WGU D685 Objective Assessment Test Bank | Questions
with Answers & Explanations | Practical Applications of
Prompt Engineering Complete Exam Review | INSTANT PDF
DOWNLOAD
The WGU D685 Practical Applications of Prompt Engineering Objective
Assessment with multiple-choice questions featuring bold answers and detailed
explanations covering AI foundations, prompt design, advanced techniques,
evaluation, bias, ethics, and real-world applications.
Section 1: Foundations of AI & Why Prompt Engineering Matters (Questions 1–50)
1. A project manager asks why the team should invest time in prompt engineering rather
than simply typing natural-language questions into a large language model, and the most
accurate professional justification is that well-designed prompts do what?
A) Guarantee that the model will never produce incorrect or fabricated information
B) Improve the relevance, consistency, and controllability of outputs for specific professional
tasks
C) Eliminate the need for any human verification of AI-generated content
D) Convert a narrow AI system into an artificial general intelligence system
Answer: B) Improve the relevance, consistency, and controllability of outputs for
specific professional tasks
Explanation: Prompt engineering shapes model behavior toward task requirements and
substantially raises output quality and reliability, but it does not remove the need for
verification or transform model capabilities into AGI.
,2. A data analyst is comparing supervised and unsupervised machine learning approaches,
and the analyst correctly notes that supervised learning differs because it does what?
A) Requires no labeled examples and discovers clusters autonomously
B) Learns a mapping from inputs to known target labels using training examples
C) Relies exclusively on reward signals from an environment without any data
D) Can only be applied to image-generation models
Answer: B) Learns a mapping from inputs to known target labels using training
examples
Explanation: Supervised learning uses labeled input-output pairs to learn a predictive
mapping, whereas unsupervised learning finds structure without labels and reinforcement
learning optimizes via rewards.
3. A student describes an AI system that plays chess at a superhuman level but cannot
transfer that skill to any other domain, and this system is best classified as which type of AI?
A) Artificial general intelligence capable of any human intellectual task
B) A fully conscious agent with human-like understanding
C) Narrow (weak) AI specialized for a single or limited set of tasks
D) An unsupervised clustering algorithm
Answer: C) Narrow (weak) AI specialized for a single or limited set of tasks
Explanation: Narrow AI excels at specific tasks such as game playing or image
classification, while general AI would match or exceed human flexibility across domains and
remains hypothetical.
4. A research team notices that a generative model sometimes invents plausible-sounding
citations that do not actually exist, and this phenomenon is most accurately known as what?
A) Tokenization of the input vocabulary
B) Successful few-shot learning
C) Hallucination, in which the model generates content not grounded in its training data or
,provided context
D) Deterministic retrieval from a knowledge base
Answer: C) Hallucination, in which the model generates content not grounded in its
training data or provided context
Explanation: Hallucinations are fluent but factually unsupported outputs, and recognizing
them is essential for evaluation and for designing prompts that reduce their occurrence.
5. An organization wants to understand the role of context when interacting with large
language models, and context is critical primarily because these models do what?
A) Condition their next-token predictions on the tokens supplied in the current prompt
window
B) Possess permanent long-term memory of every prior conversation across all users
C) Ignore all surrounding text and respond only to single keywords
D) Automatically expand every query into a multi-document research report
Answer: A) Condition their next-token predictions on the tokens supplied in the
current prompt window
Explanation: LLMs generate outputs by conditioning on the context provided in the prompt;
they do not have persistent memory of prior conversations unless that information is
included in the prompt.
6. A hospital wants to deploy an AI system to summarize patient discharge notes, and which
characteristic of generative AI makes this task well-suited for the technology?
A) The ability to perform independent clinical reasoning beyond its training data
B) The ability to process and condense unstructured natural-language text
C) The ability to guarantee complete factual accuracy without human oversight
D) The ability to replace all clinical judgment with deterministic rules
Answer: B) The ability to process and condense unstructured natural-language text
Explanation: Generative AI is well-suited for summarizing unstructured text, but it cannot
, perform independent clinical reasoning beyond its training data and still requires human
verification.
7. A financial advisor wants to implement AI to improve business processes, and which task
is an AI system best suited for in this context?
A) Making final investment decisions without human oversight
B) Processing large volumes of data in real time to identify patterns
C) Predicting stock prices with guaranteed accuracy
D) Performing independent reasoning beyond its training data
Answer: B) Processing large volumes of data in real time to identify patterns
Explanation: AI is well-suited for processing large datasets and identifying patterns in real
time, but it cannot guarantee accuracy or replace human judgment in high-stakes decisions.
8. A team is selecting between a large language model and a traditional rule-based system
for a customer support chatbot, and which advantage of LLMs makes them preferable for
handling varied customer inquiries?
A) LLMs follow deterministic rules that never produce unexpected outputs
B) LLMs can generalize to novel phrasings and intents without explicit reprogramming
C) LLMs require no training data and can operate without any context
D) LLMs are incapable of generating incorrect information
Answer: B) LLMs can generalize to novel phrasings and intents without explicit
reprogramming
Explanation: LLMs can handle varied and novel language inputs due to their pretraining on
large text corpora, whereas rule-based systems require explicit programming for each
scenario.
9. A product manager asks whether prompt engineering is necessary if the underlying
model is already highly capable, and the best answer is that prompt engineering remains
necessary because it does what?
with Answers & Explanations | Practical Applications of
Prompt Engineering Complete Exam Review | INSTANT PDF
DOWNLOAD
The WGU D685 Practical Applications of Prompt Engineering Objective
Assessment with multiple-choice questions featuring bold answers and detailed
explanations covering AI foundations, prompt design, advanced techniques,
evaluation, bias, ethics, and real-world applications.
Section 1: Foundations of AI & Why Prompt Engineering Matters (Questions 1–50)
1. A project manager asks why the team should invest time in prompt engineering rather
than simply typing natural-language questions into a large language model, and the most
accurate professional justification is that well-designed prompts do what?
A) Guarantee that the model will never produce incorrect or fabricated information
B) Improve the relevance, consistency, and controllability of outputs for specific professional
tasks
C) Eliminate the need for any human verification of AI-generated content
D) Convert a narrow AI system into an artificial general intelligence system
Answer: B) Improve the relevance, consistency, and controllability of outputs for
specific professional tasks
Explanation: Prompt engineering shapes model behavior toward task requirements and
substantially raises output quality and reliability, but it does not remove the need for
verification or transform model capabilities into AGI.
,2. A data analyst is comparing supervised and unsupervised machine learning approaches,
and the analyst correctly notes that supervised learning differs because it does what?
A) Requires no labeled examples and discovers clusters autonomously
B) Learns a mapping from inputs to known target labels using training examples
C) Relies exclusively on reward signals from an environment without any data
D) Can only be applied to image-generation models
Answer: B) Learns a mapping from inputs to known target labels using training
examples
Explanation: Supervised learning uses labeled input-output pairs to learn a predictive
mapping, whereas unsupervised learning finds structure without labels and reinforcement
learning optimizes via rewards.
3. A student describes an AI system that plays chess at a superhuman level but cannot
transfer that skill to any other domain, and this system is best classified as which type of AI?
A) Artificial general intelligence capable of any human intellectual task
B) A fully conscious agent with human-like understanding
C) Narrow (weak) AI specialized for a single or limited set of tasks
D) An unsupervised clustering algorithm
Answer: C) Narrow (weak) AI specialized for a single or limited set of tasks
Explanation: Narrow AI excels at specific tasks such as game playing or image
classification, while general AI would match or exceed human flexibility across domains and
remains hypothetical.
4. A research team notices that a generative model sometimes invents plausible-sounding
citations that do not actually exist, and this phenomenon is most accurately known as what?
A) Tokenization of the input vocabulary
B) Successful few-shot learning
C) Hallucination, in which the model generates content not grounded in its training data or
,provided context
D) Deterministic retrieval from a knowledge base
Answer: C) Hallucination, in which the model generates content not grounded in its
training data or provided context
Explanation: Hallucinations are fluent but factually unsupported outputs, and recognizing
them is essential for evaluation and for designing prompts that reduce their occurrence.
5. An organization wants to understand the role of context when interacting with large
language models, and context is critical primarily because these models do what?
A) Condition their next-token predictions on the tokens supplied in the current prompt
window
B) Possess permanent long-term memory of every prior conversation across all users
C) Ignore all surrounding text and respond only to single keywords
D) Automatically expand every query into a multi-document research report
Answer: A) Condition their next-token predictions on the tokens supplied in the
current prompt window
Explanation: LLMs generate outputs by conditioning on the context provided in the prompt;
they do not have persistent memory of prior conversations unless that information is
included in the prompt.
6. A hospital wants to deploy an AI system to summarize patient discharge notes, and which
characteristic of generative AI makes this task well-suited for the technology?
A) The ability to perform independent clinical reasoning beyond its training data
B) The ability to process and condense unstructured natural-language text
C) The ability to guarantee complete factual accuracy without human oversight
D) The ability to replace all clinical judgment with deterministic rules
Answer: B) The ability to process and condense unstructured natural-language text
Explanation: Generative AI is well-suited for summarizing unstructured text, but it cannot
, perform independent clinical reasoning beyond its training data and still requires human
verification.
7. A financial advisor wants to implement AI to improve business processes, and which task
is an AI system best suited for in this context?
A) Making final investment decisions without human oversight
B) Processing large volumes of data in real time to identify patterns
C) Predicting stock prices with guaranteed accuracy
D) Performing independent reasoning beyond its training data
Answer: B) Processing large volumes of data in real time to identify patterns
Explanation: AI is well-suited for processing large datasets and identifying patterns in real
time, but it cannot guarantee accuracy or replace human judgment in high-stakes decisions.
8. A team is selecting between a large language model and a traditional rule-based system
for a customer support chatbot, and which advantage of LLMs makes them preferable for
handling varied customer inquiries?
A) LLMs follow deterministic rules that never produce unexpected outputs
B) LLMs can generalize to novel phrasings and intents without explicit reprogramming
C) LLMs require no training data and can operate without any context
D) LLMs are incapable of generating incorrect information
Answer: B) LLMs can generalize to novel phrasings and intents without explicit
reprogramming
Explanation: LLMs can handle varied and novel language inputs due to their pretraining on
large text corpora, whereas rule-based systems require explicit programming for each
scenario.
9. A product manager asks whether prompt engineering is necessary if the underlying
model is already highly capable, and the best answer is that prompt engineering remains
necessary because it does what?
