CCAO-F Claude Model Fundamentals Practice Question
A team is building a document Q&A feature on Claude and wants to reduce hallucinations when the answer is not present in the supplied documents. Which TWO techniques are appropriate? (Choose two.)
⚠ Common exam trap
The trap here is treating sampling settings like temperature or max_tokens as hallucination controls, when grounding instructions and source citations are what actually constrain factual claims.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Instruct Claude in the system prompt to answer only from the provided documents and to say it does not know when the answer is absent.
Reducing hallucinations in document Q&A comes from constraining Claude to the supplied evidence and making that grounding verifiable. An instruction to answer only from the documents and to admit uncertainty sets the behavioral boundary, while requiring citations or quotes makes unsupported claims detectable. Sampling temperature and output-length settings do not improve factual grounding.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Instruct Claude in the system prompt to answer only from the provided documents and to say it does not know when the answer is absent.
Why this is correct
An explicit instruction to stay within the supplied documents and to admit uncertainty when the answer is missing gives Claude a clear behavioral rule to follow. This is a direct, low-cost way to reduce confident fabrication because the model is told what to do when evidence is lacking, rather than defaulting to a plausible-sounding guess.
- ✗
Remove the documents from the prompt and rely on Claude's pretrained knowledge to fill in missing details.
Why it's wrong here
Removing the source documents eliminates the grounding that the feature depends on, so Claude would answer from general training data that may be outdated or simply wrong for this corpus. This directly increases hallucination risk rather than reducing it, and it defeats the purpose of a document-based Q&A system.
- ✗
Increase max_tokens so Claude has more space to explain its reasoning before giving the final answer.
Why it's wrong here
max_tokens only limits output length; it does not improve factual grounding or reduce fabrication. A longer response can even give Claude more room to elaborate on an unsupported claim, making the hallucination more convincing. Length limits are unrelated to whether the model's statements are actually supported by the provided documents.
- ✓
Ask Claude to include a short citation or quote from the source document that supports each claim in its answer.
Why this is correct
Requiring a supporting quote forces Claude to ground each claim in retrieved text and makes unsupported statements visible to reviewers or automated checks. When no quote can be found, the absence itself signals that the answer is not in the documents, which helps catch hallucinations before they reach users.
- ✗
Raise the temperature so Claude explores a wider range of possible answers and is more likely to find the right one.
Why it's wrong here
Higher temperature increases randomness and diversity in sampling, which tends to increase the chance of invented or inconsistent content. It does not help Claude locate evidence that is absent from the documents. For factual Q&A where grounding matters, lower temperature is generally preferred to keep answers stable and tied to the source.
About these practice questions
This CCAO-F question is part of Courseiva's 259-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAO-F practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAO-F exam.