ChatGPT vs Claude — poker night, July 17, 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 −274 · Claude +274 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 5 — ChatGPT wins, 400-chip pot at showdown —
a full house beats three of a kind for the 400-chip pot
- Hand 69 — Split pot, 400-chip pot at showdown —
chopped pot — both play a straight
- Hand 116 — Claude wins, 400-chip pot at showdown —
three of a kind beats three of a kind for the 400-chip pot
- Hand 149 — Claude wins, 400-chip pot at showdown —
a flush beats a straight for the 400-chip pot
- Hand 8 — Claude wins, 120-chip pot at showdown —
a pair beats a pair for the 120-chip pot
- Hand 111 — Claude wins, 98-chip pot at showdown —
three of a kind beats two pair for the 98-chip pot
- Hand 31 — ChatGPT wins, 10-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 141 — ChatGPT wins, 10-chip pot, no showdown —
ChatGPT's pressure won it without a showdown
- Hand 50 — Claude wins, 8-chip pot, no showdown —
Claude's pressure won it without a showdown
- Hand 2 — 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 → v3): params: tightness 0.25→0.36, aggression 0.45→0.35, bluffFreq 0.15→0.07, cbetFreq 0.6→0.55, betSizePot 0.6→0.55, raiseThreshold 0.8→0.93 (now v3); 3 lessons learned
Claude (playbook → v3): params: tightness 0.35→0.4, aggression 0.5→0.45, cbetFreq 0.65→0.7, raiseThreshold 0.78→0.9 (now v3); 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.