By examining the history, philosophy, and technological pathways of artificial intelligence, this book helps readers understand the origins of the boom in large language models. It is ideal for readers seeking to develop a comprehensive understanding of the field.
When discussing artificial intelligence today, it’s easy to focus entirely on model rankings, product updates, and prompt techniques. *A Brief History of Artificial Intelligence (3rd Edition)* offers a different approach: first, examine how the field has evolved to where it is today, and then assess what the latest developments have actually changed. It’s best used to gain historical context, rather than to chase after the next new tool.
The author, Nick—whose real name is Zhang Xiaodong—has a background in computer research and development, teaching, and management. This review covers the 3rd edition, published by People’s Posts and Telecommunications Press in April 2026, with ISBN 9787115694676. When selecting and purchasing this book, please verify the edition to avoid confusion with previous editions.
书中讨论人工智能早期探索、符号方法、神经网络、语言处理及Large Model，并延伸到智能与人的关系等问题。第3版增加了大模型相关内容，把当下的变化放回更长的技术史中理解。它的阅读重心是思想与方法的演进，不是逐步搭建应用的操作手册。
This book is ideal for product managers, business executives, and readers of technology content who want to understand the industry context, as well as students new to the field of AI who need to build a knowledge base. If you need immediate solutions to specific issues related to model deployment, programming errors, or workflow setup, this book should not be your sole reference; it’s best used in conjunction with engineering tutorials and official documentation.
I suggest creating a simple technical roadmap as you read: note what problems each method aims to solve, what conditions it depends on, and what limitations it faces. When you encounter a technological trend, don’t rush to find out “who won”; instead, compare changes in data, computing power, objectives, and evaluation methods. By reading history this way, you can turn these stories into tools for evaluating new technologies.
You can also categorize the concepts you encounter while reading into three groups: “already understood,” “needs foundational review,” and “worth further verification.” The first two groups are used to determine the order of study, while the third group requires returning to the original papers or research institution materials for further verification—never treating any single general history as the final authority on all perspectives.
If you want to start by developing a broad perspective, begin with this book; if you want to understand the mathematical foundations of algorithms, read Li Hang’s *Machine Learning Methods* next; and if you’re ready to translate model capabilities into products, supplement your reading with books on AI product management. These three reading paths address “where it comes from,” “how it works,” and “how to implement it,” respectively—there’s no need to expect a single book to cover everything.
People's Post and Telecommunications Publishing House Official Store: Book Information
Book cover source:Turing Community Official Reading List.
简体中文 简体中文 
English