For executives, finance partners and transformation leads who must build, challenge or approve an AI investment case and leave able to defend the numbers in front of a CFO.
Level
Executive
Study time
8 hours
Lessons
14 in 5 modules
Exam
15 questions
About this course
Most AI business cases fail in one of three places: they count time saved as cash saved, they leave out the costs that arrive after the pilot, or they never measure what happened after launch. This course gives executives a disciplined method for each step. You will build a value driver tree that links an AI use case to revenue, cost, risk or capital; baseline the current process before anything is built; and cost the solution fully, from per-seat assistants and usage-based API pricing to committed capacity, integration, evaluation, human review and change management.
The second half covers uncertainty and proof. You will model scenarios and sensitivities (NPV, payback and the breakeven adoption rate), risk-adjust the case for error rates and slow adoption, design pilots with control groups that produce real evidence, and instrument unit economics per task or document. The final module covers funding models, presenting to a CFO or investment committee, and benefits realization reviews that decide whether to scale, fix or stop. All examples are labeled hypothetical scenarios; vendor pricing structures are described as of October 2026 and are re-checked at each quarterly review. Expect about 2 hours of reading and about 6 hours of hands-on exercises, each of which can be done with a short Python script or a spreadsheet.
What you will be able to do
Build a value driver tree that links an AI use case to a measurable financial or risk outcome
Establish a baseline for the current process before an AI solution is funded
Estimate the total cost of ownership of an AI solution across seat, usage-based and committed-capacity pricing, including hidden run costs
Distinguish released time from realized cash benefit and model scenarios, sensitivities and breakeven adoption
Design a pilot with a control group that produces decision-grade evidence
Choose a funding model and present an AI case that withstands CFO and investment committee challenge
Run a benefits realization review that leads to a scale, fix or stop decision
Syllabus
5 modules · 14 lessons · about 8 hours
1
Module 1: Framing the Investment Decision
Turning a use case into an investment thesis with an owner, a value driver and a baseline.3 lessons
From Use Case to Investment Thesis: Problem, Owner and Value Driver
Value Driver Trees: Linking AI to Revenue, Cost, Risk and Capital
Baselining Before You Build: Measuring the Process You Have Today
2
Module 2: Costing AI Honestly
The full cost of an AI solution, including the costs that appear only after the pilot.3 lessons
Total Cost of Ownership for Generative AI: Tokens, Seats, Integration and Run Costs
The Hidden Costs: Evaluation, Human Review, Data Work and Change Management
Pricing Models Compared: Per-Seat Assistants, Usage-Based APIs and Committed Capacity
3
Module 3: Modeling Benefits Under Uncertainty
Separating released time from cash, and stress-testing the case before it is approved.3 lessons
Productivity Gains Are Not Savings: Released Time Versus Realized Cash
Scenarios and Sensitivities: NPV, Payback and Breakeven Adoption
Risk-Adjusting the Case: Error Rates, Adoption Curves and Kill Thresholds
4
Module 4: Proving Value With Evidence
Pilots and instrumentation that produce numbers a CFO will accept.2 lessons
Designing Pilots That Produce Evidence: Control Groups and Staged Rollouts
Instrumenting Unit Economics: Cost and Value per Task, Ticket or Document
5
Module 5: Funding, Approval and Benefits Realization
How AI investments are funded, approved and held to account after go-live.3 lessons
Funding Models: Central AI Fund, Business-Unit Chargeback and Tranche Funding
Presenting the Case to the CFO and Investment Committee
Benefits Realization Reviews: Scale, Fix or Stop
Exam and certificate
The exam unlocks once you have completed every lesson. It draws 15 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 30 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.
Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.
The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by Microsoft, Google, OpenAI, Anthropic, or Amazon Web Services. Product and company names are trademarks of their respective owners.
Follow one company's AI decisions through a winter: three vendor pitches, a test on its own calls, a first pilot that worked, a chatbot that made promises nobody approved, and the plan it took to the board. Free, practical, and built around decisions you make yourself.
Work through one company's AI year as its executive sponsor. Test vendor claims on your own data, rank a portfolio by defended value, make the sourcing and GPU calls, rescue a stalled rollout, govern risk and law, and prove ROI to a skeptical CFO, then play seven Decision Labs and build the board pack.
A practical briefing on what boards owe on AI and how to discharge it. Covers oversight duties, committee ownership, NIST AI RMF and ISO/IEC 42001 at board level, the EU AI Act after the 2026 Digital Omnibus, US federal, state and sector rules, disclosure, risk appetite, inventory and dashboards, third-party AI, incident escalation and an annual oversight calendar.