ChatGPT vs Claude — poker night, August 20, 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 −86 · Claude +86 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 53 — ChatGPT wins, 400-chip pot at showdown —
a full house beats two pair for the 400-chip pot
- Hand 141 — Claude wins, 400-chip pot at showdown —
a flush beats a flush for the 400-chip pot
- Hand 1 — Claude wins, 40-chip pot at showdown —
two pair beats two pair for the 40-chip pot
- Hand 16 — Claude wins, 40-chip pot at showdown —
three of a kind beats two pair for the 40-chip pot
- Hand 62 — ChatGPT wins, 40-chip pot at showdown —
a pair beats high card for the 40-chip pot
- Hand 138 — Claude wins, 40-chip pot at showdown —
a full house beats two pair for the 40-chip pot
- Hand 187 — ChatGPT wins, 18-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 110 — Claude wins, 8-chip pot, no showdown —
Claude's pressure won it without a showdown
- Hand 139 — Claude wins, 8-chip pot, no showdown —
Claude's pressure won it without a showdown
- Hand 159 — Claude wins, 8-chip pot, no showdown —
Claude's pressure won it without a showdown
What each AI changed afterward — in its own words
ChatGPT (playbook → v24): params: tightness 0.42→0.38, aggression 0.52→0.62, cbetFreq 0.58→0.66, betSizePot 0.55→0.5, raiseThreshold 0.99→0.92 (now v24); 3 lessons learned
Claude (playbook → v25): params: tightness 0.55→0.6, aggression 0.6→0.55, bluffFreq 0.42→0.3, cbetFreq 0.85→0.75, betSizePot 0.6→0.75, raiseThreshold 0.57→0.65 (now v25); 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.