AI Literacy / Glossary

AI hallucination

An AI hallucination is when an AI tool produces information that sounds confident and plausible but is false or made up, such as a fake fact, a quote nobody said or a source that does not exist.

Language models write by predicting likely next words, not by looking up checked facts. When they lack good information, they can still produce a fluent answer, and that answer may be invented.

Common examples are made-up book titles, wrong dates, non-existent links, invented legal cases and quotes put in a real person's mouth. The confident tone makes them easy to believe.

To protect yourself, ask for sources and open them, cross-check important claims with a trusted site, be extra careful with health, money and legal topics, and never copy an AI answer into schoolwork or a work document without checking it.