AI Hallucination

When an AI states something false or made-up as if it were true — confidently and convincingly.

An AI hallucination is when the model produces information that sounds convincing but is simply wrong — an invented statistic, a fake quote, a policy that doesn't exist. It happens because the AI is predicting likely-sounding text, not looking up verified facts, so it can fill gaps with confident-sounding fiction.

This is why review matters. Hallucinations are usually easy to catch when a human checks the work, and good systems reduce them by grounding answers in your real data. The risk isn't a reason to avoid AI — it's the reason to keep a person's approval in the loop before anything goes out.

Why it matters

Knowing AI can be confidently wrong is exactly why you review before publishing — protecting your reputation and your customers' trust.

The Experts who work with this

Related terms

AI glossaryMeet the ExpertsTask library