Safety · Concept
Chain-of-Thought Monitoring: Answer Access vs Reasoning Verification
Study showing trusted reference answers inflate CoT monitor accuracy mainly on wrong-answer traces while hurting critical correct-answer-with-bad-trace cases.
Forecast horizon
Oversight protocols should separate answer checking from process verification.
Connections
Connections · 2
How this node ties into the rest of the map, and the evidence behind each link.
to · constrains 3/10
Science of AI Evaluation Answer-access artifacts show process monitors can look better while missing critical wrong-trace correct-answer failures.
+3 growth to · constrains 3/10
Verification-Time Dependency on a Disappearing Evaluator Both show post-hoc oversight assumptions fail when monitors see answers or original evaluators disappear.
+3 growthSignal sources
Signal sources
Dated facts from primary sources in this direction.
US evaluation centre Jun 2025
In June 2025 the US AI Safety Institute was renamed the Center for AI Standards and Innovation (CAISI), pivoting toward security, standards and adversary-model assessment.
NIST → Frontier safeguards May 2025
Anthropic activated its ASL-3 deployment and security standard with Claude Opus 4 on 22 May 2025 — the first real-world trigger of a responsible-scaling tier, focused on blocking bio-weapon uplift.
Anthropic → Cross-border testing 2025
The International Network of AI Safety Institutes (launched Nov 2024) ran a third joint testing exercise focused on agentic AI systems across cyber and fraud strands.
European Commission — AI Office →