Safety Guardrails
Safety guardrails are controls built into or around an AI system to keep its behaviour within acceptable limits, such as refusing harmful requests or filtering unsafe outputs.
What is Safety Guardrails?
Guardrails include safety training, system prompts, input and output classifiers, policy rules and human review.
How it works
Requests and responses pass through checks that block, modify or flag content that breaks policy.
← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions