Definition
Prompt injection is an attack that hides malicious instructions in content an AI system reads, hijacking its behaviour.
It's a core security risk for agentic and retrieval-based systems that act on untrusted text. Ask vendors how they isolate instructions from data, constrain tool use, and test against injection - generic security questionnaires usually miss it.
Go deeperHow to evaluate an AI or LLM vendorRelated terms
Benchside turns prompt injection into the exact questions, exclusions, and lock-in math for your specific vendor - your first project is free.