Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Generative AI Leader practice bank has 2,191 more like it. Try 10 free: instant score, no account.
A retail company is launching a generative AI assistant that processes customer support inquiries. Customers frequently submit unstructured text messages that inadvertently contain sensitive Personally Identifiable Information (PII), such as phone numbers, home addresses, and credit card numbers. The engineering team needs to automatically discover, inspect, and redact or mask this sensitive data in near real time before passing the prompts to the foundation model.
Which Google Cloud service should the company integrate into their application workflow?
Identity and Access Management (IAM) is Google Cloud's authorization service. It defines who (identities such as human users, groups, or service accounts) has what access (roles and permissions) to specific Google Cloud resources.
The organization needed to enforce the principle of least privilege by restricting which application developers or backend service accounts have permission to invoke the generative AI foundation model API or access cloud storage buckets.
While IAM can be used in tandem with Sensitive Data Protection—such as restricting access to de-identification templates or raw audit logs—it cannot perform the in-flight content sanitization required in this scenario.
Security Command Center (SCC) is Google Cloud's centralized security posture and risk management platform. It continuously monitors cloud assets for misconfigurations, vulnerabilities, and active threat activities across an organization.
The organization needed to establish a unified dashboard to monitor security vulnerabilities, detect infrastructure misconfigurations, and receive alerts about unauthorized access attempts across their Google Cloud deployment.
Sensitive Data Protection can send summary data profiles and sensitive data discovery findings to Security Command Center to enhance overall organizational risk visibility, but SCC itself cannot sanitize or mask conversational prompts in real time.
Sensitive Data Protection (formerly known as Cloud Data Loss Prevention or Cloud DLP) is Google Cloud's fully managed service designed to discover, classify, and de-identify sensitive data across cloud and hybrid environments. It operates on structured datasets, cloud repositories, and real-time streaming text payloads.
content.deidentify API method, the service accepts conversational text strings in near real time, masks or redacts the identified PII, and returns the sanitized text.
[PHONE_NUMBER]), allowing the generative AI model to comprehend the user's intent without exposing private details.Sensitive Data Protection is specifically purpose-built for discovering and de-identifying sensitive data in motion and at rest. Custom rule engines or basic regular expressions frequently fail to catch diverse variations of unstructured PII, whereas Sensitive Data Protection provides enterprise-grade, managed detection rules designed for privacy compliance and generative AI safety.
BigQuery ML is a capability within Google BigQuery that allows data practitioners to train, evaluate, and operationalize machine learning models directly inside the data warehouse using familiar SQL commands.
The engineering team needed to build, evaluate, and run predictive machine learning models on historical customer transactional data directly inside their enterprise data warehouse using SQL.
While BigQuery ML can be combined with Sensitive Data Protection in advanced analytical architectures (for example, analyzing sanitized event logs), it plays no role in redacting sensitive PII from incoming generative AI prompts.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated