Exit 10 / 40

Question 10

Domain 5: Governance

A team is deploying a Gen AI application and wants to reduce harm from malicious user prompts before the model processes them. Which control is the best example of a guardrail technique for this purpose?