You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Nicolas Cravino edited this page Apr 15, 2026
·
2 revisions
id: oscal-phantom-compliance
title: "From article to artefacts Vol 2 - When AI Agents Write Your regulatory Compliance Reports: The Urgent Reality of Phantom Controls"
author: Nicolas Cravino
date: 2026-02-11
source_file: "2026-02-11 04:16:37.0-From article to artefacts Vol 2 - When AI Agents Write Your regulatory Compliance Reports: The Urgent Reality of Phantom Controls.md"
ingested: 2026-04-11
category: oscal-compliance
stacks: [oscal, compliance]
tags: [compliance, langgraph, nist, oscal]
Summary
AI agents writing compliance artifacts risk generating false controls that pass validation yet don't exist. OSCAL systems require guardrails, evidence reconciliation, and confidence scoring.