{
  "id": 131,
  "url": "https://arxiv.org/abs/2607.07229v1",
  "title": "Reasoning Consistency Scanning: A Framework for Auditing Chain-of-Thought Validity in AI Safety Evaluations",
  "summary": "Prior work has shown that chain-of-thought (CoT) reasoning is often unfaithful: a model's stated reasoning does not reliably reflect the process that produced its output. Detecting unfaithfulness, though, requires controlled experimental interventions, which cannot be applied to evaluation transcripts after the fact. We turn instead to a more tractable question that has received less attention: whether the stated reasoning is logically consistent with the answer it accompanies. Unlike faithfulne",
  "authors": "Silvia Santano",
  "category": "research",
  "topics": "safety-alignment,transparency",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-08T10:13:19.000Z",
  "fetched_at": "2026-07-14T14:14:19.968Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/131",
  "original_url": "https://arxiv.org/abs/2607.07229v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}