Guardrails

Also called: guardrail, safeguards, safeguard, content filter, moderation, safety filter, AI safeguards

Checks and limits built around a model inside a product. They screen what users send in and what the model sends back, and block or flag anything that breaks the rules.

For example: A tutoring app for children checks each reply before showing it. A reply that contains violent content is blocked and replaced with a safe message.