When an AI model generates information that sounds confident and plausible but is factually wrong or completely made up.
The model isn’t lying – it has no concept of truth or falsehood. It predicts the most statistically likely next word based on its training data. When it lacks reliable information, it fills the gap with something that sounds right rather than flagging the uncertainty.
The risk isn’t that hallucinations sound wrong. It’s that they often sound entirely correct.