AI engineering glossary · Evals guardrails AI security
What is ASR (attack success rate)?
Attack success rate (also called ASR): The percentage of adversarial attempts, such as jailbreak or injection prompts, that get a system to do what it should not. It is the main number for comparing defences and models in security testing, and lower is better.
Attack success rate: an example
Of 200 jailbreak prompts tried, 14 produce a prohibited answer, so the attack success rate is 7 percent.
Related terms
More Evals guardrails AI security terms
- Attack surface
- Automated red teaming
- Backdoor attack
- Benchmark saturation
- Canary token
- Chatbot Arena
- Citation checking
- Constitutional classifiers
Where you learn it
Attack success rate is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.