AI engineering glossary · Evals guardrails AI security
What is defence in depth?
Defence in depth (also called layered security): A security strategy of stacking several independent layers of protection so that if one fails, others still limit the harm. For AI systems this combines input checks, safer prompts, permission limits, output filters, monitoring and human review, since no single measure stops all attacks.
Defence in depth: an example
A bot uses an input classifier, a restricted tool list and an output filter, so a jailbreak that beats the first still hits the others.
Related terms
More Evals guardrails AI security terms
- Denial of wallet
- Direct prompt injection
- Egress filtering
- Elo rating
- Encoding attack
- Eval set
- Evaluation harness
- Evasion attack
Where you learn it
Defence in depth is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.