{
  "schema": "vedokrok.public-item.v1",
  "release_id": "MHC-RPUB-20260920-75ad787a",
  "url": "/knowledge/let-the-model-propose-let-deterministic-code-commit",
  "id": "MHC-D-RESEARCH-1176",
  "version": "0.1.0",
  "title": "Let the model propose; let deterministic code commit",
  "summary": "Use the model for judgment; use the control plane for authority.",
  "kind": "principle",
  "body": "Treat model output as a proposed action. Deterministic infrastructure validates schema, permissions, limits, idempotency and current state before committing the mutation. Record the decision and result so failures can be audited, retried or reconciled without asking the model what probably happened.",
  "limits": [
    "Deterministic guards constrain effects but cannot guarantee the model chose the right business objective; consequential intent still needs appropriate review."
  ],
  "topics": [
    "union-ai-runtime-verification-2026"
  ],
  "intents": [],
  "source_ids": [
    "RS-CC5DC111315BEF2B"
  ],
  "evidence": [
    {
      "claim": "A deterministic control plane can treat model outputs as proposals, validate them, commit permitted actions and keep a decision history for recovery and audit.",
      "source_id": "RS-CC5DC111315BEF2B",
      "role": "supports",
      "note": "Deterministic infrastructure cannot make model judgment deterministic; it constrains effects and records decisions rather than eliminating model uncertainty.",
      "locator": "3:08-6:27, model proposals, control-plane decision history, coordinated state and distributed-systems protections"
    }
  ],
  "use_when": [
    "An agent can trigger consequential mutations directly from model output."
  ],
  "avoid_when": [
    "Deterministic guards constrain effects but cannot guarantee the model chose the right business objective; consequential intent still needs appropriate review."
  ],
  "example": "An agent proposes updating 500 customer records. The control plane validates the allowed fields, batch size and authorization, assigns an operation ID and performs the write.",
  "check": "A malformed or unauthorized proposal fails before mutation, and a later operator can reconstruct why an accepted action ran.",
  "sources": [
    {
      "id": "RS-CC5DC111315BEF2B",
      "title": "Building Deterministic Infrastructure for Non-Deterministic AI Agents",
      "url": "https://ai.engineer/talks/APh1Vx0oLmQ-building-deterministic-infrastructure-non"
    }
  ],
  "relations": [
    {
      "from": "MHC-D-RESEARCH-1176",
      "to": "MHC-D-RESEARCH-0315",
      "type": "useful_with",
      "url": "/knowledge/enforce-permissions-outside-the-model"
    }
  ],
  "collections": [
    {
      "id": "RC-ACD6385FB70207D3",
      "title": "Operate AI agents as systems you can replay, verify and constrain",
      "url": "/collections/operate-ai-agents-as-systems-you-can-replay-verify-and-constrain"
    }
  ]
}