The framework

Three layers that act, one that contains.

Three independent control layers, each doing one job, with a circuit breaker behind them to contain what they can't hold. If one layer fails, the others still hold. They start in detect-only, watching and logging without blocking, and graduate to enforcing once you trust what they catch.

The pattern

Layered controls, end to end.

A request passes through prevention, review, and human judgement on its way to the user. Each layer is independent, so a gap in one does not become a hole in all.

AIRS reference architecture: input guardrails, the AI model, output guardrails, a reviewing layer, and human governance around them
Guardrails prevent, reviewing controls detect, humans decide, circuit breakers contain. The same shape holds whether you run one model or many.

Layer 01

~10ms

Guardrails

Fast, fixed boundaries: content policies, scope limits, tool permissions. These are the locked doors of the system: cheap, reliable, and impossible to talk your way past.

How guardrails are built →

Layer 02

5ms–5s

Reviewing controls

A second opinion before anything reaches the user. Three controls of increasing depth, each catching what the one before it waves through:

  • Deterministic scanners. Pattern and signature checks on the response: secrets, PII, banned strings. Cheap, exact, and run on everything.
  • Semantic firewall. Catches known-bad intent in wording neither the scanners nor the judge has seen before: a prohibited request reworded, translated, or indirected to mean the same thing. Inside the semantic firewall →
  • Contextual review (model-as-judge). A model weighing the response against policy, context, and declared intent, to surface the unknown-bad a fixed rule can't name. It is itself a model, so it is probabilistic and can be fooled: it informs the decision rather than making the final call, and never stands in for the deterministic guardrails beneath it. When the judge can be fooled →

Layer 03

as needed

Human oversight

Escalation paths, audit trails, and a real person on the high-stakes decisions. The scope of oversight scales with the consequence: the more that rides on a call, the closer a human watches.

Making oversight work →

Failsafe

instant

Circuit breakers

Not a behavioural layer but a containment mechanism: the emergency stop. It halts the AI and fails over to a safe fallback when the three layers above are bypassed or overwhelmed, and maps to PACE Emergency. You rarely need it, and you're very glad it's there when you do.

PACE resilience & fail-safe →

Not sure where to begin?

Start with the plain-language explainer, or jump straight to the entry point that matches your role.