WGU D685 OBJECTIVE ASSESSMENT
AND PA NEWEST 2025/ 2026 TEST
BANK| D685 PRACTICAL
APPLICATIONS OF PROMPT OA & PA
EXAM WITH COMPLETE 200 REAL
EXAM QUESTIONS AND CORRECT
VERIFIED ANSWERS/ ALREADY
GRADED A+ (MOST RECENT!!)
1. What is the primary goal of prompt engineering?
A) To make AI models run faster
B) To craft inputs that elicit the most accurate and useful
outputs from an AI
C) To train AI models from scratch
D) To prevent AI from generating any text
✅ Correct Answer: B
Rationale: Prompt engineering focuses on designing input queries
to optimize AI responses. (A) is about performance optimization,
,not prompting. (C) is model training. (D) is censorship, not
engineering.
2. Which term describes an AI generating factually incorrect
but confident-sounding information?
A) Overfitting
B) Hallucination
C) Gradient descent
D) Tokenization
✅ Correct Answer: B
Rationale: Hallucination refers to AI generating false or
nonsensical information confidently. (A) is a training error. (C) is
optimization. (D) is text preprocessing.
3. You want the AI to act as a career coach. Which prompt
component is being used?
A) Output format
B) Persona
C) Context
D) Token limit
✅ Correct Answer: B
Rationale: Assigning a role (career coach) is persona-setting. (A)
,defines response structure. (C) provides background. (D) is length
constraint.
4. What does "zero-shot prompting" mean?
A) Giving the AI many examples before asking the task
B) Asking the AI to perform a task without any examples
C) Providing only incorrect examples
D) Using images instead of text
✅ Correct Answer: B
Rationale: Zero-shot means no examples are provided. (A) is few-
shot. (C) is counterproductive. (D) is multimodal prompting.
5. Which prompting technique is best for solving multi-step
math problems?
A) Zero-shot
B) Chain-of-thought
C) Persona-based
D) Role-playing
✅ Correct Answer: B
Rationale: Chain-of-thought (CoT) asks the model to reason step-
by-step, ideal for math. (A) often fails on complex logic. (C) and
(D) are not reasoning-specific.
, 6. What is a major ethical risk when entering employee data
into a public AI model?
A) Faster processing
B) Privacy violation
C) Improved accuracy
D) Reduced bias
✅ Correct Answer: B
Rationale: Public AI models may store or train on input data,
risking sensitive employee info leaks. (A) and (C) are not risks. (D)
is unlikely without safeguards.
7. Which prompt would likely produce the most relevant
image?
A) "Draw something nice"
B) "Create a sunset"
C) "Generate a photorealistic image of a golden retriever
playing fetch on a beach at sunset, high resolution"
D) "Make art"
✅ Correct Answer: C
Rationale: Specificity (breed, action, setting, style, resolution)
yields better image outputs. (A), (B), (D) are vague and
ambiguous.
AND PA NEWEST 2025/ 2026 TEST
BANK| D685 PRACTICAL
APPLICATIONS OF PROMPT OA & PA
EXAM WITH COMPLETE 200 REAL
EXAM QUESTIONS AND CORRECT
VERIFIED ANSWERS/ ALREADY
GRADED A+ (MOST RECENT!!)
1. What is the primary goal of prompt engineering?
A) To make AI models run faster
B) To craft inputs that elicit the most accurate and useful
outputs from an AI
C) To train AI models from scratch
D) To prevent AI from generating any text
✅ Correct Answer: B
Rationale: Prompt engineering focuses on designing input queries
to optimize AI responses. (A) is about performance optimization,
,not prompting. (C) is model training. (D) is censorship, not
engineering.
2. Which term describes an AI generating factually incorrect
but confident-sounding information?
A) Overfitting
B) Hallucination
C) Gradient descent
D) Tokenization
✅ Correct Answer: B
Rationale: Hallucination refers to AI generating false or
nonsensical information confidently. (A) is a training error. (C) is
optimization. (D) is text preprocessing.
3. You want the AI to act as a career coach. Which prompt
component is being used?
A) Output format
B) Persona
C) Context
D) Token limit
✅ Correct Answer: B
Rationale: Assigning a role (career coach) is persona-setting. (A)
,defines response structure. (C) provides background. (D) is length
constraint.
4. What does "zero-shot prompting" mean?
A) Giving the AI many examples before asking the task
B) Asking the AI to perform a task without any examples
C) Providing only incorrect examples
D) Using images instead of text
✅ Correct Answer: B
Rationale: Zero-shot means no examples are provided. (A) is few-
shot. (C) is counterproductive. (D) is multimodal prompting.
5. Which prompting technique is best for solving multi-step
math problems?
A) Zero-shot
B) Chain-of-thought
C) Persona-based
D) Role-playing
✅ Correct Answer: B
Rationale: Chain-of-thought (CoT) asks the model to reason step-
by-step, ideal for math. (A) often fails on complex logic. (C) and
(D) are not reasoning-specific.
, 6. What is a major ethical risk when entering employee data
into a public AI model?
A) Faster processing
B) Privacy violation
C) Improved accuracy
D) Reduced bias
✅ Correct Answer: B
Rationale: Public AI models may store or train on input data,
risking sensitive employee info leaks. (A) and (C) are not risks. (D)
is unlikely without safeguards.
7. Which prompt would likely produce the most relevant
image?
A) "Draw something nice"
B) "Create a sunset"
C) "Generate a photorealistic image of a golden retriever
playing fetch on a beach at sunset, high resolution"
D) "Make art"
✅ Correct Answer: C
Rationale: Specificity (breed, action, setting, style, resolution)
yields better image outputs. (A), (B), (D) are vague and
ambiguous.
