Generative AI and Finance

This one-day course shows investment professionals how to use Generative AI—ChatGPT, Copilot, and similar tools—to think faster, write clearer, and make better decisions. Instead of building technical systems, participants learn practical prompting techniques to turn news, data snippets, and notes into hypotheses, scenarios, IC memos, and client-ready narratives. The focus is real-world workflow acceleration, bias-aware thinking, and compliance-safe use of GenAI.

• Portfolio Managers
• Traders
• Investment Analysts
• Financial Executives and Team Leads
• Risk and IC Committee members

• Generate clear hypotheses, scenarios, and investment narratives in minutes.
• Produce high-quality IC memos with risks, triggers, and exit criteria.
• Use GenAI to summarise, synthesise, and challenge investment views
• Improve factuality and reduce hallucinations with simple QA techniques
• Streamline daily work using Copilot in Outlook, Excel, PowerPoint, and Teams
• Build a repeatable decision-making rhythm using GenAI as a structured assistant.

• Basic ability to read Python (no coding required in class)
• Working understanding of financial markets and investment concepts
• Access to ChatGPT or Microsoft Copilot; Excel and PowerPoint recommended

Welcome, Goals & Guardrails

• What “good” looks like for PM desks (speed, clarity, decision hygiene)
• Safety & compliance quick start: confidentiality, factuality, “don’t-know” defaults
• Deliverable: 1-page “compliance-safe prompting” card.

Prompting for Insight (Theory + Micro-Labs)

• The executive prompt stack: Role → Context → Task → Constraints → Checks
• Five reusable patterns:
  • Brief → Hypotheses → Tests
  • Devil’s Advocate (steel-man the opposite)
  • Two-Column Synthesis (bull vs. bear with evidence)
  • Decision Memo (risks, triggers, kill-switch)
  • Meeting Prep (5 smart questions + likely answers)
  • Micro-Labs (in ChatGPT/Copilot): turn a news digest into 3 testable hypotheses; attack your own stance; produce a 1-page IC draft
  • Output: Copy/paste prompt library

Portfolio Narrative & IC Pack Acceleration (Hands-on)

• Anchoring on your numbers without tooling: paste tables; “read-this-first”; cite-what-you-used
• Style control for IC, Risk, Client variants
• Lab: From a small “PM Pack” (CSV + excerpts), create:
  • A 1-page IC memo (Setup, Thesis, 3 Risks, 2 Triggers, Base/Alt)
  • Two rewrites (Risk Chair, Client)
  • “What would change my mind?” checklist
  • Output: Three ready-to-edit documents

Scenario & Risk Thinking with GenAI (Hands-on)

• From macro narrative to trade implications: channels, signs, timing, exits
• Bias traps (confirmation, narrative fallacy) + prompts that counteract them
• Lab: Generate 2–3 scenario trees, map to expressions (futures/options/packages), write a 6-line playbook (enter/monitor/exit)
• Output: Reusable scenario-to-playbook template

Speed-Runs in Copilot Apps (No-Code)

• Outlook/Teams: summarise threads to actions, extract asks & deadlines
• PowerPoint: turn IC one-pager into 5 slides; iterate speaker notes
• Excel: natural-language transforms on small tables (filter, label, quick stats)
• Output: Three “from messy to usable” drills

Quality, Factuality & Anti-Hallucination Tactics

• The 4 Cs: Context, Citations, Counters, Calibration (confidence + next checks)
• Red-teaming your own outputs; refusal/“insufficient info” clauses
• Lightweight QA rubric (no code)
• Lab: Run three QA passes on your IC memo (factual claims, missing data, style/compliance)
• Output: QA rubric + three prompt macros

Operating Rhythms & Team Playbooks

• Daily 30-minute rhythm (AM hypotheses, midday scenario check, EOD memo)
• Weekly IC cadence: questions to always ask (and ones to avoid)
• Team setup: shared prompt libraries, versioned memos, decision logs
• Exercise: Personalise a desk rhythm (who/when/artifacts/storage)
• Output: 1-page operating playbook

Wrap-Up & First-Week Challenge

• Ship one improved IC memo + one scenario playbook using today’s templates
• Q&A