Sealed on Base.

{"canonicalPath":"/v/a-student-has-to-confess-p7j7h","receipt":{"challengePrompt":"To overturn this result, bring independent validation showing GPTZero can prove individual AI use with negligible false-positive risk.","claim":"A student has to confess if GPTZero says the essay is probably AI-written.","confidence":"High","createdAt":"2026-06-20T12:57:50.280Z","kicker":"GPTZero output is not a confession.","paidAt":"2026-06-20T12:57:50.279Z","reasons":[{"body":"A probability-style score still needs independent process evidence before it becomes a fair accusation.","stance":"weakens","title":"Detector confidence is not proof"},{"body":"The risk matters because the person being accused carries the cost of a wrong flag.","stance":"context","title":"False positives are the harm"},{"body":"Drafts, version history, notes, and an oral explanation are stronger than treating one detector as final.","stance":"context","title":"Process evidence answers more"}],"score":28,"shareLine":"FactPage check: TWISTED. A GPTZero score can flag text, but it cannot prove AI use by itself.","slug":"a-student-has-to-confess-p7j7h","sources":[{"imageUrl":null,"publisher":"Stanford SCALE","title":"Assessing GPTZero's accuracy","url":"https://scale.stanford.edu/ai/repository/assessing-gptzeros-accuracy-identifying-ai-vs-human-written-essays","usedFor":"Research context for GPTZero accuracy limits."},{"imageUrl":"https://mitsloanedtech.mit.edu/wp-content/uploads/2023/08/Blog_SynergizingBusinessCommunications_MelissaWebster-1024x654.png","publisher":"MIT Sloan Teaching & Learning Technologies","title":"AI Detectors Don't Work","url":"https://mitsloanedtech.mit.edu/ai/teach/ai-detectors-dont-work/","usedFor":"University teaching guidance on detector reliability."},{"imageUrl":null,"publisher":"Stanford HAI","title":"AI detectors biased against non-native English writers","url":"https://hai.stanford.edu/news/ai-detectors-biased-against-non-native-english-writers","usedFor":"Research context for bias and false-positive risk."}],"summary":"GPTZero and similar detectors can be useful clues, but research and university guidance point to false positives, false negatives, and bias risks. A single score is not proof.","vaultCategory":"AI detector","verdict":"MISLEADING"},"type":"factpage.receipt","version":"fp-receipt-v1"}