Why Does AI Make Things Up? Hallucinations Explained With Examples You Can Check

Why does AI make things up even when an answer sounds confident? This guide explains how a language model can generate a plausible statement without establishing that it is true, and why errors in facts, summaries and inferences need different checks. We examine research on hallucinations, retrieval-augmented generation, long-context reliability and the limits of generated explanations. A fictional museum passage supplies a small, reproducible exercise: which questions can the source answer, and which require the system to acknowledge missing information? No model results or general accuracy rates are invented. Learn how to verify a citation, distinguish a budget from actual expenditure and record a real test without overclaiming. Read the companion article and sources on TaylorTailored.co.uk, and follow Taylor Tailored for practical AI literacy.