Hallucination
A hallucination is model output that is presented confidently as fact but is fabricated — an invented citation, statistic, quotation, function name, or event.
Hallucination is most likely where the model is asked for specifics it does not have: named sources, precise figures, recent events, or obscure API details. Prompts reduce it by supplying the facts rather than requesting them, and by explicitly permitting "I do not know" as an answer.
Related terms
- Grounding
- Grounding is constraining a model to answer only from supplied source material, rather than from its training data.
- RAG (retrieval-augmented generation)
- RAG is a technique that retrieves relevant documents from an external store and places them in the prompt, so the model answers from that material rather than from memory.
- Stop condition
- A stop condition is an instruction telling a model when to stop and ask for clarification instead of proceeding on assumptions.
More on failure modes
See the full glossary, read the guides, or put it into practice in the prompt builder.