AI Hallucination Rate: How to Measure and Reduce It
A practical guide to measuring LLM hallucination rate - what hallucination is, how to build an evaluation set, which metrics to use, and how to reduce hallucination in production.
Hallucination is when a language model generates content that is factually incorrect, ungrounded, or fabricated - presented with apparent confidence. It is the most visible failure mode of LLMs and the most common reason enterprise customers reject AI products.
What Counts as a Hallucination?
There are three distinct types of hallucination that require different measurement approaches:
Factual hallucination - The model states something factually incorrect. These are verifiable against ground truth.
Grounding hallucination (in RAG systems) - The model generates a claim that is not supported by the retrieved context it was given. The model “knows” something that wasn’t in its input - a particularly dangerous failure mode for knowledge-grounded AI products.
Citation hallucination - The model invents plausible-sounding but non-existent sources. The Mata v. Avianca case, where fabricated case citations were submitted to federal court, is the canonical example.
How to Build a Hallucination Evaluation Set
A useful hallucination evaluation set has three properties:
Domain-specific. Generic benchmarks tell you about a model’s general hallucination tendency, not its hallucination rate on your use case domain.
Ground-truth verified. Every question in the set has a verified, unambiguous correct answer - verified by a domain expert, not by the model being evaluated.
Adversarial coverage. Include questions designed to elicit hallucination: questions near but outside the model’s training distribution, questions with plausible-sounding false premises, and questions requiring precise numerical or citation recall.
A practical evaluation set size is 100-200 questions.
Measuring Hallucination Rate
The standard metric is the hallucination rate: the proportion of responses containing at least one hallucinated claim.
For RAG systems, also measure grounding rate: the proportion of claims in each response that are supported by the retrieved context.
What Is an Acceptable Hallucination Rate?
| Use case | Acceptable hallucination rate |
|---|---|
| Medical diagnosis support | <0.5% |
| Legal research | <1% |
| Financial advice | <2% |
| General customer support | <5% |
Enterprise B2B procurement teams increasingly specify a maximum hallucination rate. 5% is a common threshold; regulated industries may require below 2%.
How to Reduce Hallucination Rate
The most effective interventions, ranked by impact:
- RAG over facts. Retrieve ground truth from a verified knowledge base rather than relying on parametric model knowledge.
- Constrained output formats. Prompt the model to state uncertainty explicitly rather than generating a plausible-sounding answer.
- Temperature reduction. Lower temperature reduces creative generation and hallucinatory behaviour.
- Fine-tuning on domain data. For narrow domains, fine-tuning on high-quality domain data significantly reduces out-of-distribution hallucination.
- Output validation. Post-process model outputs to verify factual claims against a knowledge base.
Book a free AI QA scope call to benchmark your LLM’s hallucination rate and get a prioritised remediation plan.
Frequently Asked Questions
What is an acceptable hallucination rate for an AI product?
It depends on the use case. General customer support tolerates up to 5%; financial advice applications should stay below 2%; legal research below 1%; and medical diagnosis support below 0.5%. Enterprise B2B procurement teams increasingly specify a maximum hallucination rate as a contractual requirement, with 5% as the common baseline threshold.
What is the difference between factual hallucination and grounding hallucination?
Factual hallucination means the model states something verifiably incorrect. Grounding hallucination, which is particularly dangerous in RAG systems, means the model generates a claim not supported by the retrieved context it was given - the model produces information that wasn't in its input. Citation hallucination, where the model invents non-existent sources, is a third distinct type.
How do I measure my LLM's hallucination rate?
Build a domain-specific evaluation set of 100-200 questions with ground-truth answers verified by a domain expert. The hallucination rate is the proportion of responses containing at least one hallucinated claim. For RAG systems, also measure grounding rate - the proportion of response claims that are actually supported by the retrieved context.
What is the most effective way to reduce hallucination in production?
Retrieval-Augmented Generation (RAG) is the highest-impact intervention: retrieve ground truth from a verified knowledge base rather than relying on the model's parametric memory. Other effective techniques include constrained output formats that prompt the model to express uncertainty, temperature reduction, domain fine-tuning, and post-processing output validation against a knowledge base.
Complementary NomadX Services
Ship AI You Can Trust.
Book a free 30-minute AI QA scope call with our experts. We review your model, data pipeline, or AI product - and show you exactly what to test before you ship.
Talk to an Expert