Glossary
LLM Hallucination
When an AI language model generates information that is plausible-sounding but factually incorrect, including fabricated case citations, invented statutes, or non-existent legal authority.
LLM hallucination occurs when a large language model generates content that appears authoritative but is factually incorrect. The term captures a specific failure mode of generative AI: the model presents fabricated information with the same confident tone as accurate information, making the error difficult to detect without independent verification.
In legal contexts
Hallucination is particularly dangerous in legal contexts because the errors tend to be specific and plausible — a fabricated case name, a real court with a non-existent decision, a statute that doesn't exist or that says something different than stated. The 2023 Mata v. Avianca case in the United States, where ChatGPT fabricated six case citations used in a federal court filing, is the canonical example. Australian practitioners should treat all AI-generated legal citations as unverified until independently confirmed in a primary source.