MCQs on AI Agents 1. An AI agent perceives its environment through: A. Sensors B. Actuators C. Programs D. Processors ✅ Correct Answ...
MCQs on AI Agents
1. An AI agent perceives its environment through:
A. Sensors
B. Actuators
C. Programs
D. Processors
✅ Correct Answer: A. Sensors
Explanation: Sensors are used by agents to perceive the environment.
2. An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through:
A. Memory
B. Effectors
C. Processors
D. Actuators
✅ Correct Answer: D. Actuators
Explanation: Actuators allow agents to take actions in the environment.
3. A rational agent always chooses the action that:
A. Is fastest
B. Maximizes the expected performance measure
C. Is simplest
D. Takes the least resources
✅ Correct Answer: B. Maximizes the expected performance measure
Explanation: Rationality is defined in terms of performance measure, percept sequence, and knowledge.
4. Which of the following is NOT a type of agent in AI?
A. Simple reflex agent
B. Model-based reflex agent
C. Learning agent
D. Cloud-based agent
✅ Correct Answer: D. Cloud-based agent
Explanation: It's a technology, not an agent type. AI agents include reflex, model-based, goal-based, utility-based, and learning agents.
5. The task environment is described by PEAS, which stands for:
A. Perception, Environment, Agent, Sensor
B. Performance measure, Environment, Actuators, Sensors
C. Performance, Environment, Automation, State
D. Planning, Execution, Action, Strategy
✅ Correct Answer: B. Performance measure, Environment, Actuators, Sensors
Explanation: PEAS is used to describe the agent's interaction with its environment.
MCQs on Agentic AI (Advanced Concept)
6. Agentic AI primarily refers to systems that are:
A. Programmed using Java
B. Autonomous, goal-directed, and capable of self-reflection
C. Dependent on human input only
D. Non-interactive systems
✅ Correct Answer: B. Autonomous, goal-directed, and capable of self-reflection
Explanation: Agentic AI refers to AI that can operate as agents — independently planning and executing tasks.
7. A key distinction of Agentic AI compared to traditional AI systems is:
A. It requires no data
B. It reacts without reasoning
C. It uses goal-driven planning and adaptive behavior
D. It does not use sensors or actuators
✅ Correct Answer: C. It uses goal-driven planning and adaptive behavior
Explanation: Agentic AI emphasizes long-term goal pursuit and adaptability.
8. Which of the following best describes autonomy in Agentic AI?
A. Following fixed rules without learning
B. Operating only under human supervision
C. Making decisions and acting without human intervention
D. Random decision making
✅ Correct Answer: C. Making decisions and acting without human intervention
Explanation: Autonomy means the agent acts on its own, based on its goals and environment.
9. Agentic AI systems like AutoGPT or BabyAGI typically:
A. Wait for user instructions at every step
B. Operate without a memory or goal
C. Create, evaluate, and execute subgoals independently
D. Only classify data
✅ Correct Answer: C. Create, evaluate, and execute subgoals independently
Explanation: These systems exhibit agentic behavior — they plan tasks on their own to fulfill a main goal.
10. In Agentic AI, feedback loops are used to:
A. Slow down decision making
B. Enhance human control
C. Improve the agent’s performance over time
D. Eliminate all goals
✅ Correct Answer: C. Improve the agent’s performance over time
Explanation: Feedback helps agentic systems refine actions based on outcomes and adapt to new information.
11. A goal-based agent differs from a reflex agent because it:
A. Ignores the environment
B. Has a specific goal and makes decisions to achieve it
C. Does not require sensors
D. Performs only random actions
✅ Correct Answer: B. Has a specific goal and makes decisions to achieve it
Explanation: Goal-based agents plan actions to reach a desired goal state.
12. What kind of agent keeps track of the world using an internal model?
A. Simple reflex agent
B. Model-based agent
C. Utility-based agent
D. Learning agent
✅ Correct Answer: B. Model-based agent
Explanation: These agents maintain a model of the world to handle partially observable environments.
13. Which agent type can learn from experience and adapt?
A. Reflex agent
B. Model-based agent
C. Utility-based agent
D. Learning agent
✅ Correct Answer: D. Learning agent
Explanation: Learning agents improve their performance over time using feedback.
14. Which of the following best defines an environment in AI?
A. Internal memory of the agent
B. Everything external to the agent it can interact with
C. Database of agent knowledge
D. Fixed program written in Python
✅ Correct Answer: B. Everything external to the agent it can interact with
Explanation: The environment includes all surroundings and factors the agent senses and responds to.
15. The performance of an agent is measured by:
A. The agent’s code
B. Its training data
C. The environment's complexity
D. A performance measure defined in PEAS
✅ Correct Answer: D. A performance measure defined in PEAS
Explanation: The performance measure is the standard used to judge how well an agent performs.
MCQs on Agentic AI (Advanced)
16. Which of the following is NOT a feature of Agentic AI?
A. Independent goal pursuit
B. Reactive behavior only
C. Planning and memory usage
D. Task decomposition
✅ Correct Answer: B. Reactive behavior only
Explanation: Agentic AI is proactive, not just reactive.
17. Agentic AI systems often utilize:
A. Rule-based hardcoded instructions only
B. Static databases
C. Large Language Models (LLMs) for dynamic reasoning
D. Game engines only
✅ Correct Answer: C. Large Language Models (LLMs) for dynamic reasoning
Explanation: Many agentic frameworks use LLMs like GPT-4 or Claude for flexible decision making.
18. Which one is an example of Agentic AI in real-world tools?
A. ChatGPT in default mode
B. Microsoft Excel macros
C. AutoGPT or Devin AI
D. Photoshop filters
✅ Correct Answer: C. AutoGPT or Devin AI
Explanation: These tools can take a goal, plan multiple steps, and adapt to outcomes autonomously.
19. One challenge in Agentic AI development is:
A. Making agents reactive
B. Preventing internet connectivity
C. Aligning goals with human values
D. Disabling autonomy features
✅ Correct Answer: C. Aligning goals with human values
Explanation: Ensuring safe and aligned behavior is a major research area in agentic systems.
20. What allows Agentic AI to perform long-horizon tasks?
A. Pre-trained answers only
B. Manual user input for each step
C. Planning, memory, and recursive reasoning
D. Only visual processing
✅ Correct Answer: C. Planning, memory, and recursive reasoning
Explanation: These capabilities enable agentic systems to work on complex tasks over time.AI Agents – MCQs (Basic to Intermediate)
21. In AI, the percept sequence refers to:
A. A list of programs the agent has
B. A sequence of environmental states
C. Complete history of what the agent has perceived
D. Steps taken by the environment
✅ Correct Answer: C. Complete history of what the agent has perceived
22. What is the key characteristic of a utility-based agent?
A. Acts without reasoning
B. Chooses randomly
C. Makes decisions based on maximizing a utility function
D. Operates without a goal
✅ Correct Answer: C. Makes decisions based on maximizing a utility function
23. What is the role of actuators in an agent?
A. Thinking
B. Observing the environment
C. Taking action in the environment
D. Creating goals
✅ Correct Answer: C. Taking action in the environment
24. Which of the following is not a valid agent program?
A. Reflex program
B. Goal-based program
C. Learning program
D. Repetition program
✅ Correct Answer: D. Repetition program
25. In a vacuum cleaner world, what does a simple reflex agent do?
A. Remembers past actions
B. Always moves left
C. Acts based on current location and dirt sensor
D. Plans a full route
✅ Correct Answer: C. Acts based on current location and dirt sensor
26. The environment for a chess-playing AI is:
A. Fully observable, deterministic
B. Partially observable and stochastic
C. Static and episodic
D. Dynamic and single-agent
✅ Correct Answer: A. Fully observable, deterministic
27. Which of these is a property of an environment in AI?
A. Reactive
B. Sensor-based
C. Observable or unobservable
D. Logically programmed
✅ Correct Answer: C. Observable or unobservable
28. What is the correct PEAS description for a self-driving car?
A. Performance: speed; Sensors: clock
B. Performance: safety, legality; Environment: roads; Sensors: cameras; Actuators: wheels
C. Performance: look; Environment: apps; Sensors: radar only
D. None of the above
✅ Correct Answer: B. Performance: safety, legality; Environment: roads; Sensors: cameras; Actuators: wheels
29. Which of the following types of agents does NOT use internal state?
A. Simple reflex agent
B. Model-based agent
C. Goal-based agent
D. Utility-based agent
✅ Correct Answer: A. Simple reflex agent
30. Which AI agent type can deal with uncertainty better?
A. Simple reflex
B. Model-based
C. Utility-based
D. Static agent
✅ Correct Answer: C. Utility-based
✅ Agentic AI – MCQs (Emerging Concepts)
31. Agentic AI systems are often designed to:
A. Wait for step-by-step instructions
B. Mimic human emotions
C. Autonomously pursue goals and adapt
D. Delete their own memory
✅ Correct Answer: C. Autonomously pursue goals and adapt
32. What makes Agentic AI more powerful than traditional chatbots?
A. More emojis
B. Reciting memorized facts
C. Goal-setting, planning, and reflection
D. Loud voice interaction
✅ Correct Answer: C. Goal-setting, planning, and reflection
33. What is a "replanning loop" in Agentic AI?
A. Fixed execution path
B. Restarting the app
C. Updating the plan when conditions change
D. Breaking the system intentionally
✅ Correct Answer: C. Updating the plan when conditions change
34. An Agentic AI system working on a research report would typically:
A. Finish in one step
B. Ask a user what to do next every time
C. Break it into subtasks and solve them
D. Only fetch facts
✅ Correct Answer: C. Break it into subtasks and solve them
35. Which is an example of agentic memory usage?
A. Forgetting past tasks
B. Logging goals, subgoals, and observations
C. Only storing commands
D. External video storage
✅ Correct Answer: B. Logging goals, subgoals, and observations
36. Which of these is a key risk in Agentic AI systems?
A. GPU overheating
B. Misaligned goal pursuit
C. Slow boot time
D. Using too few variables
✅ Correct Answer: B. Misaligned goal pursuit
37. AutoGPT is considered an agentic system because it:
A. Creates images
B. Asks random questions
C. Plans and executes tasks without constant input
D. Plays games
✅ Correct Answer: C. Plans and executes tasks without constant input
38. What’s the difference between Agentic AI and reactive AI?
A. Speed
B. Vision system
C. Agentic AI plans and adjusts; reactive just responds
D. Programming language
✅ Correct Answer: C. Agentic AI plans and adjusts; reactive just responds
39. A long-horizon task refers to:
A. A task with one step
B. A task requiring multi-step planning over time
C. A real-time computation
D. Zooming into images
✅ Correct Answer: B. A task requiring multi-step planning over time
40. Agentic AI is inspired by:
A. Hard-coded systems
B. Human cognition and decision-making
C. Social media algorithms
D. Fixed control loops
✅ Correct Answer: B. Human cognition and decision-making
✅ Mixed Conceptual Questions
41. An agent that can reason about its own reasoning is called:
A. Reflexive
B. Meta-cognitive agent
C. Random
D. Non-intelligent
✅ Correct Answer: B. Meta-cognitive agent
42. Multi-agent systems consist of:
A. A single AI agent
B. Hardware components
C. Multiple AI agents interacting
D. Random decision-makers
✅ Correct Answer: C. Multiple AI agents interacting
43. The environment of a robotic vacuum cleaner is:
A. Static, fully observable
B. Dynamic, partially observable
C. Episodic and isolated
D. Deterministic and noisy
✅ Correct Answer: B. Dynamic, partially observable
44. A bounded rational agent is one that:
A. Acts randomly
B. Ignores goals
C. Makes the best decision given its limitations
D. Has infinite computing power
✅ Correct Answer: C. Makes the best decision given its limitations
45. A reflex agent fails when:
A. Environment is static
B. There's no goal
C. It faces unfamiliar situations
D. Memory is large
✅ Correct Answer: C. It faces unfamiliar situations
46. One example of Agentic AI in productivity is:
A. Photoshop
B. Google Sheets
C. Devin AI (autonomous software engineer)
D. Microsoft Paint
✅ Correct Answer: C. Devin AI
47. In multi-agent systems, cooperation is used for:
A. Competing only
B. Increasing memory
C. Solving tasks jointly
D. Debugging code
✅ Correct Answer: C. Solving tasks jointly
48. The key difference between learning agents and others is:
A. Use of programming
B. Memory access
C. Ability to improve performance over time
D. Actuation
✅ Correct Answer: C. Ability to improve performance over time
49. A stochastic environment is one where:
A. The outcome is predictable
B. Everything is random
C. There is uncertainty in outcomes
D. Agents never act
✅ Correct Answer: C. There is uncertainty in outcomes
50. The main benefit of Agentic AI in the future is:
A. Limited use
B. Automating complex goals with minimal human input
C. Replacing entertainment
D. Storing images
✅ Correct Answer: B. Automating complex goals with minimal human input
COMMENTS