Generative AI Leader Practice Question: Business Strategies for Generative AI Solutions
A telecommunications provider is preparing a business case for a generative AI virtual assistant that handles billing inquiries. Executives want the proposal to address financial viability, not just technical feasibility. Which two elements should the business case include to demonstrate responsible financial planning? (Choose two.)
⚠ Common exam trap
The trap here is filling a business case with technical artifacts such as model lists, prompt counts, or network diagrams instead of quantified unit costs and sensitivity-tested savings.
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
✓
A model of cost per resolved inquiry that includes inference, grounding, and human escalation.
A credible financial case for the billing assistant needs unit economics and a forward-looking benefit estimate. Cost per resolved inquiry ties inference, grounding, and human escalation into one comparable figure, while projected handling-cost reduction with sensitivity ranges shows how savings vary under different automation and volume assumptions. Together they let executives judge viability and set targets, whereas model inventories, prompt counts, and network diagrams do not quantify financial impact.
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 prompt variants tested during internal prototyping.
Why it's wrong here
Counting prompt variants describes experimentation effort, not financial viability. It offers no view of costs, savings, or return on investment, and a large number of variants could indicate churn rather than progress. For a business case focused on whether the billing assistant is financially sound, this metric does not help executives weigh investment against expected benefit or understand the assumptions behind projected results.
- ✓
A model of cost per resolved inquiry that includes inference, grounding, and human escalation.
Why this is correct
A cost-per-resolved-inquiry model captures the true unit economics of the assistant by combining inference charges, retrieval or grounding costs, and the expense of escalations to human agents. Because not every conversation is fully automated, ignoring escalations would overstate savings. This element gives executives a defensible basis for comparing the assistant against current billing support costs and for setting a target automation rate that keeps the program financially viable.
- ✗
A list of every foundation model available in Vertex AI Model Garden.
Why it's wrong here
Cataloging all available foundation models is an inventory exercise, not a financial planning element. It does not estimate costs, savings, or return, and it may distract from the few models actually suited to billing inquiries. A business case needs quantified economics and assumptions, so listing the Model Garden would add length without helping executives judge whether the virtual assistant is worth funding.
- ✗
A diagram of the virtual private cloud network topology used by the assistant.
Why it's wrong here
Network topology is an architecture artifact that matters for security and connectivity reviews, but it does not demonstrate financial planning. It cannot show cost per inquiry, expected savings, or the range of plausible outcomes. Including it in the financial section of the business case would not help executives assess viability, though it may belong in a separate technical appendix once funding decisions move forward.
- ✓
A projected reduction in average handling cost with assumptions and sensitivity ranges.
Why this is correct
Projecting the reduction in average handling cost, with explicit assumptions and sensitivity ranges, shows how the assistant affects the economics of billing support. Sensitivity ranges reveal how savings change if automation rates, escalation rates, or inquiry volumes differ from the base case, which is essential for responsible planning. This element lets executives see best, expected, and worst outcomes rather than a single optimistic number.
Visual reference
Go deeper
Related to this question
About these practice questions
This Generative AI Leader question is part of Courseiva's 1,008-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 Google Cloud exam blueprint
This Generative AI Leader practice question is part of Courseiva's free Google Cloud 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 Generative AI Leader exam.