Your agent returns something odd. Was it the prompt, a tool call that timed out, or a response your code could not parse? Without traces, you are guessing.

A customer asks a company’s new AI-based support assistant whether a subscription purchased 15 days ago qualifies for a refund. The assistant responds by explaining that the company offers a 30-day refund window, describes the cancellation process, and promises that the money will arrive within five working days. In reality, the company only allows refunds within 14 days. The assistant has invented a fake commitment out of thin air.

Hallucination in the context of LLMs is generated information that is factually incorrect, invented, or inconsistent with the material the model is supposed to use. The entire response may not be wrong, but a small fabricated detail may be the part the reader relies upon.

Errors fall into three categories: a factual hallucination contradicts reality; a faithfulness hallucination concerns the relationship between the answer and its supplied evidence; fabrication involves inventing something like a policy section or a research paper. These categories overlap.