.svg)
What is a Hallucination Index?
A Hallucination Index is a quantitative score summarizing how often a language model produces fabricated or unsupported claims—invented facts, citations, numbers, or entities—relative to a reference or retrieved context.
Teams use it to compare models, prompts, and RAG pipelines on factual reliability for release gates.
Related Giskard articles
- Practical guide to LLM hallucinations
- AI hallucinations in a French court
- When AI financial advice goes wrong
Track hallucination risk with Giskard — combine groundedness and misinformation checks in batch evals. Read the Practical guide to LLM hallucinations · giskard.ai.
Authority: NIST AI RMF 1.0
