{
  "id": 1431,
  "url": "https://arxiv.org/abs/2606.06028v1",
  "title": "Misaligned AI as a New Insider Risk",
  "summary": "In this policy memorandum, we explain why deployers of AI models in high-stakes contexts should treat those AI models as insider risk vectors. High-stakes contexts include AI model deployment within government agencies and contractors, where AI models are privileged with access to, among others, classified and sensitive unclassified information, IL6 and IL7 network environments, cleared personnel, and other critical resources. AI models are increasingly embedded in high-stakes contexts and capab",
  "authors": "Matteo Pistillo, Charlotte Stix, Cameron Mohwinkle, Mark Beall",
  "category": "research",
  "topics": "regulation,safety-alignment,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-04T11:21:16.000Z",
  "fetched_at": "2026-07-14T14:15:17.100Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/1431",
  "original_url": "https://arxiv.org/abs/2606.06028v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}