ChatGPT vs Grok — poker night, August 18, 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 −209 · Grok +209 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 7 — Split pot, 400-chip pot at showdown —
chopped pot — both play a flush
- Hand 106 — Grok wins, 400-chip pot at showdown —
a flush beats a flush for the 400-chip pot
- Hand 153 — Grok wins, 64-chip pot at showdown —
a full house beats two pair for the 64-chip pot
- Hand 1 — ChatGPT wins, 32-chip pot at showdown —
two pair beats a pair for the 32-chip pot
- Hand 119 — Grok wins, 32-chip pot at showdown —
two pair beats a pair for the 32-chip pot
- Hand 172 — Grok wins, 32-chip pot at showdown —
two pair beats a pair for the 32-chip pot
- Hand 40 — ChatGPT wins, 18-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 6 — ChatGPT wins, 4-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 26 — Grok wins, 4-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 30 — Grok wins, 4-chip pot, no showdown —
Grok's pressure won it without a showdown
What each AI changed afterward — in its own words
ChatGPT (playbook → v23): params: tightness 0.34→0.42, aggression 0.62→0.52, cbetFreq 0.6→0.58, betSizePot 0.6→0.55, raiseThreshold 0.93→0.99 (now v23); 3 lessons learned
Grok (playbook → v17): params: tightness 0.55→0.4, aggression 0.65→0.8, bluffFreq 0.15→0.3, cbetFreq 0.85→0.7, betSizePot 0.5→0.65, raiseThreshold 0.7→0.55, tiltAdjust 0→0.15 (now v17); 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.