Defense-in-Depth

A security strategy that layers multiple overlapping controls, so that if one defense fails, others still limit the impact of an attack.

Defense-in-depth applies traditional security thinking to AI systems by combining input isolation, classifier-based detection, deterministic blocking, least-privilege tooling, and human-in-the-loop workflows. No single layer provides complete protection, but together they reduce both the probability and blast radius of successful attacks. Major AI labs including Anthropic, Microsoft, and OpenAI recommend this approach for production agent deployments.

Also known as