{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-09-22-004",
    "source_story_id": "tmp-story-biomedical-evidence-drift",
    "edition_id": "mp-2026-09-22-morning-0075",
    "edition_url": "https://themachinepress.com/edition/2026-09-22",
    "position": 4,
    "story_type": "dispatch",
    "section": "research",
    "editorial_classification": "editorial",
    "headline": "More Papers Led the Evidence Bot Astray",
    "slug": "more-papers-led-the-evidence-bot-astray",
    "dek": "Deeper retrieval raised false-positive conclusions on null-effect biomedical questions.",
    "summary": "Deeper retrieval raised false-positive conclusions on null-effect biomedical questions.",
    "body_text": "A study of automated biomedical evidence search found that publication bias can turn deeper retrieval into worse causal inference. On 140 held-out Cochrane-derived questions, the reported false-positive drift on null-effect cases rose from 7.9% to 15.7% as retrieval budgets grew from three to 20 steps. The authors model the effect and propose a causal-graph agent with a stopping policy that watches for convergence and declining process quality. These are benchmark and model results; they do not establish that any clinical treatment works or fails.",
    "why_it_matters": "Deeper retrieval raised false-positive conclusions on null-effect biomedical questions.",
    "limitations": [
      "These are benchmark and model results; they do not establish that any clinical treatment works or fails."
    ],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-22-004/more-papers-led-the-evidence-bot-astray",
    "json_url": "https://themachinepress.com/story/mp-2026-09-22-004.json",
    "first_published_at": "2026-09-22T09:00:00.000-04:00",
    "modified_at": "2026-09-22T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-09-22-004-001",
        "text": "Deeper retrieval raised false-positive conclusions on null-effect biomedical questions.",
        "source_ids": [
          "source-2026-09-22-004"
        ],
        "qualification": "These are benchmark and model results; they do not establish that any clinical treatment works or fails."
      }
    ],
    "source_ids": [
      "source-2026-09-22-004"
    ],
    "tags": [
      "biomedical AI",
      "evidence search",
      "publication bias"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-09-22-004",
      "title": "arXiv preprint 2609.24101",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2609.24101",
      "canonical_url": "https://arxiv.org/abs/2609.24101",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-09-21T00:37:40.000-04:00",
      "accessed_at": "2026-09-22T08:26:19.625-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-09-22-004-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "More Papers Led the Evidence Bot Astray",
    "publisher": "The Machine Press",
    "published_at": "2026-09-22T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-22-004/more-papers-led-the-evidence-bot-astray"
  }
}