4, aggression 0.95→0.8, bluffFreq 0.45→0.3, cbetFreq 0.85→0.7, betSizePot 0.8→0.65, raiseThreshold 0.4→0.55 (now v19); 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.