CCAR-P Practice Question: Stakeholder Communication and Lifecycle Management
During a project milestone review, a stakeholder asks why the Anthropic model output occasionally varies. How do you explain this in a way that manages expectations?
⚠ Common exam trap
Candidates often suggest trying to eliminate output variance entirely by forcing zero temperature, failing to explain the fundamental probabilistic nature of LLMs to stakeholders.
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
✓
Explain that LLMs use probabilistic sampling, which creates variance.
Explaining the concept of 'temperature' and probabilistic nature of LLMs is essential to manage expectations. By framing variation as a feature that allows for creative and diverse responses, the architect aligns the stakeholder with the technology's strengths. This is crucial for avoiding frustration when stakeholders expect deterministic software-like behavior from generative AI models, which can lead to misaligned quality assessments.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
State it is a bug and that a patch will be deployed.
Why it's wrong here
Describing probabilistic output as a bug is fundamentally incorrect and sets unrealistic expectations. LLMs are designed to generate creative content, which naturally involves variance. Labeling standard behavior as a defect forces the team into an impossible position of trying to 'fix' a core characteristic of the model's architecture.
- ✓
Explain that LLMs use probabilistic sampling, which creates variance.
Why this is correct
Providing a technical explanation of how sampling parameters influence output helps stakeholders understand the underlying logic. It shifts the discussion from 'broken vs. working' to 'tuning for desired outcomes.' This transparency builds confidence and allows stakeholders to participate in defining the desired balance between creativity and consistency.
- ✗
Suggest setting the temperature to 0 for all future requests.
Why it's wrong here
Setting temperature to 0 eliminates variance but may also decrease the reasoning depth or creativity required for complex tasks. This is a potential optimization, not a universal solution. The architect must explain the trade-offs of this setting rather than prescribing it as a fix for the stakeholder's perceived problem.
- ✗
Ignore the concern as it is common knowledge in AI.
Why it's wrong here
Assuming stakeholders have the same domain knowledge as an AI architect is a common communication failure. Ignoring concerns prevents the stakeholder from understanding the tool, which often leads to poor utilization and dissatisfaction. Proactive education is a fundamental responsibility of the architect to ensure successful adoption.
About these practice questions
One of 262 original CCAR-P practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. 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 CCAR-P 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 CCAR-P exam.