QuestionQ60

VerifiedSecuring and governing agentic workflows

You are building a customer support agent. To meet corporate governance policies, you must ensure that the agent never exposes users’ personally identifiable information (PII) in its responses. You want to enforce these guardrails without changing the agent’s core app code.

How should you configure this governance?

Explanation

Model Armor can use Sensitive Data Protection to detect configured types of sensitive information in agent responses. Applied through Agent Gateway with enforcement enabled, it can block or redact detected violations without adding filtering logic to the agent’s core code. For redaction, configure advanced Sensitive Data Protection settings with a de-identification template and use INSPECT_AND_BLOCK. Detection is not a guarantee that every possible PII instance will be identified. (docs.cloud.google.com)

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!