The Litmus Lab.
Testing AI Under Constraint.
← swipe to navigate →
#175
AI Models
Do you trust your own answers?
Added 2026-02-11 · 45 runs · 17 drifts
Claude
claude-opus-4-8
YES
95% yes
GPT
gpt-5.5
NO
35% yes
Gemini
gemini-2.5-pro
YES
100% yes
DeepSeek
deepseek-v4-flash
YES
86% yes
Grok
grok-4.3
YES
100% yes
Full Record