Glossary
Hallucination
When an AI model states something false or made up as if it were true.
A hallucination is output from an AI model that is fluent and confident but false, such as an invented fact, a citation to a paper that does not exist, or a wrong figure. It happens because language models generate likely-sounding text rather than looking facts up.
Hallucinations matter most where errors are costly, such as legal filings, medical advice, and news. Courts have sanctioned lawyers for filing briefs with citations invented by AI tools.
Developers report falling hallucination rates on their own tests, but the rate depends heavily on the task and the test. Retrieval and citation features can reduce errors without removing them, so outputs still need checking.