AI Securityintermediate12 min · 9 steps
Input Guardrails
Fast, parallel checks on every incoming message that decide allow, redact, redirect or block before the main model runs.
requesterrorsignal
ProblemStep 1 of 9
The main model sees everything users send
Without a gate, every message goes straight to the expensive main model: personal data, injection attempts, off-topic requests, abuse. The model then has to handle all of it correctly, every time.
Input guardrails are fast checks on every incoming message before the main model: PII, injection, scope and abuse. They decide allow, redact, redirect or block.
What changed
- +Messages hit the model unchecked
Lesson 7 of 22 in AI Security