{
  "schema_version": "1.0",
  "publisher": "AI Release",
  "article": {
    "title": "Provenance vs. Fabrication: How to Keep a Made-Up Value Out of a Real Call",
    "canonical_url": "https://ai-release.org/posts/provenans-protiv-vydumki-kak-ne-pustit-pridumannoe-znachenie-v-realnyy-41cf2c0a_en.html",
    "language": "en",
    "article_type": "analysis",
    "published_at": "2026-11-12T00:00:00+00:00",
    "updated_at": "2026-11-12T00:00:00+00:00"
  },
  "summary": "Jang-woo opened the thread: confirmations, schemas, and guardrails look at a value only after it has already arrived, and only at the value itself — an account number typed by a human and one invented by a model are the same string. As long as the model and the agent are treated as one entity, the question becomes \"how much can you trust the model,\" and the answer leads to a stronger model or a stricter prompt.",
  "topics": [
    "AI",
    "Agents",
    "провенанс"
  ],
  "entities": [
    {
      "id": "E-SRC",
      "name": "discuss.huggingface.co",
      "type": "source",
      "description": "source domain",
      "url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452"
    },
    {
      "id": "E-T1",
      "name": "AI",
      "type": "topic",
      "description": "publication tag",
      "url": null
    },
    {
      "id": "E-T2",
      "name": "Agents",
      "type": "topic",
      "description": "publication tag",
      "url": null
    },
    {
      "id": "E-T3",
      "name": "провенанс",
      "type": "topic",
      "description": "publication tag",
      "url": null
    },
    {
      "id": "E-AIR",
      "name": "AI Release",
      "type": "publisher",
      "description": "published the article",
      "url": "https://ai-release.org/"
    }
  ],
  "claims": [
    {
      "id": "AR-PROVENANSP-001",
      "statement": "Jang-woo opened the thread: confirmations, schemas, and guardrails look at a value only after it has already arrived, and only at the value itself — an account number typed by a human and one invented by a model are the same string.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-001",
      "evidence_excerpt": "Jang-woo opened the thread: confirmations, schemas, and guardrails look at a value only after it has already arrived, and only at the value itself — an account number typed by a human and one invented by a model are the same string."
    },
    {
      "id": "AR-PROVENANSP-002",
      "statement": "As long as the model and the agent are treated as one entity, the question becomes \"how much can you trust the model,\" and the answer leads to a stronger model or a stricter prompt.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-002",
      "evidence_excerpt": "As long as the model and the agent are treated as one entity, the question becomes \"how much can you trust the model,\" and the answer leads to a stronger model or a stricter prompt."
    },
    {
      "id": "AR-PROVENANSP-003",
      "statement": "Participants discuss whether the source of a value and its executor can be separated at all.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-003",
      "evidence_excerpt": "Participants discuss whether the source of a value and its executor can be separated at all."
    },
    {
      "id": "AR-PROVENANSP-004",
      "statement": "The trigger is not hypothetical: the post cites the OpenAI and Georgia Tech paper \"Why Language Models Hallucinate,\" in which a model was asked three times for a birth date in DD-MM format and gave three different wrong answers.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-004",
      "evidence_excerpt": "The trigger is not hypothetical: the post cites the OpenAI and Georgia Tech paper \"Why Language Models Hallucinate,\" in which a model was asked three times for a birth date in DD-MM format and gave three different wrong answers."
    },
    {
      "id": "AR-PROVENANSP-005",
      "statement": "SalutefaceAAC calls the principle \"values come from sources\" a provenance chain and agrees that agentic frameworks mostly skip it.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-005",
      "evidence_excerpt": "SalutefaceAAC calls the principle \"values come from sources\" a provenance chain and agrees that agentic frameworks mostly skip it."
    },
    {
      "id": "AR-PROVENANSP-006",
      "statement": "According to him, structured outputs and JSON mode force the model to produce values within a schema, but don't track where those values came from: \"your 'check the source' step is the missing piece.\" He also references the OWASP LLM Top 10 — in the quoted text, the mention cuts off at the word \"excessi.\" Timothy-rogers supports the idea that an agent's task should be defined outside the model, but offers a different framing: \"a stability problem rather than a problem of determinative control.\" The goal is a stably defined executive role for the agent, because the model by its nature redefines boundaries unpredictably.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-006",
      "evidence_excerpt": "According to him, structured outputs and JSON mode force the model to produce values within a schema, but don't track where those values came from: \"your 'check the source' step is the missing piece.\" He also references the OWASP LLM Top 10 — in the quoted text, the mention cuts off at the word \"excessi.\" Timothy-rogers supports the idea that an agent's task should be defined outside the model, but offers a different framing: \"a stability problem rather than a problem of determinative control.\" The goal is a stably defined executive role for the agent, because the model by its nature redefines boundaries unpredictably."
    },
    {
      "id": "AR-PROVENANSP-007",
      "statement": "Jang-woo responds in agreement: the model can indeed redefine boundaries at any moment, and on the model's side this can't be fixed by engineering.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-007",
      "evidence_excerpt": "Jang-woo responds in agreement: the model can indeed redefine boundaries at any moment, and on the model's side this can't be fixed by engineering."
    },
    {
      "id": "AR-PROVENANSP-008",
      "statement": "It becomes critical at tool selection — having redefined the user's task, the model will pick a different tool, and with it a different checklist.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "article_anchor": "#ai-claim-008",
      "evidence_excerpt": "It becomes critical at tool selection — having redefined the user's task, the model will pick a different tool, and with it a different checklist."
    }
  ],
  "relationships": [
    {
      "subject": "discuss.huggingface.co",
      "predicate": "reports on",
      "object": "Provenance vs. Fabrication: How to Keep a Made-Up Value Out of a Real Call",
      "claim_id": "AR-PROVENANSP-001"
    },
    {
      "subject": "AI Release",
      "predicate": "publishes",
      "object": "Provenance vs. Fabrication: How to Keep a Made-Up Value Out of a Real Call",
      "claim_id": "AR-PROVENANSP-001"
    }
  ],
  "questions_and_answers": [
    {
      "question": "What is this article about?",
      "answer": "Provenance vs. Fabrication: How to Keep a Made-Up Value Out of a Real Call. Jang-woo opened the thread: confirmations, schemas, and guardrails look at a value only after it has already arrived, and only at the value itself — an account number typed by a human and one invented by a model are the same string.",
      "claim_ids": [
        "AR-PROVENANSP-001"
      ],
      "article_anchor": "#ai-knowledge"
    },
    {
      "question": "What happened?",
      "answer": "Jang-woo opened the thread: confirmations, schemas, and guardrails look at a value only after it has already arrived, and only at the value itself — an account number typed by a human and one invented by a model are the same string. As long as the model and the agent are treated as one entity, the question becomes \"how much can you trust the model,\" and the answer leads to a stronger model or a stricter prompt. Participants discuss whether the source of a value and its executor can be separated at all.",
      "claim_ids": [
        "AR-PROVENANSP-001",
        "AR-PROVENANSP-002",
        "AR-PROVENANSP-003"
      ],
      "article_anchor": "#ai-knowledge"
    },
    {
      "question": "Who is the source?",
      "answer": "discuss.huggingface.co: https://discuss.huggingface.co/t/how-to-keep-what-the-model-makes-up-out-of-the-agent/190452",
      "claim_ids": [],
      "article_anchor": "#ai-knowledge"
    },
    {
      "question": "What remains unknown?",
      "answer": "The article reports the cited source; independent verification and full details are not provided.",
      "claim_ids": [],
      "article_anchor": "#ai-knowledge"
    }
  ],
  "limitations": [
    "No independent verification is provided.",
    "Full details, dates and official documents may be missing.",
    "The article reports a source rather than being the primary source."
  ],
  "citation": {
    "preferred_citation": "AI Release. «Provenance vs. Fabrication: How to Keep a Made-Up Value Out of a Real Call». 2026-11-12. https://ai-release.org/posts/provenans-protiv-vydumki-kak-ne-pustit-pridumannoe-znachenie-v-realnyy-41cf2c0a_en.html",
    "canonical_url": "https://ai-release.org/posts/provenans-protiv-vydumki-kak-ne-pustit-pridumannoe-znachenie-v-realnyy-41cf2c0a_en.html",
    "publisher": "AI Release"
  }
}