AI engineering glossary · Evals guardrails AI security
What is over-refusal?
Over-refusal (also called false refusal, exaggerated safety): When a model declines or waters down a request that is actually harmless because it superficially resembles a dangerous one. It frustrates users and reduces usefulness, so safety evaluation tracks it alongside harmful compliance to keep the balance.
Over-refusal: an example
A model refuses to explain how to 'kill a Linux process' because of the word kill.
Related terms
More Evals guardrails AI security terms
- Overreliance
- OWASP Top 10 for LLM Applications
- Pairwise comparison
- Payload splitting
- Perplexity filter
- Pickle deserialization risk
- PII detection
- PII redaction
Where you learn it
Over-refusal is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.