What Are AI Hallucinations? Why LLMs Make Things Up

Hallucinations

When a model produces confident but false information — a made-up quote, a fake book title, a wrong date — it is called a hallucination.

Hallucinations happen because the model generates plausible text, not verified facts. They are more likely for niche topics, exact numbers, citations, and recent events.

Reduce the risk by giving the model trusted source text, asking it to say "I don't know" when unsure, and checking important facts yourself.