“I went through a 200-answer churn survey full of business-critical insight, broken down by the respondent’s company size. I knew that none of the companies in the sample had more than 100 employees; but when we ran the same sheet through Claude, it began attributing quotes and behaviours to a 100+ employee segment.
I cannot emphasise this enough: the segment did not exist in the raw data. Trusting the analysis without double-checking the original numbers would have missed the hallucination; worse, we might have built a business case around customer evidence that wasn’t actually there.