A risk-augmented financial model: AI-identified assumptions to stress test, three scenario narratives, and a risk register template populated from your business context.
How AI Improves Risk Thinking
AI doesn't do the math in your risk models — your financial model does that. What AI does is improve the thinking: identifying assumptions you've made without realizing it, surfacing risks you didn't think to include, and structuring your risk narrative clearly.
The most valuable AI use in risk analysis is the "devil's advocate" prompt. After you've built a model or scenario, you ask AI to challenge it systematically.
You are a skeptical CFO reviewing this financial projection.
Your job is to find weaknesses, not validate assumptions.
For each key assumption, tell me:
1. Why this assumption might be too optimistic
2. What would have to be true for it to be wrong
3. What data I could check to validate or invalidate it
Key assumptions in our FY2026 plan:
[list your 5-8 most important assumptions]
Be specific. Don't just say "market conditions could change."Scenario Building with AI
Every financial plan should have a base case, bull case, and bear case. The numbers come from your model — AI helps you build the narrative logic that makes each scenario internally consistent and credible.
Build three scenario narratives for our FY2026 plan.
For each scenario, write 1 paragraph explaining:
- What business and macro conditions lead to this outcome
- The key leading indicators that signal we're in this scenario
- What we should do differently if this scenario materializes
Base case: Revenue $18M, 15% growth, margin 22%
Bull case: Revenue $22M, 35% growth, margin 26%
Bear case: Revenue $14M, -7% decline, margin 16%
Business context:
[2-3 sentences about what your business does and key drivers]Build a Risk Register
A risk register is a documented list of identified risks with likelihood, impact, and mitigation plans. Creating one from scratch is slow. With AI, you can generate a first draft in under 10 minutes.
Generate a risk register for the following business.
For each risk include: description, likelihood (H/M/L),
impact (H/M/L), leading indicator, and mitigation action.
Business context:
[Describe your business: size, industry, revenue model,
key dependencies, top customers, key suppliers]
Cover these risk categories:
- Revenue/customer concentration
- Operational/execution
- Financial/liquidity
- Regulatory/compliance
- Technology/systems
- People/talent
Generate 3-4 specific risks per category.The AI-generated risk register is your starting point. Review each item: remove risks that don't apply, add risks specific to your context that AI missed, adjust likelihood and impact ratings based on your judgment. This calibration step is essential.
What You Learned Today
- How the 'devil's advocate' prompt surfaces weaknesses in financial plans
- How to generate internally consistent scenario narratives from financial projections
- How to build a risk register from business context in minutes
- Why reviewing and calibrating AI output is always required before using it professionally
Go Further on Your Own
- Run your current operating plan through the devil's advocate prompt. How many of the challenges AI identifies are ones you hadn't already considered?
- Ask AI to generate an 'early warning system' — for each scenario in your plan, what are the specific KPIs you'd monitor to know which scenario you're actually in?
- Use AI to build a sensitivity table narrative: you provide the variables, AI explains in plain English what the financial impact of each assumption change means in business terms
Nice work. Keep going.
Day 5 is ready when you are.
Continue to Day 5Want live instruction and hands-on projects? Join the AI bootcamp — 3 days, 5 cities.