Fundamentals of AI in Equity Research: Building Analyst Workflows for the Future

Part One: Understanding AI-Augmented Research Workflows

Session 1: The Modern Equity Research Process

• AI in equity covers the end-to-end research workflow — from idea discovery and validation to modelling, valuation, and final report communication

• Understand how AI integrates into each stage of this process: data gathering, summarisation, analysis, visualisation, and drafting

• Learn the specific strengths of major AI tools (GPT-5, Claude, Gemini, Perplexity, Copilot) and how they complement traditional analyst work

• Distinguish clearly between tasks best handled by automation and those that require human insight, judgement, and market context

• Discuss real-world examples of blended human-AI workflows in professional research teams

Session 2: Fundamentals of Prompting in Research

• Master the core principles of effective prompting — clarity, context, reasoning, and refinement

• Apply the “Plan – Prompt – Polish” framework to translate traditional research steps into efficient AI-assisted workflows

• Learn how to transform a typical analyst assignment (e.g., an earnings call review or thematic summary) into an AI-supported task flow

• Identify prompt patterns, structures, and modifiers that consistently improve depth, factual precision, and style

• Discuss common pitfalls of shallow prompting and how to guide AI models toward verifiable, professional-grade output

• Mini-Case 1: Use AI to extract and synthesise key strategic themes from a company’s annual report, comparing results across multiple models for accuracy and tone

Part Two: Building the AI-Enhanced Research Note

Session 3: From Idea to Insight – Workflow Design

• Fundamentals of AI in equity research deconstructs complex research tasks into modular, prompt-driven stages that mirror the real analyst workflow

• Learn to design iterative AI workflows for company analysis, peer benchmarking, and basic valuation

• Compare “Single-Shot” versus “Multi-Shot” prompting approaches to achieve progressively higher quality and analytical precision

• Document the evolution of prompts to build traceability and show analytical reasoning

• Explore how to use AI feedback loops to refine assumptions and strengthen conclusions

• Mini-Case 2: Iterative prompting exercise – build, test, and refine a concise one-page company note using structured AI feedback

Session 4: Structuring and Drafting the Report

• Review the standard architecture of an equity research note — Executive Summary, Investment Thesis, Financials, Valuation, and Risks

• Practise using AI to draft concise, verifiable, and well-reasoned sections that combine data and narrative

• Integrate qualitative insights (strategy, management tone) with quantitative metrics (margins, growth, multiples) for balanced analysis

• Employ visual prompting for generating charts, peer comparisons, and valuation commentary that communicate findings effectively

• Understand how iterative refinement improves flow, tone, and analytical coherence

• Mini-Case 3: Prompt-chaining exercise – generate and refine valuation commentary, peer tables, and catalyst summaries

Part Three: Best Practice, Verification, and Future Readiness

Session 5: Risk, Ethics, and Verification

• Recognise and manage common AI pitfalls — hallucinations, bias, and data confidentiality breaches

• Apply cross-model validation to check accuracy and triangulate findings between GPT-5, Claude, and Gemini

• Build verification checklists for numerical integrity, citation traceability, and transparent disclosure

• Discuss compliance, governance, and responsible-use frameworks shaping AI deployment in investment research

Session 6: Embedding AI in Research Teams

• Learn how to create and maintain  reusable prompt templates and workflow libraries  for institutional use

• Explore strategies for  scaling AI adoption  while preserving analyst judgement, oversight, and accountability

• Align AI tools with internal data systems, compliance requirements, and quality-control standards

• Design review frameworks to measure improvements in speed, consistency, and analytical depth

• Mini-Case 4: Audit and refine a flawed AI-generated report — identify weaknesses, verify data, and rebuild it using structured prompting best practices

Prerequisites and Tools

Participants should have basic familiarity with Excel, PowerPoint, and financial statement interpretation for AI in equity research training.

All exercises use open-source or readily accessible tools; paid versions are optional for deeper functionality.

AI Tools:

Pro Tier (Preferred for full capability and workflow integration)

• GPT-5 (ChatGPT Pro): Core tool for modelling, valuation, drafting notes, and summarising complex data

• Gemini 1.5 Pro: Quick fact-finding, macro trend analysis, and Google Sheets integration

• Perplexity Pro: Real-time search, consensus validation, and alt-data triangulation

• Notion AI: Structured prompt management, note iteration, and workflow documentation

Free Tier (Suitable for foundational use

• ChatGPT Free / Claude Instant / Gemini Basic: Light summarisation, drafting, and brainstorming

• Claude 4: Deep analysis of filings, transcripts, and risk disclosures (accessible under free usage limits)

• Microsoft Copilot: Excel automation, comp-table generation, and slide creation within Office 365 environments

• Perplexity Free: Rapid fact-checking and public-data cross-verification

• Google Sheets or Excel Online: For simple ratio and trend analysis

Financial / Data Platforms

(Used for input workbooks and sourced financial information to be worked on via case studies and demonstrations; participants will have some access via materials provided during the session.)

• TIKR Pro: Company fundamentals, estimates, and peer benchmarking

• Koyfin Pro: Charting, screening, macro dashboards, and time-series analysis

• Seeking Alpha Pro: Earnings transcripts, management commentary, sentiment and crowd analysis

As this is a foundational course, we will use the above tools to iterate, prompt, and build the key workstreams that support idea generation, ongoing idea monitoring, and core equity research processes. However, since we are not using an integrated or out-of-the-box AI software suite, the course will not cover fully automated financial model builds. The focus is on mastering the workflows, prompting techniques, and analytical reasoning that underpin high-quality professional research using AI assistance.

Training Objectives

• Produce an AI-augmented equity research note, combining both qualitative and quantitative analysis of a chosen company while following professional research standards.

• Use AI tools to accelerate data gathering, summarisation, and drafting while maintaining data integrity, transparency, and analytical rigour.

• Deliver well-structured sections covering company overview, industry context, strategy, financials, valuation, risks, ESG considerations, catalysts, and conclusion.

• Design efficient prompting workflows that reflect how analysts work in practice, from sourcing and analysis through drafting and review.

• Develop hands-on skills in multi-step prompting and report iteration, learning how to improve quality through context, structure, and feedback loops.

• Understand the ethical and regulatory aspects of using AI in research, including disclosure, bias control, and data privacy.

• Build and present a final AI-powered research report applying all techniques learned to analyse a listed company and generate clear investment insights.

Training Course Summary

This AI in Equity Research course teaches finance professionals how to use AI and machine learning in equity research and portfolio management. It covers data extraction, sentiment analysis, stock screening, risk assessment, and portfolio optimisation using AI.

Sessions include hands-on exercises with open-source AI tools and discuss ethical considerations and future trends in AI-driven finance.