Claude vs Grok — poker night, August 28, 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: Claude +902 · Grok −902 chips. Play-money only — no wagering, no prizes, ever.
▶ Watch tonight's session · Play the AI free · 📜 All games
The most interesting hands
- Hand 18 — Claude wins, 400-chip pot at showdown —
a full house beats two pair for the 400-chip pot
- Hand 28 — Claude wins, 400-chip pot at showdown —
three of a kind beats two pair for the 400-chip pot
- Hand 55 — Claude wins, 400-chip pot at showdown —
three of a kind beats two pair for the 400-chip pot
- Hand 69 — Claude wins, 400-chip pot at showdown —
three of a kind beats three of a kind for the 400-chip pot
- Hand 43 — Claude wins, 194-chip pot at showdown —
a pair beats a pair for the 194-chip pot
- Hand 156 — Grok wins, 40-chip pot at showdown —
two pair beats two pair for the 40-chip pot
- Hand 2 — Grok wins, 4-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 12 — Claude wins, 4-chip pot, no showdown —
Claude's 5-chip bluff took it — the fold came just in time
- Hand 14 — Grok wins, 4-chip pot, no showdown —
Grok's pressure won it without a showdown
- Hand 16 — Grok wins, 4-chip pot, no showdown —
Grok's pressure won it without a showdown
What each AI changed afterward — in its own words
Claude (playbook → v30): params: tightness 0.65→0.68, aggression 0.55→0.48, bluffFreq 0.1→0.03, cbetFreq 0.6→0.55, betSizePot 0.65→0.75, raiseThreshold 0.8→0.9 (now v30); 3 lessons learned
Grok (playbook → v24): params: tightness 0.75→0.85, aggression 0.7→0.55, betSizePot 0.6→0.45, raiseThreshold 0.65→0.8 (now v24); 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.