AIF-C01 Fundamentals of Generative AI Practice Question
A financial services firm is evaluating foundation models for a customer-facing assistant. Compliance requires that prompts and completions never leave the firm's own AWS account boundary. Which characteristic of a foundation model deployment should the team evaluate FIRST against this constraint?
⚠ Common exam trap
The trap here is evaluating model quality metrics first, when a hard compliance constraint about where data is processed must drive the architecture decision.
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
✓
Whether the model is consumed as a fully managed service or deployed into infrastructure the firm controls.
Data-residency and account-boundary requirements are determined by where inference actually executes, so the deployment and consumption model must be assessed before capability metrics. Fully managed endpoints process requests on provider-managed infrastructure, while models deployed into the firm's own account keep prompt and completion data inside that boundary.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
The number of parameters in the model and its published benchmark scores.
Why it's wrong here
Parameter count and benchmark scores describe capability and quality, not where data is processed or stored. A very capable model can still be hosted outside the firm's account boundary, so these metrics do not address the residency requirement and should be evaluated only after the deployment-location constraint is satisfied.
- ✗
The maximum number of tokens the model can accept in a single request.
Why it's wrong here
Context length governs how much text can be supplied per call and affects cost and summarization strategies, but it says nothing about where data is processed. A model with a huge context window can still transmit prompts to a third-party endpoint, so context length is irrelevant to satisfying the account-boundary constraint.
- ✗
The licensing terms that govern commercial redistribution of the model weights.
Why it's wrong here
Licensing determines whether the firm may redistribute or embed the model in products it sells, which is a legal question distinct from data residency. A permissive license does not guarantee that inference stays inside the firm's account, so license review does not resolve the compliance constraint described in the scenario.
- ✓
Whether the model is consumed as a fully managed service or deployed into infrastructure the firm controls.
Why this is correct
The consumption model determines whether inference happens on shared managed endpoints or inside resources inside the firm's own account, which is the decisive factor for the boundary requirement. Fully managed APIs process prompts outside the customer account, whereas self-hosted deployments keep data within the firm's environment, so this must be settled before comparing model quality.
Go deeper
Related to this question
About these practice questions
One of 862 original AIF-C01 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 Amazon Web Services exam blueprint
This AIF-C01 practice question is part of Courseiva's free Amazon Web Services 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 AIF-C01 exam.