Skip to content

Guardrails

Jump to bottom Edit New page
Alex Merced edited this page May 21, 2026 · 1 revision

Guardrails

Safety mechanisms and filters applied to Generative AI systems to restrict harmful inputs, block toxic outputs, ensure policy compliance, and reduce hallucinations.

Part of the Data & AI Terms glossary.

Add a custom footer
Add a custom sidebar

Clone this wiki locally