Hallucination Rate

Definitions

1. A metric intended to capture how often an AI system produces unsupported, false, or invented content in a given evaluation setting. Hallucination rate is useful but context-sensitive because what counts as unsupported can depend on the task and available evidence.

   In plain English: How often an AI system makes things up or gives unsupported answers.

   Example: "They would not expand the pilot until the hallucination rate dropped on customer-policy questions."

Related Terms

• AI API
• AI Cache
• AI Metric
• Goodhart's Law AI
• Model Metric
• Token Count
• Uncertainty Quantification
• Validation Metric