ARTIFICIAL INTELLIGENCE FOR GAMES
EXAMINATION QUESTIONS AND CORRECT
ANSWE WITH EXPLANATION GRADED
Section 1: Introduction to Game AI
1. Which of the following best describes the primary distinction between academic AI and
game AI?
A. Academic AI focuses on theoretical optimality, while game AI focuses on creating a believable
and engaging player experience.
B. Academic AI is always more computationally expensive than game AI.
C. Game AI is exclusively concerned with pathfinding, while academic AI covers all other
domains.
D. Academic AI is written in Python, while game AI is written in C++.
Correct Answer: A
Rationale: The goal of game AI is not necessarily to find the perfect solution but to create
an illusion of intelligence that is fun, fair, and challenging for the player. Academic AI, in
contrast, often strives for optimal solutions to problems, regardless of whether that is "fun." A is
the most accurate distinction. B is false; many academic AI problems are simplified for games. C
is false; game AI encompasses many domains. D is false; language choice is not the defining
distinction.
2. In game AI, what is the concept of "the illusion of intelligence"?
A. The process of making AI agents appear smarter than they are through clever tricks and
cheats.
B. A technique for rendering intelligent-looking characters.
C. The philosophical question of whether AI can truly be intelligent.
D. A method for debugging AI decision-making processes.
Correct Answer: A
Rationale: The "illusion of intelligence" is a core concept in game AI. Since computational
resources are limited and perfect AI can be boring or frustrating, developers often use
techniques like cheating (e.g., seeing through walls in limited ways) or simplification to make an
agent appear intelligent to the player. B, C, and D are incorrect definitions.
3. Which of the following are common goals of AI in a video game? (Select All That Apply)
A. To provide a challenging and fair opponent.
,B. To solve the P vs. NP problem.
C. To create a believable and immersive world.
D. To replace human game designers entirely.
E. To adapt to player skill and behavior.
Correct Answers: A, C, E
Rationale: Game AI aims to enhance the player's experience by providing challenge (A),
creating believable worlds and characters (C), and adapting to the player (E). Academic AI is
more concerned with theoretical problems like P vs. NP (B). While AI can assist designers, the
goal is not full replacement (D).
4. What is the role of a "game agent" in AI?
A. A human player controlling a character.
B. An autonomous entity that perceives its environment and acts upon it to achieve goals.
C. A software tool used by developers to test game balance.
D. The central processing unit of a game console.
Correct Answer: B
Rationale: In AI, an agent is defined as an entity that perceives its environment through
sensors and acts upon that environment through actuators to achieve specific goals. This is the
standard definition used in game AI for characters. A, C, and D are incorrect.
5. Which of the following are typical inputs to a game AI agent's "perception" system? (Select
All That Apply)
A. Player position.
B. The game's source code.
C. Line-of-sight checks.
D. Sound events.
E. The AI's own internal state (e.g., health, ammo).
Correct Answers: A, C, D, E
Rationale: An AI agent's perception system gathers information from the game world (A, C,
D) and its own internal state (E) to make decisions. The agent would not have access to the
game's source code (B) at runtime.
6. What is a common criticism of using simple, reactive AI (like pure finite state machines) for
complex characters?
A. They are too difficult to implement.
B. They are too computationally expensive.
,C. They can appear predictable and lack long-term planning.
D. They are impossible to debug.
Correct Answer: C
Rationale: Simple reactive systems map a stimulus directly to a response. While easy to
implement and debug, they lack the ability to plan ahead, making them predictable and often
unconvincing for complex roles. A, B, and D are generally not true; simple FSMs are usually easy
to implement and debug and are computationally cheap.
7. The term "AI Director" is most commonly associated with which game genre?
A. Real-Time Strategy (RTS).
B. First-Person Shooter (FPS).
C. Massively Multiplayer Online (MMO).
D. Puzzle.
Correct Answer: B
Rationale: The "AI Director" is a system famously used in the Left 4 Dead series (an FPS) to
dynamically manage pacing, enemy spawns, and item placement based on player performance
and stress levels. While other genres have similar systems, the term is most iconic in the FPS
genre.
8. Which of the following is a key challenge in creating AI for multiplayer online games?
A. Ensuring the AI has perfect knowledge of all players' positions at all times.
B. Balancing AI difficulty for players of vastly different skill levels.
C. Preventing the AI from experiencing lag.
D. The AI must be written in a specific programming language.
Correct Answer: B
Rationale: In a multiplayer environment, a key challenge is creating an AI that is challenging
for a skilled player but not overwhelming for a new player. This often requires dynamic difficulty
adjustment. Perfect knowledge (A) would be cheating and unfair. AI doesn't experience lag in
the same way a networked player does (C). Language (D) is irrelevant.
9. What is the primary purpose of "debugging tools" in game AI development?
A. To make the AI unbeatable.
B. To allow developers to visualize and inspect the internal state and decision-making of an AI
agent.
C. To automatically generate new AI behaviors.
D. To reduce the final file size of the game.
, Correct Answer: B
Rationale: Debugging tools for AI are crucial for understanding why an agent is behaving in
a certain way. They often provide visualizations for things like pathfinding (showing the path),
state machines (showing the current state), and perception (showing what the AI can see).
10. What is the difference between "strong AI" and "weak AI" in the context of games?
A. Strong AI is for main characters, weak AI is for background characters.
B. Strong AI refers to general, human-level intelligence; weak AI (or narrow AI) refers to AI that
is specialized for a specific task.
C. Strong AI is written in C++, weak AI is written in Python.
D. Strong AI is for single-player, weak AI is for multiplayer.
Correct Answer: B
Rationale: This is a fundamental distinction in AI. Weak AI (or Narrow AI) is what all game AI
is today—it is designed to perform a specific task (e.g., play chess, navigate a level). Strong AI
(Artificial General Intelligence) would possess human-like reasoning and problem-solving
abilities across a wide range of domains, which does not currently exist in games.
Section 2: Movement and Pathfinding
11. Which pathfinding algorithm is best suited for a static, unchanging environment where
you need to find a path between any two points quickly?
A. A* (A-Star) Search.
B. Dijkstra's Algorithm.
C. Breadth-First Search (BFS).
D. The Floyd-Warshall Algorithm.
Correct Answer: D
Rationale: The Floyd-Warshall algorithm computes the shortest paths between all pairs of
nodes in a single run. For a static environment, this pre-computation allows for extremely fast
pathfinding queries (O(1) lookups). A* and Dijkstra's (B, C) are used for single-source queries
and would need to be run for every request. A* is generally faster than Dijkstra for a single
query due to its heuristic, but Floyd-Warshall is superior for the "any-to-any" scenario in a static
graph.
12. What is the primary advantage of using a navigation mesh (navmesh) over a grid for
pathfinding?
A. Navmeshes are always easier to generate automatically.
B. Navmeshes provide a more memory-efficient and often more natural representation of
walkable space.
EXAMINATION QUESTIONS AND CORRECT
ANSWE WITH EXPLANATION GRADED
Section 1: Introduction to Game AI
1. Which of the following best describes the primary distinction between academic AI and
game AI?
A. Academic AI focuses on theoretical optimality, while game AI focuses on creating a believable
and engaging player experience.
B. Academic AI is always more computationally expensive than game AI.
C. Game AI is exclusively concerned with pathfinding, while academic AI covers all other
domains.
D. Academic AI is written in Python, while game AI is written in C++.
Correct Answer: A
Rationale: The goal of game AI is not necessarily to find the perfect solution but to create
an illusion of intelligence that is fun, fair, and challenging for the player. Academic AI, in
contrast, often strives for optimal solutions to problems, regardless of whether that is "fun." A is
the most accurate distinction. B is false; many academic AI problems are simplified for games. C
is false; game AI encompasses many domains. D is false; language choice is not the defining
distinction.
2. In game AI, what is the concept of "the illusion of intelligence"?
A. The process of making AI agents appear smarter than they are through clever tricks and
cheats.
B. A technique for rendering intelligent-looking characters.
C. The philosophical question of whether AI can truly be intelligent.
D. A method for debugging AI decision-making processes.
Correct Answer: A
Rationale: The "illusion of intelligence" is a core concept in game AI. Since computational
resources are limited and perfect AI can be boring or frustrating, developers often use
techniques like cheating (e.g., seeing through walls in limited ways) or simplification to make an
agent appear intelligent to the player. B, C, and D are incorrect definitions.
3. Which of the following are common goals of AI in a video game? (Select All That Apply)
A. To provide a challenging and fair opponent.
,B. To solve the P vs. NP problem.
C. To create a believable and immersive world.
D. To replace human game designers entirely.
E. To adapt to player skill and behavior.
Correct Answers: A, C, E
Rationale: Game AI aims to enhance the player's experience by providing challenge (A),
creating believable worlds and characters (C), and adapting to the player (E). Academic AI is
more concerned with theoretical problems like P vs. NP (B). While AI can assist designers, the
goal is not full replacement (D).
4. What is the role of a "game agent" in AI?
A. A human player controlling a character.
B. An autonomous entity that perceives its environment and acts upon it to achieve goals.
C. A software tool used by developers to test game balance.
D. The central processing unit of a game console.
Correct Answer: B
Rationale: In AI, an agent is defined as an entity that perceives its environment through
sensors and acts upon that environment through actuators to achieve specific goals. This is the
standard definition used in game AI for characters. A, C, and D are incorrect.
5. Which of the following are typical inputs to a game AI agent's "perception" system? (Select
All That Apply)
A. Player position.
B. The game's source code.
C. Line-of-sight checks.
D. Sound events.
E. The AI's own internal state (e.g., health, ammo).
Correct Answers: A, C, D, E
Rationale: An AI agent's perception system gathers information from the game world (A, C,
D) and its own internal state (E) to make decisions. The agent would not have access to the
game's source code (B) at runtime.
6. What is a common criticism of using simple, reactive AI (like pure finite state machines) for
complex characters?
A. They are too difficult to implement.
B. They are too computationally expensive.
,C. They can appear predictable and lack long-term planning.
D. They are impossible to debug.
Correct Answer: C
Rationale: Simple reactive systems map a stimulus directly to a response. While easy to
implement and debug, they lack the ability to plan ahead, making them predictable and often
unconvincing for complex roles. A, B, and D are generally not true; simple FSMs are usually easy
to implement and debug and are computationally cheap.
7. The term "AI Director" is most commonly associated with which game genre?
A. Real-Time Strategy (RTS).
B. First-Person Shooter (FPS).
C. Massively Multiplayer Online (MMO).
D. Puzzle.
Correct Answer: B
Rationale: The "AI Director" is a system famously used in the Left 4 Dead series (an FPS) to
dynamically manage pacing, enemy spawns, and item placement based on player performance
and stress levels. While other genres have similar systems, the term is most iconic in the FPS
genre.
8. Which of the following is a key challenge in creating AI for multiplayer online games?
A. Ensuring the AI has perfect knowledge of all players' positions at all times.
B. Balancing AI difficulty for players of vastly different skill levels.
C. Preventing the AI from experiencing lag.
D. The AI must be written in a specific programming language.
Correct Answer: B
Rationale: In a multiplayer environment, a key challenge is creating an AI that is challenging
for a skilled player but not overwhelming for a new player. This often requires dynamic difficulty
adjustment. Perfect knowledge (A) would be cheating and unfair. AI doesn't experience lag in
the same way a networked player does (C). Language (D) is irrelevant.
9. What is the primary purpose of "debugging tools" in game AI development?
A. To make the AI unbeatable.
B. To allow developers to visualize and inspect the internal state and decision-making of an AI
agent.
C. To automatically generate new AI behaviors.
D. To reduce the final file size of the game.
, Correct Answer: B
Rationale: Debugging tools for AI are crucial for understanding why an agent is behaving in
a certain way. They often provide visualizations for things like pathfinding (showing the path),
state machines (showing the current state), and perception (showing what the AI can see).
10. What is the difference between "strong AI" and "weak AI" in the context of games?
A. Strong AI is for main characters, weak AI is for background characters.
B. Strong AI refers to general, human-level intelligence; weak AI (or narrow AI) refers to AI that
is specialized for a specific task.
C. Strong AI is written in C++, weak AI is written in Python.
D. Strong AI is for single-player, weak AI is for multiplayer.
Correct Answer: B
Rationale: This is a fundamental distinction in AI. Weak AI (or Narrow AI) is what all game AI
is today—it is designed to perform a specific task (e.g., play chess, navigate a level). Strong AI
(Artificial General Intelligence) would possess human-like reasoning and problem-solving
abilities across a wide range of domains, which does not currently exist in games.
Section 2: Movement and Pathfinding
11. Which pathfinding algorithm is best suited for a static, unchanging environment where
you need to find a path between any two points quickly?
A. A* (A-Star) Search.
B. Dijkstra's Algorithm.
C. Breadth-First Search (BFS).
D. The Floyd-Warshall Algorithm.
Correct Answer: D
Rationale: The Floyd-Warshall algorithm computes the shortest paths between all pairs of
nodes in a single run. For a static environment, this pre-computation allows for extremely fast
pathfinding queries (O(1) lookups). A* and Dijkstra's (B, C) are used for single-source queries
and would need to be run for every request. A* is generally faster than Dijkstra for a single
query due to its heuristic, but Floyd-Warshall is superior for the "any-to-any" scenario in a static
graph.
12. What is the primary advantage of using a navigation mesh (navmesh) over a grid for
pathfinding?
A. Navmeshes are always easier to generate automatically.
B. Navmeshes provide a more memory-efficient and often more natural representation of
walkable space.
