6.2 Explain AI Agent Design Principles, Autonomous Capabilities, and Orchestration AI-TECHNICAL-PRACTITIONER Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A logistics company deploys an AI agent to monitor warehouse inventory. The agent must reorder stock when levels drop below a threshold, but the finance team requires that every purchase order be approved by a human before submission to suppliers. Which design principle should the agent incorporate to satisfy both requirements?

   • AA goal specification that instructs the agent to only generate a recommendation and never take action
   • BAn environment interaction model where the agent monitors inventory but delegates all reordering decisions to a separate human-operated system
   • CA feedback loop that pauses the action-execution phase until a human approves the generated purchase order
   • DA perception module that alerts a human whenever inventory is low, leaving the reorder decision entirely to the human
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A cybersecurity firm deploys an AI agent to detect and respond to network intrusions. The agent's goal is to 'protect the network.' During a test, the agent responds to a minor anomaly by disconnecting the entire network, causing a major outage. What is the most likely cause of this behavior?

   • AThe agent's feedback loop was not configured to learn from past incidents
   • BThe orchestration system failed to delegate the response to a more specialized agent
   • CThe agent's perception module misidentified the minor anomaly as a critical threat
   • DThe goal specification lacks constraints on the severity of actions, so the agent chose the most drastic action that satisfies 'protect the network'
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   What is the primary role of orchestration in a multi-agent AI system?

   • ATo coordinate the activities of multiple agents to achieve a common goal
   • BTo replace the need for individual agents to have goals
   • CTo ensure that only one agent operates at a time
   • DTo eliminate the need for feedback loops in agents
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   In AI agent design, what is the primary purpose of a feedback loop?

   • ATo replace the need for a defined goal
   • BTo adjust future behavior based on outcomes
   • CTo prevent the agent from interacting with the environment
   • DTo define the agent's initial state
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   What does it mean for an AI agent to have autonomous capabilities?

   • AIt can only perceive its environment but cannot act on it
   • BIt is limited to a single predefined task with no variation
   • CIt can operate without human intervention for perception, decision-making, and action execution
   • DIt requires human approval for every decision it makes
   Reveal Solution Discussion Select an answer first