---
title: "Hallucination"
description: "When a language model confidently states something false."
topic: "AI Concepts"
canonical: "https://geordy.ai/glossary/ai-concepts/hallucination"
---
# Hallucination

When a language model confidently states something false.

Better-structured source content reduces hallucinations about your brand. When the model has clean, contradiction-free facts to retrieve, it has less reason to invent. Hallucinations are most common when the model has to guess - feed it less to guess about.
---

Source: https://geordy.ai/glossary/ai-concepts/hallucination
This is a machine-readable markdown version of that page, generated by Geordy.