{
  "id": 266,
  "url": "https://arxiv.org/abs/2607.03510v1",
  "title": "CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI",
  "summary": "Enterprise artificial intelligence is moving from experimentation into operational workflows. Early programs focused on model access and retrieval-augmented generation, but enterprises are now beginning to deploy agents that plan, retrieve, remember, call tools, update systems, and coordinate work across applications. This changes the evaluation problem. Leaders are no longer asking only whether an answer is accurate or fluent. They need to know who authorized an action, which policy applied, wh",
  "authors": "Roopam W. Sure",
  "category": "research",
  "topics": "regulation,agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-03T17:37:10.000Z",
  "fetched_at": "2026-07-14T14:14:24.248Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/266",
  "original_url": "https://arxiv.org/abs/2607.03510v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}