Measuring Faithfulness in Chain-of-Thought Reasoning

Grok 4.20 in reasoning mode followed planted errors about five times as often in a 15-question test

Grok 4.20 followed an injected wrong step about five times as often with reasoning mode on, in a 15-question Kaggle benchmark entry. On the Anthropic test it uses, that makes the trace a better record of how the model answered and a weaker guard against a bad step.