{
  "schema": "vedokrok.public-item.v1",
  "release_id": "VEDOKROK-SEL-5d86e1b23401e578",
  "url": "/knowledge/use-ai-to-widen-a-premortem-then-verify-the-risks-yourself",
  "id": "MHC-D-BRALI-0070",
  "version": "0.1.0",
  "title": "Use AI to widen a premortem, then verify the risks yourself",
  "summary": "An AI can produce failure modes quickly. That is useful for coverage, not truth. Ask for plausible ways the plan could fail, then discard generic or impossible suggestions and verify the few that could actually change the plan.",
  "kind": "checklist",
  "body": "An AI can produce failure modes quickly. That is useful for coverage, not truth. Ask for plausible ways the plan could fail, then discard generic or impossible suggestions and verify the few that could actually change the plan.",
  "limits": [
    "Do not send secrets or regulated data to an unsuitable model. Generated risks are hypotheses, not incidents, probabilities or expert review."
  ],
  "topics": [
    "critical-thinking"
  ],
  "intents": [],
  "source_ids": [
    "FP-BRALI-SRC-8892f2fd9b6700a0addd"
  ],
  "evidence": [
    {
      "claim": "First-party source of this practical exercise.",
      "source_id": "FP-BRALI-SRC-8892f2fd9b6700a0addd",
      "role": "contextualizes",
      "note": "Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding.",
      "locator": "data/life-os-content/ai-premortem-risk-planner.json"
    }
  ],
  "use_when": [
    "You are reviewing a plan and want a broader list of possible failure modes without delegating the risk decision."
  ],
  "avoid_when": [
    "Do not send secrets or regulated data to an unsuitable model. Generated risks are hypotheses, not incidents, probabilities or expert review."
  ],
  "checklist": [
    "Describe the plan and constraints without sensitive information.",
    "Ask for concrete ways the plan could fail, including assumptions and dependencies.",
    "Group duplicates and remove risks that contradict known facts.",
    "Verify the important survivors against real data, owners or documentation.",
    "Choose mitigations only for risks that survive verification."
  ],
  "check": "The plan has a shorter verified risk list with owners, checks or mitigations where justified.",
  "template": "Plan: ___.\nImportant constraints: ___.\nAssume it failed. List plausible failure modes, the assumption behind each, and what evidence would distinguish a real risk from a generic possibility.",
  "published_at": "2026-11-12T18:37:40Z",
  "review": {
    "kind": "agent",
    "label": "Agent editorial and scoped source review",
    "reviewed_at": "2026-11-12T18:37:40Z"
  },
  "knowledge_origin": {
    "release_id": "MHC-RPUB-20261112-20287590",
    "corpus_commit": "20287590ddd3999b7492df232822b6a0f7d114b1",
    "manifest_sha256": "2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95",
    "projection_sha256": "2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0",
    "published_at": "2026-11-12T18:37:40Z"
  },
  "editorial_overlay": null,
  "editorial": null,
  "notices": [],
  "sources": [
    {
      "id": "FP-BRALI-SRC-8892f2fd9b6700a0addd",
      "title": "AI-Assisted Premortem",
      "url": "https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-premortem-risk-planner.json"
    }
  ],
  "relations": [],
  "collections": [
    {
      "id": "RC-MHC-PRESSURE-TEST-A-PLAN-OR-EXPLANATION",
      "title": "Pressure-test the plan before the story hardens",
      "url": "/collections/pressure-test-the-plan-before-the-story-hardens"
    }
  ],
  "topic_hubs": [
    {
      "id": "critical-thinking",
      "title": "Critical thinking and evidence",
      "url": "/topics/critical-thinking"
    }
  ],
  "solution_packs": []
}
