CCAR-P Governance, Safety, and Risk Management Practice Question
An enterprise wants to deploy an AI-powered customer service agent. What is the most important governance consideration when integrating with internal customer databases?
⚠ Common exam trap
Test-takers frequently select broad administrative permissions or full database access for the AI agent, failing to apply the principle of least privilege required for database integrations.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Use a service account with scoped, read-only permissions to the necessary database views.
The most important consideration is implementing a robust access control layer between the AI agent and the database. The agent should only have read-only access to a strictly defined, non-sensitive subset of data. This prevents the model from inadvertently surfacing sensitive info or executing unauthorized data modifications, ensuring that the integration adheres to the principle of least privilege and maintains corporate data integrity standards.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Ensure the model has write access to the database to update customer profiles.
Why it's wrong here
Granting write access to an AI agent significantly increases the risk of data corruption, unauthorized changes, or accidental deletion. Following the principle of least privilege, the agent should only possess read-only permissions for the specific information required to provide accurate customer support responses.
- ✓
Use a service account with scoped, read-only permissions to the necessary database views.
Why this is correct
Scoped, read-only permissions minimize risk by ensuring the AI agent can only access exactly what it needs to perform its function. This prevents unauthorized data exposure and ensures that any potential model hallucination or injection cannot result in unintended modifications to the underlying customer databases.
- ✗
Store the database connection string directly in the prompt for ease of access.
Why it's wrong here
Storing connection strings in prompts is a catastrophic security vulnerability that leads to credential exposure. Connection details must be handled via secure secret managers, never in plain text within system prompts or code, as this would grant anyone with prompt access the ability to compromise the database.
- ✗
Configure the agent to query the entire production database to ensure comprehensive answers.
Why it's wrong here
Querying the entire production database violates the principle of data minimization. The agent should only access specific, pre-approved views. Exposing the entire database increases the risk of PII leakage, potential performance degradation, and data over-exposure, which is unacceptable in an enterprise production environment.
About these practice questions
Courseiva writes every CCAR-P question from scratch — 262 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAR-P practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAR-P exam.