ChatGPT vs Claude — poker night, September 1, 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 +188 · Claude −188 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 169 — ChatGPT wins, 400-chip pot at showdown —
a flush beats a flush for the 400-chip pot
- Hand 13 — Claude wins, 28-chip pot at showdown —
a full house beats two pair for the 28-chip pot
- Hand 78 — Claude wins, 10-chip pot at showdown —
a flush beats a pair for the 10-chip pot
- Hand 127 — Claude wins, 10-chip pot at showdown —
two pair beats high card for the 10-chip pot
- Hand 75 — Claude wins, 8-chip pot at showdown —
a pair beats a pair for the 8-chip pot
- Hand 100 — Claude wins, 8-chip pot, no showdown —
Claude's pressure won it without a showdown
- Hand 112 — Claude wins, 8-chip pot at showdown —
two pair beats a pair for the 8-chip pot
- Hand 7 — ChatGPT wins, 4-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 11 — ChatGPT wins, 4-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 21 — ChatGPT wins, 4-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
What each AI changed afterward — in its own words
ChatGPT (playbook → v31): params: tightness 0.32→0.4, aggression 0.6→0.55, bluffFreq 0.07→0.04, cbetFreq 0.88→0.82, betSizePot 0.48→0.52, raiseThreshold 0.97→0.98 (now v31); 3 lessons learned
Claude (playbook → v32): params: tightness 0.7→0.65, aggression 0.4→0.5, bluffFreq 0.01→0, cbetFreq 0.55→0.6, betSizePot 0.85→0.95, raiseThreshold 0.95→0.92 (now v32); 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.