AI for Investment Banking: Leverage ChatGPT & DeepSeek

In this lecture, you'll learn how to use ChatGPT to draft professional emails, create IPO roadmaps, brainstorm team bonding ideas, and organize schedules. Specifically, you'll see how to write year-end emails, develop IPO timelines, suggest icebreakers, and generate staffing schedules for training programs.

In this lecture, you'll learn how to use ChatGPT to analyze call transcripts. Using Target Corp 3Q2024 earnings call as an example, you'll see how to extract and organize equity research analysts' questions and identify their key concerns.

In this lecture, you’ll learn how to use ChatGPT to personalize client pitch decks. You'll see how to tailor a generic M&A pitch to Blackstone Infrastructure by incorporating client-specific information, refine messaging, and generate speaker notes to ensure a more impactful presentation.

In this lecture, you'll learn how to use Microsoft Copilot to analyze PDF documents such as earnings reports. We will specifically look at Baker Hughes quarterly earnings release and summarize their key financials and generate key takeaways. Additionally, we'll explore how Copilot integrates with Microsoft 365 tools for finance professionals, ensuring security and seamless workflow optimization.

In this lecture, you'll learn how to use DeepSeek chat to build a Discounted Cash Flow (DCF) model from a PDF earnings release. We'll work on Unilever's 4Q2024 earnings release and explore how to structure prompts for financial modeling, apply Chain of Thought (COT) prompting for transparency and verify assumptions using Perplexity AI.

In this lecture, we test DeepSeek chat’s ability to build a Discounted Cash Flow (DCF) model using Excel financial data as a source. While AI can extract insights from Unilever's Excel file, this session highlights its limitations in making correct assumptions. You'll learn why Excel remains the gold standard for financial modelling.

This lecture covers AI prompting techniques, including Chain-of-Thought (COT) for step-by-step reasoning transparency, Tree-of-Thought (TOT) for exploring multiple solutions, and Role-Based Prompting for expert-like responses.

This lecture explores how AI can act as a personal agent to simulate investor Q&A sessions. By uploading Citigroup’s Fixed Income Investor Presentation to DeepSeek, you can prompt the AI to play the role of a PIMCO Fixed Income Fund Manager, generating realistic and challenging questions, plus reviewing feedback at the end.

In this lecture, we categorize AI use cases for finance into four key areas:

1) Data Extraction from Files,

2) Creative Writing for Finance,

3) Financial Analysis, and

4) Financial Data Sourcing from the Web.

We evaluate how effectively AI handles each category and rank individual use cases we went through in this course based on the quality and usefulness of AI-generated responses, scoring them from 1 to 5. Finally, we identify the best AI tools for each category, highlighting where AI delivers the most value and where human expertise remains essential. This last lecture acts as a recap of this course.