Crucial Exams

GCP Professional Cloud Architect Practice Question

Your company is launching a Vertex AI-based chat assistant that will be exposed publicly over HTTP. Security testing shows that attackers can coerce the underlying foundation model into revealing sensitive training data and producing extremist content. Compliance also requires that personal identifiers in both prompts and responses be redacted before they are stored for analytics. You need to design a control plane solution that mitigates these risks while adding the least possible runtime latency to online predictions. What should you do?

  • Store all prompts and responses unmodified in BigQuery and run Cloud DLP jobs hourly to detect and redact sensitive data retroactively.

  • Enable Model Armor on the endpoint with both prompt and response policies in enforcement mode, and attach a Sensitive Data Protection rule set for PII redaction.

  • Place Cloud Armor in front of the endpoint with custom WAF rules and enable Sensitive Data Protection inspection on Cloud Armor.

  • Enable Model Armor only in monitoring mode and perform daily review of violation logs to update allow-lists.

GCP Professional Cloud Architect
Designing for security and compliance
Your Score:
Settings & Objectives
Random Mixed
Questions are selected randomly from all chosen topics, with a preference for those you haven’t seen before. You may see several questions from the same objective or domain in a row.
Rotate by Objective
Questions cycle through each objective or domain in turn, helping you avoid long streaks of questions from the same area. You may see some repeat questions, but the distribution will be more balanced across topics.

Check or uncheck an objective to set which questions you will receive.

Wipe Score
Bash, the Crucial Exams Chat Bot
AI Bot