Generative AI represents the most significant development in legal technology in the past century. We wrote this book to help our colleagues understand not just what these systems can do, but how they actually work and how to deploy them effectively in discovery and investigations.

When we published the third edition, we had 65 pages covering the fundamentals of Large Language Models and their basic applications in legal work. Within months, the technology had evolved dramatically. AI systems gained the ability to plan multi-step investigations, use tools strategically, and reason through complex problems autonomously. The emergence of agentic AI and advanced reasoning models demanded a fundamental expansion of the book.

This fourth edition spans nearly 200 pages and includes an entirely new section on agentic AI systems and reasoning models—covering single-agent and multi-agent architectures, tool use, iterative refinement, and the ethical implications of autonomous AI in legal practice. We've also expanded our practical applications section with detailed examples from actual discovery work, including complete workflows, cost analyses, and defensibility considerations.

The book is published through EDRM and distributed with a Creative Commons license because we believe this knowledge should be freely available to the legal profession. Whether you're a practitioner trying to understand these tools, a student preparing for practice, or simply someone interested in how AI is transforming legal work, we hope this book provides the foundation you need.