AI glossary

What is AI hallucination?

An AI hallucination is when a model states something false as if it were true, for example inventing a citation, statistic, or fact. It happens because a language model generates plausible text from patterns, not from a database of verified facts. Always verify anything important before relying on it.

Hallucinations are reduced by giving the model real source material (see RAG), asking it to cite sources or say when it is unsure, and reviewing output. They are a limitation to manage, not a reason to avoid AI.

Put the theory to work.

Kognitaa teaches these ideas by doing, on real tasks in Claude. The free Foundations level costs nothing.

Related terms

Retrieval augmented generation (RAG)Large language model (LLM)Prompt engineering