A financial services firm is deploying an AI agent that gives investment advice to external clients. The agent must avoid giving inappropriate advice and must be able to explain its decisions. What should the firm implement as part of its responsible AI strategy?

A disclaimer that the advice is not professional financial advice.

A human review process for all advice before it is sent to clients.

A set of predefined rules that filter out high-risk topics and a logging system that records the agent's reasoning.

A feedback loop where clients can rate the agent's responses.

A company is deploying an AI agent that handles employee onboarding. The agent will be deployed in a development environment, then to a user acceptance testing (UAT) environment, and finally to production. The company has strict change management policies that require approval at each stage. What deployment strategy should you plan?

Use automated pipelines with manual approval gates between environments, and maintain separate configurations for each environment.

Use the same configuration for all environments to simplify management.

Deploy directly from development to production to speed up delivery.

Manually deploy to each environment without automation.

A university is designing an AI agent to help prospective students with admissions questions. The agent will be publicly accessible. What design consideration is most important for this external audience?

Require users to create an account before using the agent.

Provide clear, simple language and a fallback to human support.

Limit the agent's availability to business hours.

Ensure the agent uses the university's internal terminology.

A retail company wants to deploy an AI agent to assist customers with product recommendations and order tracking. They want the agent to be available on their website and also through a mobile app. The company has a strict responsible AI policy that requires transparency about AI usage. Which channel plan should you implement?

Embed the agent in the website using a web chat control and integrate the same agent into the mobile app using the mobile SDK.

Deploy the agent only on the mobile app to reduce complexity.

Create two separate agents, one for the website and one for the mobile app, to tailor responses per channel.

Use email as the primary channel and also provide a web chat.

Which security measure is essential to protect data transmitted between an AI agent and its users?

Encryption in transit using TLS

Disabling authentication for ease of access

Storing passwords in plain text

Microsoft Certified:AI Agent Builder Associate

Domain 1Objective 1

Plan an Agent Solution AB-620 Practice Questions (Page 5)

Part of the Plan and configure agent solutions domain, which accounts for 30–35% of the AB-620 exam. Microsoft does not publish an official question count, but from its 120-minute exam (~50–80 total, ~15–28 in this domain), expect 5–9 from this objective — we provide 50 practice questions to prepare you well beyond it. (estimate)

50questions here

10free pages

9concepts

30–35%of the exam

Questions 21–25

21application · medium
A financial services firm is deploying an AI agent that gives investment advice to external clients. The agent must avoid giving inappropriate advice and must be able to explain its decisions. What should the firm implement as part of its responsible AI strategy?
Select an answer first
22expert · hard
A company is deploying an AI agent that handles employee onboarding. The agent will be deployed in a development environment, then to a user acceptance testing (UAT) environment, and finally to production. The company has strict change management policies that require approval at each stage. What deployment strategy should you plan?
Select an answer first
23application · medium
A university is designing an AI agent to help prospective students with admissions questions. The agent will be publicly accessible. What design consideration is most important for this external audience?
Select an answer first
24application · medium
A retail company wants to deploy an AI agent to assist customers with product recommendations and order tracking. They want the agent to be available on their website and also through a mobile app. The company has a strict responsible AI policy that requires transparency about AI usage. Which channel plan should you implement?
Select an answer first
25foundation · easy
Which security measure is essential to protect data transmitted between an AI agent and its users?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to AB-620

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “AB-620” is a trademark of its owner, used for identification only.

Plan an Agent Solution AB-620 Practice Questions (Page 5)

Questions 21–25

A financial services firm is deploying an AI agent that gives investment advice to external clients. The agent must avoid giving inappropriate advice and must be able to explain its decisions. What should the firm implement as part of its responsible AI strategy?

A university is designing an AI agent to help prospective students with admissions questions. The agent will be publicly accessible. What design consideration is most important for this external audience?

Which security measure is essential to protect data transmitted between an AI agent and its users?

Keep studying — Plan and configure agent solutions