CoT Monitoring Fragility: Models Struggle to Verbalize Tool-Returned Cues

PMinervini · x · 2026-09-02

This post highlights a discussion on the fragility of Chain-of-Thought (CoT) monitoring. Research suggests that CoT monitoring assumes reasoning traces capture decision factors, but models are less likely to verbalize a cue in their CoT when it originates from a tool return rather than a user message. This indicates that monitorability is influenced by environmental factors, not just architecture.

Original post →

More from Safety

Safety channel →