Plan an Agent Solution AB-620 Practice Questions (Page 8)

Questions 36–40

1. 36foundation · easy

   Which deployment strategy involves gradually rolling out an AI agent to a small subset of users before full deployment?

   • ARecreate deployment
   • BBig bang deployment
   • CCanary deployment
   • DBlue-green deployment
   Reveal Solution Discussion Select an answer first
2. 37expert · hard

   A hospital is deploying an AI agent that allows patients to book appointments. The agent will access patient records to suggest available times. The hospital must comply with health data regulations and ensure that only the patient can access their own records. What security measure is essential?

   • AStore patient records in a separate database for each patient.
   • BUse multi-factor authentication for patients when they access the agent.
   • CImplement a secure authentication mechanism that uniquely identifies the patient and scopes the agent's data access to that patient's records.
   • DEncrypt all data transmitted between the agent and the patient.
   Reveal Solution Discussion Select an answer first
4. 38foundation · easy

   When designing an AI agent for an external audience, which consideration is most important?

   • AUsing internal jargon and technical terms without explanation.
   • BEnsuring the agent can handle a wide variety of user inputs and provide clear, helpful responses.
   • CLimiting the agent to a small set of predefined responses.
   • DRequiring users to have deep knowledge of the underlying system.
   Reveal Solution Discussion Select an answer first
5. 39expert · hard

   A healthcare organization is deploying an AI agent that provides appointment scheduling. They must comply with strict data residency requirements and need to validate the agent's behavior before production. They have a development environment in a different region. What deployment strategy should they use?

   • AUse a canary deployment in production with a small percentage of real users.
   • BUse a staging environment in the same region as production, with production data masked, then promote to production after validation.
   • CDeploy the agent directly to production in the required region and test with synthetic data.
   • DDevelop and test in the development region, then deploy the same code to production in the required region.
   Reveal Solution Discussion Select an answer first
6. 40application · medium

   A healthcare organization is deploying an AI agent that provides general wellness tips to employees. The agent is internal and does not handle personal health information. However, the organization wants to ensure the agent does not give harmful advice and that there is accountability for its responses. Which responsible AI strategy should you plan?

   • AOnly log interactions that are flagged as problematic to reduce storage costs.
   • BAllow the agent to respond freely without filters to provide the most helpful advice.
   • CImplement content filters, provide disclaimers, and log all interactions for review.
   • DRely on the agent's built-in safety features without additional configuration.
   Reveal Solution Discussion Select an answer first