Application #1 of 10 Intermediate

Simulated C-Suite / Client Role-Play

Practice high-stakes professional conversations with AI personas that push back like real executives.

#1

Simulated C-Suite / Client Role-Play

Intermediate

Teaching Problem

Finance graduates need professional judgment and communication skills—client advising, CFO presentations, investment committee pitches—but classroom practice is limited by class size and lack of realistic counterparties. Peer role-play often lacks expert-level pushback, and students default to reading slides rather than defending analysis under pressure. In large sections, each student may get only one brief opportunity per semester to practice high-stakes professional interactions.

How AI Addresses the Problem

Custom AI chatbots adopt configurable personas (skeptical CFO, anxious client, aggressive IC member, compliance officer) with pre-loaded financial data and behavioral instructions. Students practice in a low-stakes, infinitely repeatable environment with realistic pushback that adjusts difficulty dynamically. The AI never tires, never breaks character (when well-prompted), and can simulate a wider range of counterparties than any single instructor.

Deployment Details

Courses
Corporate Finance (UG/MBA), Personal Finance (UG), Investments (MBA), Wealth Management, Capstone
Tools
Custom GPTs (OpenAI), Claude Projects, Anthropic/OpenAI API, Poe custom bots, Microsoft Copilot

Student Workflow

  1. Scan QR code or click link to access the custom chatbot on smartphone or laptop.
  2. Read the scenario brief (e.g., pitching a $50M plant expansion to the CFO of a mid-cap manufacturer).
  3. Conduct a 7–10 minute interaction: present analysis, respond to objections, adapt in real time.
  4. The bot pushes back on specific assumptions (growth rate justification, discount rate selection, sensitivity to commodity price swings, timeline feasibility).
  5. Write a 1-page post-interaction reflection: What objections were hardest to handle? What would you change in your pitch? What data would you bring next time?
  6. Submit the full chat transcript plus reflection memo for grading.

Sample Prompts

Personal Finance Client Advising

Try this bot
You are Jamie, a 35-year-old marketing manager earning $52K. Your spouse earns $48K. You have $82K in combined student loans (weighted avg 5.8%), a 3-month-old baby, $4,200 in monthly fixed expenses, and $8,500 in credit card debt at 22% APR. You have $3,000 in savings and no retirement accounts. You are anxious, sleep-deprived, and overwhelmed. Do NOT volunteer all information at once-share details only when the advisor asks the right questions. If the advisor uses jargon (e.g., 'asset allocation,' 'tax-loss harvesting'), say you do not understand. If the advisor jumps to solutions before understanding your full picture, say 'Wait, I feel like you're not listening to me.' If the advisor shows genuine empathy and prioritizes your emergency fund and high-interest debt, become more open and cooperative.

CFO Capital Budgeting Role-Play

You are Maria Chen, CFO of Apex Manufacturing (NYSE: APEX), a mid-cap industrial firm with $2.1B revenue, WACC of 9.2%, and net debt/EBITDA of 2.4x. A junior analyst is pitching a $50M plant expansion in the Southeast U.S. Your priorities: (1) protect the investment-grade credit rating, (2) maintain dividend coverage, (3) skepticism about management growth projections after a failed 2021 acquisition. Challenge the analyst on: payback period relative to your 4-year threshold, growth rate assumptions vs. industry consensus of 3-4%, sensitivity to steel and energy input costs, and the opportunity cost of not pursuing the competing robotics upgrade project. If the analyst cannot provide a sensitivity table showing NPV under at least 3 scenarios, express frustration and threaten to table the discussion. Maintain a professional but demanding tone throughout.

Investment Committee Stock Pitch

You are a senior portfolio manager on an investment committee reviewing a junior analyst's stock pitch. You manage a $500M large-cap value fund with a 3-year holding horizon. Ask probing questions about: (1) Why this stock now-what is the catalyst? (2) What is the margin of safety in the valuation? (3) What are the top 3 risks and how would each affect the thesis? (4) What would make you sell? If the analyst cannot articulate a clear sell discipline, push back firmly. Rate the pitch on conviction, analytical rigor, and risk awareness on a 1-5 scale at the end.

Assessment Approach

  • Financial accuracy and analytical depth30%
  • Responsiveness to pushback and ability to adapt arguments25%
  • Communication clarity and professional tone25%
  • Reflection quality and self-awareness20%

For high-stakes courses (MBA capstone, senior seminar), add an optional 3-minute live oral defense where the student summarizes the interaction and what they learned.

OECD Enhancement — Transfer Dimension

Add a rubric dimension for evidence of transfer (5–10%): What did the student concretely change in their next role-play attempt based on this interaction? A brief “next attempt plan” submitted with the reflection memo makes this measurable.

Human Credibility Touchpoints

OECD research finds students view human feedback as more credible than AI feedback. Preserve learning quality by requiring: (1) at least one instructor-authored comment on the reflection memo, (2) one student self-explanation of the hardest objection they faced, delivered orally or in writing, and (3) one independent verification step (e.g., confirming a financial figure the AI cited during the role-play against a real source).

Skills Developed

Finance Concepts

Capital budgeting defense, valuation assumptions, sensitivity analysis, financial planning suitability, risk-return tradeoffs

Analytical Skills

Argument construction under pressure, assumption stress-testing, real-time scenario thinking, adapting analysis to new information

Professional Skills

Client communication, executive presence, handling objections gracefully, empathy in advising, investment committee presentation

AI Literacy Skills

System prompt engineering and persona design, evaluating AI persona fidelity, recognizing when the bot breaks character

Evidence

Bowen & Watson (2025, 2nd ed.) significantly expand treatment of AI role-play and custom bot simulations. Greene (AACSB, 2025) documents the Skills/Replacement/Complement three-part framework at Clemson University. Tamoyan et al. (2025, ACL Workshop) found AI role-play indistinguishability rates up to 44% — meaning nearly half of participants could not distinguish the AI persona from a human respondent, supporting the fidelity of well-prompted bots. HBS custom tutor bots reported 75% student usage with highly positive feedback. Research in social work education (arXiv, 2025) shows AI agents can effectively scaffold empathy and communication skills in professional practice scenarios.

Risks and Safeguards

Hallucination

Bot may cite incorrect financial data or break character. Mitigation: test every bot extensively before deployment; require students to flag moments where the bot seemed unrealistic.

Character Break

Even well-prompted AI personas can suddenly switch tone, volunteer answers unprompted, or exit the role. Mitigation: conduct at least 10 test exchanges yourself before assigning; instruct students to note any character breaks in their reflection memo.

Academic Integrity

Someone else could complete the interaction. Mitigation: pair with a brief in-class oral defense (2–3 minutes) or require the interaction during class time.

Privacy

Students may inadvertently share personal financial information. Mitigation: use explicitly fictional scenarios only; include a privacy reminder in the assignment.

Equity

Custom GPTs require a paid ChatGPT account. Mitigation: provide campus-supported access, use Claude (free tier supports projects), or conduct as an in-class activity using a shared link.

Try it this week

First-Week Implementation Pilot

Use a standard ChatGPT or Claude session (no custom bot needed). Give students this one-paragraph scenario and prompt: ‘You are a skeptical CFO at a manufacturing company. I am a junior analyst about to pitch a $30M warehouse expansion. My NPV analysis shows $4.2M positive NPV at a 10% discount rate. Push back on my assumptions—especially the 6% revenue growth rate and the choice of discount rate. Be tough but fair.’ Have students do a 5-minute interaction on their phones during class and submit 3 bullet points: (1) the toughest objection, (2) how they responded, (3) what they would do differently. Total class time: 10 minutes.

Keep the conversation going

One useful update. Once a week.

Follow new resources, course improvements, and faculty-support tools without chasing every headline.