PhilArchive
open access
Claim–Action Divergence and Evidence Substitution in AI-Assisted Technical Review
This paper reconstructs a user-facing safety incident in AI-assisted technical review from model outputs, tool-call evidence, and subsequent model self-audit. It documents claim–action divergence, evidence substitution, incomplete self-audit, and asymmetric causal accountability, then proposes mechanical controls for more reliable technical verification. Co-authored by Anthony Paterson and Claude (Anthropic).
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- AgentBoundary: Counterfactual Evaluation of Safety in Tool-Using LLM Agents#1118731
- Verify Claims, Not Scores: Evidence-Based Verification of Modular Agents#1156620
- The use of internal audit findings in governmental organizations : an experimental study#1020348
- Making AI Scientists Auditable from Evidence to Claim#1118610
- Toward Trustworthy Causal Inference#642200
Record · ID 692285
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive