Full Explanation
A hallucination is what happens when an AI model produces plausible-sounding text that isn't actually true. The word suggests something unusual — a drift, a malfunction. But the model isn't doing anything unusual. It generates the most plausible continuation of its input, exactly as it was trained to. When that output happens to match reality, we call it correct. When it doesn't, we call it a hallucination. The model was doing the same thing both times.
There is no built-in step in the core generation process where the model checks its output against reality. Modern systems can reduce hallucinations by adding external checks — retrieving documents, verifying against sources, applying constraints — but these work around the model, not inside it. The underlying behavior doesn't change: the model produces fluent text with no mechanism for knowing whether that text is true. That gap between fluency and truth is the thing to understand.
---
