Skip to content

Quiz: Intelligent Agents and Generative AI

Test your understanding of intelligent agents and generative ai with these review questions.

1. What term names the following idea: “Artificial intelligence that produces new text, code, images, data structures, or other artifacts from prompts and context.”?

  1. Tool Use (Agent)
  2. Generative AI
  3. Iterative Workflow
  4. LLM-Generated Dependency Graph
Show Answer

The correct answer is B. Generative AI is artificial intelligence that produces new text, code, images, data structures, or other artifacts from prompts and context. The chapter's terminology supports the same distinction in practice: An AI-assisted workflow can use generative ai to generate artifacts while keeping the result inspectable by humans. Tool Use (Agent), Iterative Workflow, and LLM-Generated Dependency Graph name neighboring ideas, but none has this defining purpose. The decisive clue is the function described in the stem, not merely the fact that all four terms belong to Intelligent Agents and Generative AI.

Concept Tested: Generative AI

See: Intelligent Agents and Generative AI

2. Which statement best explains what Intelligent Agent contributes to an agentic workflow grounded in graph-structured course knowledge?

  1. The practice of designing prompts, examples, constraints, and context to guide generative AI output.
  2. A conversational software interface that accepts user messages and returns generated or retrieved responses.
  3. A software system that uses goals, context, tools, and feedback to decide and perform actions.
  4. A generation pattern that retrieves relevant source material before producing an answer or artifact.
Show Answer

The correct answer is C. The requirement points to Intelligent Agent: A software system that uses goals, context, tools, and feedback to decide and perform actions. In an agentic workflow grounded in graph-structured course knowledge, that function is what separates it from Prompt Engineering, Chatbot, and Retrieval-Augmented Generation. Those alternatives may participate in the same workflow, but substituting one of them would change the role the question asks the practitioner to identify.

Concept Tested: Intelligent Agent

See: Intelligent Agents and Generative AI

3. A project team building an agentic workflow grounded in graph-structured course knowledge needs the capability described as “The degree to which an agent can plan, decide, and act without step-by-step human direction.” Which concept should the team apply?

  1. Agent Autonomy
  2. GraphRAG
  3. Learning Agent
  4. Generative AI
Show Answer

The correct answer is A. The degree to which an agent can plan, decide, and act without step-by-step human direction. That is the chapter's specific meaning of Agent Autonomy. By contrast, GraphRAG, Learning Agent, and Generative AI solve different parts of the larger problem. A sound choice therefore follows the stated capability or relationship rather than selecting the most familiar term from Intelligent Agents and Generative AI.

Concept Tested: Agent Autonomy

See: Intelligent Agents and Generative AI

  1. A software system that uses goals, context, tools, and feedback to decide and perform actions.
  2. A workflow in which multiple agents coordinate specialized tasks to complete a larger objective.
  3. The degree to which an agent can plan, decide, and act without step-by-step human direction.
  4. An agent capability that allows software tools, files, APIs, or external systems to be called during work.
Show Answer

The correct answer is B. Multi-Agent Collaboration matches because it is a workflow in which multiple agents coordinate specialized tasks to complete a larger objective. The stem describes exactly that responsibility within an agentic workflow grounded in graph-structured course knowledge. The three distractors—Intelligent Agent, Agent Autonomy, and Tool Use (Agent)—remain plausible because they are related, yet their definitions do not satisfy the stated criterion as directly.

Concept Tested: Multi-Agent Collaboration

See: Intelligent Agents and Generative AI

5. How should a practitioner interpret Tool Use (Agent) when working with an agentic workflow grounded in graph-structured course knowledge?

  1. An agent capability that allows software tools, files, APIs, or external systems to be called during work.
  2. A repeated cycle of generation, review, revision, and validation.
  3. A dependency graph whose nodes or edges are produced by a large language model from prompts and source material.
  4. The practice of designing prompts, examples, constraints, and context to guide generative AI output.
Show Answer

The correct answer is A. The key distinction is that Tool Use (Agent) is an agent capability that allows software tools, files, APIs, or external systems to be called during work. Iterative Workflow, LLM-Generated Dependency Graph, and Prompt Engineering refer to other mechanisms or structures discussed in the chapter. Reading the stem as a functional requirement makes Tool Use (Agent) the only defensible match and prevents a choice based on surface vocabulary alone.

Concept Tested: Tool Use (Agent)

See: Intelligent Agents and Generative AI

6. A project team building an agentic workflow grounded in graph-structured course knowledge needs the capability described as “A repeated cycle of generation, review, revision, and validation.” Which concept should the team apply?

  1. Chatbot
  2. Retrieval-Augmented Generation
  3. GraphRAG
  4. Iterative Workflow
Show Answer

The correct answer is D. In Intelligent Agents and Generative AI, Iterative Workflow means a repeated cycle of generation, review, revision, and validation. The other options describe adjacent concepts rather than synonyms. Because the prompt asks for this exact function in an agentic workflow grounded in graph-structured course knowledge, their conceptual proximity does not make them interchangeable with Iterative Workflow.

Concept Tested: Iterative Workflow

See: Intelligent Agents and Generative AI

7. Which recommendation should reviewers accept if LLM-Generated Dependency Graph is the governing criterion for an agentic workflow grounded in graph-structured course knowledge?

  1. A dependency graph whose nodes or edges are produced by a large language model from prompts and source material.
  2. An intelligent agent that uses learner data and educational goals to recommend, generate, or adapt instruction.
  3. Artificial intelligence that produces new text, code, images, data structures, or other artifacts from prompts and context.
  4. A software system that uses goals, context, tools, and feedback to decide and perform actions.
Show Answer

The correct answer is A. A practitioner would select LLM-Generated Dependency Graph because A dependency graph whose nodes or edges are produced by a large language model from prompts and source material. The scenario requires that capability specifically. Learning Agent, Generative AI, and Intelligent Agent can still matter elsewhere in an agentic workflow grounded in graph-structured course knowledge, but each would answer a different design or analysis question.

Concept Tested: LLM-Generated Dependency Graph

See: Intelligent Agents and Generative AI

8. A reviewer is tracing how the parts of an agentic workflow grounded in graph-structured course knowledge fit together. Which description identifies Prompt Engineering's specific role?

  1. The degree to which an agent can plan, decide, and act without step-by-step human direction.
  2. The practice of designing prompts, examples, constraints, and context to guide generative AI output.
  3. A workflow in which multiple agents coordinate specialized tasks to complete a larger objective.
  4. An agent capability that allows software tools, files, APIs, or external systems to be called during work.
Show Answer

The correct answer is B. The evidence in the stem corresponds to Prompt Engineering, defined here as the practice of designing prompts, examples, constraints, and context to guide generative AI output. This evidence does not establish Agent Autonomy, Multi-Agent Collaboration, or Tool Use (Agent); those concepts require different defining features. The distinction matters when analyzing how the parts of an agentic workflow grounded in graph-structured course knowledge fit together.

Concept Tested: Prompt Engineering

See: Intelligent Agents and Generative AI

9. A project team building an agentic workflow grounded in graph-structured course knowledge needs the capability described as “A conversational software interface that accepts user messages and returns generated or retrieved responses.” Which concept should the team apply?

  1. Iterative Workflow
  2. LLM-Generated Dependency Graph
  3. Prompt Engineering
  4. Chatbot
Show Answer

The correct answer is D. Chatbot is the precise term because it is a conversational software interface that accepts user messages and returns generated or retrieved responses. Choosing Iterative Workflow, LLM-Generated Dependency Graph, or Prompt Engineering would broaden or redirect the requirement. The keyed option preserves the narrower meaning established in Intelligent Agents and Generative AI and applies it to the professional scenario in the stem.

Concept Tested: Chatbot

See: Intelligent Agents and Generative AI

10. Which design choice would most directly build Retrieval-Augmented Generation into an agentic workflow grounded in graph-structured course knowledge?

  1. A conversational software interface that accepts user messages and returns generated or retrieved responses.
  2. A retrieval-augmented generation approach that uses graph structure to organize, retrieve, or reason over context.
  3. A generation pattern that retrieves relevant source material before producing an answer or artifact.
  4. An intelligent agent that uses learner data and educational goals to recommend, generate, or adapt instruction.
Show Answer

The correct answer is C. The proposed design implements Retrieval-Augmented Generation, which is a generation pattern that retrieves relevant source material before producing an answer or artifact. The alternatives emphasize Chatbot, GraphRAG, or Learning Agent instead. They could complement the design, but they do not by themselves create the capability the question requires for an agentic workflow grounded in graph-structured course knowledge.

Concept Tested: Retrieval-Augmented Generation

See: Intelligent Agents and Generative AI