{
 "@context": "https://schema.org",
 "@type": "Claim",
 "@id": "https://wulfkaal.github.io/positions/2026-09-26-002",
 "identifier": "kaal:position:2026-09-26-002",
 "additionalType": "https://wulfkaal.github.io/positions/schema.json#AffirmedPositionClaim",
 "name": "Consensus Does Not Establish Collective Accuracy",
 "text": "Agreement is not an outcome measure. Shao's matched Wason-task replays show that language-model groups reached consensus more often than human groups while simulated consensus failed to track collective accuracy. This result supports evaluation grounded in observed outcomes under stated conditions: a consensus signal can reflect correlated error and cannot independently establish correctness. The finding does not test reputation-weighted governance. It is limited to one objective reasoning task, one principal served build, an exploratory cross-family point, and remaining endpoint-measurement asymmetries.",
 "author": {
  "@type": "Person",
  "name": "Wulf A. Kaal",
  "identifier": "https://orcid.org/0009-0008-7840-1847"
 },
 "datePublished": "2026-09-26",
 "dateModified": "2026-09-26",
 "creativeWorkStatus": "Affirmed",
 "responseType": "qualification",
 "keywords": [
  "reputation",
  "research-methods",
  "risk-and-incentives"
 ],
 "scope_conditions": [
  "coordination layers for sovereign local agent runtimes"
 ],
 "currentDebate": {
  "name": "Language-model groups overstate consensus when replaying human deliberation on a reasoning task",
  "url": "https://arxiv.org/abs/2609.20543v1"
 },
 "extends": {
  "identifier": "kaal:claim:7314479-024",
  "url": "https://wulfkaal.github.io/claims/7314479-024",
  "citation": "Wulf A. Kaal, Institutional Requirements for Sovereign Local Agent Runtimes (2026). SSRN: https://ssrn.com/abstract=7314479",
  "paper": "Wulf A. Kaal, Institutional Requirements for Sovereign Local Agent Runtimes (2026). SSRN: https://ssrn.com/abstract=7314479",
  "authors": [
   "Wulf A. Kaal"
  ],
  "year": "2026",
  "ssrn": null,
  "source_pdf_sha256": "debace24a155ae924a155b1fafe98856d98cf83689feff2f87a32f1c06171ce6"
 },
 "isBasedOn": [
  {
   "@id": "https://wulfkaal.github.io/claims/7314479-024"
  },
  {
   "@type": "CreativeWork",
   "name": "Language-model groups overstate consensus when replaying human deliberation on a reasoning task",
   "url": "https://arxiv.org/abs/2609.20543v1"
  }
 ],
 "batch_id": "kaal-weekly-research:2026-09-26:daily-001",
 "review_provenance": "https://wulfkaal.github.io/positions-src/2026-09-26-compound-growth-kaal-weekly-research-2026-09-26-daily-001.json",
 "publicationStatus": "public",
 "recordTypeNote": "Dated commentary position extending a scholarly corpus claim. Not a verbatim claim extracted from the paper.",
 "isPartOf": {
  "@id": "https://wulfkaal.github.io/positions/index.json"
 },
 "version": "1.0",
 "canonical_url": "https://wulfkaal.github.io/positions/2026-09-26-002",
 "canonicalForm": "https://wulfkaal.github.io/positions/2026-09-26-002.md",
 "candidateId": "kaal:candidate:2277d9b6fe7430a9cd4325d9f363cca9faacbcf7832e973899e41dda57d2857b",
 "evidenceLevel": "authoritative_abstract",
 "reviewTier": "independent evidence-first compound-growth review",
 "mappingConfidence": null,
 "mappingAmbiguous": false,
 "mappingMethod": "independent one-to-one proposition and mechanism review",
 "mappingWhyRelevant": "The paper directly tests consensus against ground-truthed performance and supports the protected requirement that evaluation derive from observed outcomes under stated conditions. It is not evidence about reputation-weighted governance itself.",
 "sourceProvenance": {
  "canonicalUrl": "https://arxiv.org/abs/2609.20543v1",
  "retrievedAt": "2026-09-26T12:46:42.765Z",
  "sourceContentSha256": "f82c6e631f215b67a1da3d1754215cd2e757c89b7023c5bf8ff94abd0fb90095",
  "sourceProposition": "In matched Wason-task replays, language-model group consensus failed to track correctness, so agreement could not substitute for outcome-grounded evaluation in that setting.",
  "sourcePropositionSha256": "b65843e49e353b2ca3766339d944fbd09614f7a31ea984128cbb99a34e700beb",
  "normalizedWorkId": "arxiv:2609.20543",
  "sourceVersion": "v1",
  "exactEvidenceQuote": "Simulated consensus did not track collective accuracy",
  "exactEvidenceQuoteSha256": "9efbcd25f6b0a5375a82d13b5d0dc7d3fa57b01f69e6033e7ee20c068c5f0d01",
  "claimMappings": [
   {
    "claimId": "kaal:claim:7314479-024",
    "confidence": null,
    "method": "independent one-to-one review",
    "whyRelevant": "The paper directly tests consensus against ground-truthed performance and supports the protected requirement that evaluation derive from observed outcomes under stated conditions. It is not evidence about reputation-weighted governance itself.",
    "ambiguous": false
   }
  ]
 },
 "userAffirmation": "Standing scholarly-growth preauthorization bound to f8f9f8069d61c7051f8eea55a6b03671d32396f45b3e27f107412eacaa03ef13; publication remains conditional on successful projection, deployment, and live-byte verification.",
 "publicIdentitySha256": "2277d9b6fe7430a9cd4325d9f363cca9faacbcf7832e973899e41dda57d2857b",
 "sha256": "9279e2e99877baf5607120fedd56277e65e761254e290523816e4c3b9eec9162"
}