What do you think?
Rate this book
How AI Works: From Sorcery to Science
AI isn’t magic. How AI Works demystifies the explosion of artificial intelligence by explaining—without a single mathematical equation—what happened, when it happened, why it happened, how it happened, and what AI is actually doing "under the hood."
Artificial intelligence is everywhere—from self-driving cars, to image generation from text, to the unexpected power of language systems like ChatGPT—yet few people seem to know how it all really works. How AI Works unravels the mysteries of artificial intelligence, without the complex math and unnecessary jargon.
You’ll
The relationship between artificial intelligence, machine learning, and deep learningThe history behind AI and why the artificial intelligence revolution is happening nowHow decades of work in symbolic AI failed and opened the door for the emergence of neural networksWhat neural networks are, how they are trained, and why all the wonder of modern AI boils down to a simple, repeated unit that knows how to multiply input numbers to produce an output number.The implications of large language models, like ChatGPT and Bard, on our society—nothing will be the same again
AI isn’t magic. If you’ve ever wondered how it works, what it can do, or why there’s so much hype, How AI Works will teach you everything you want to know.
Artificial intelligence is everywhere—from self-driving cars, to image generation from text, to the unexpected power of language systems like ChatGPT—yet few people seem to know how it all really works. How AI Works unravels the mysteries of artificial intelligence, without the complex math and unnecessary jargon.
You’ll
The relationship between artificial intelligence, machine learning, and deep learningThe history behind AI and why the artificial intelligence revolution is happening nowHow decades of work in symbolic AI failed and opened the door for the emergence of neural networksWhat neural networks are, how they are trained, and why all the wonder of modern AI boils down to a simple, repeated unit that knows how to multiply input numbers to produce an output number.The implications of large language models, like ChatGPT and Bard, on our society—nothing will be the same again
AI isn’t magic. If you’ve ever wondered how it works, what it can do, or why there’s so much hype, How AI Works will teach you everything you want to know.
192 pages, Kindle Edition
Published October 24, 2023
173 people are currently reading
2065 people want to read
About the author
Ronald T. Kneusel
17 books32 followersMy infatuation with computers began in 1981 with an Apple II. I've been active in machine learning since 2003, and deep learning since before AlexNet was a thing.
My background includes a Ph.D. in computer science from the University of Colorado, Boulder (deep learning), and an M.S. in physics from Michigan State University.
By day, I work in industry building deep learning systems. By night, I type away on my keyboard generating the books you see here. I sincerely hope that if you explore my books, you gain as much enjoyment from them as I had in creating them.
Please see my full list of books here: www.rkneusel.com
Questions? Comments? Let me know at rkneuselbooks@gmail.com
My background includes a Ph.D. in computer science from the University of Colorado, Boulder (deep learning), and an M.S. in physics from Michigan State University.
By day, I work in industry building deep learning systems. By night, I type away on my keyboard generating the books you see here. I sincerely hope that if you explore my books, you gain as much enjoyment from them as I had in creating them.
Please see my full list of books here: www.rkneusel.com
Questions? Comments? Let me know at rkneuselbooks@gmail.com
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 24 of 24 reviews
January 11, 2024
Though challenging to digest at times, Kneusel's introduction to artificial intelligence and machine learning is worth the effort. I particularly enjoyed the section on large language models (LLMs) like ChatGPT and musings on the future of AI in society.
I recommend this book to anyone desiring to have a fundamental grasp of artificial intelligence and machine learning.
I recommend this book to anyone desiring to have a fundamental grasp of artificial intelligence and machine learning.
October 27, 2025
AI books are coming out in record numbers but most seem to be either too technical (heavy on math & code) or too non-technical (user advice, social impact, etc.). This is the first book I've found that hits the sweet spot of being technical enough but still digestible. I came away with a much better understanding of how today's AI systems really work internally.
March 13, 2025
Overall highly recommended to learn the basic concepts, ideas, methods of AI starting from
classical AI methods and the current generation neural network based methods.
The author gives very good rundown on different methods, approaches taken by AI researchers.
complete with very clear explanations on SVN, kNN, and random forest. Then, also gives good
explanations on NNs, CNN, GANs and diffusion models. However, the part focused on LLMs seems
to be much limited in explanations but more on capabilities of the outputs. Also I believe
the musings on future of AI are well... musings, you can agree with them or disagree with them.
classical AI methods and the current generation neural network based methods.
The author gives very good rundown on different methods, approaches taken by AI researchers.
complete with very clear explanations on SVN, kNN, and random forest. Then, also gives good
explanations on NNs, CNN, GANs and diffusion models. However, the part focused on LLMs seems
to be much limited in explanations but more on capabilities of the outputs. Also I believe
the musings on future of AI are well... musings, you can agree with them or disagree with them.
June 8, 2025
How AI Works” is an excellent entry-level guide for anyone curious about artificial intelligence. It delivers a clear and engaging historical overview of the field, helping readers appreciate how far we’ve come and what milestones have shaped current developments. The chronological structure of the book is particularly well-done, offering a smooth progression from the foundations of AI to modern-day applications.
One of the book’s standout features is its concise yet insightful presentation of current AI algorithms and methodologies. The content is remarkably up to date, especially regarding the state of large language models (LLMs). The author does a great job not just explaining these models, but also critically analyzing them—highlighting both their capabilities and limitations in a balanced and informed manner. This level of analysis is rare in introductory texts and makes the book especially valuable.
That said, the book falls a bit short when it comes to explaining some of the more complex topics. In particular, sections on Generative Adversarial Networks (GANs) and diffusion models attempt to address important concepts but end up being overly compressed or unclear for beginners. These parts would benefit greatly from more foundational explanations or simplified visual aids—perhaps even a short tutorial-style walkthrough to help demystify the mathematics and intuition behind these algorithms.
One of the book’s standout features is its concise yet insightful presentation of current AI algorithms and methodologies. The content is remarkably up to date, especially regarding the state of large language models (LLMs). The author does a great job not just explaining these models, but also critically analyzing them—highlighting both their capabilities and limitations in a balanced and informed manner. This level of analysis is rare in introductory texts and makes the book especially valuable.
That said, the book falls a bit short when it comes to explaining some of the more complex topics. In particular, sections on Generative Adversarial Networks (GANs) and diffusion models attempt to address important concepts but end up being overly compressed or unclear for beginners. These parts would benefit greatly from more foundational explanations or simplified visual aids—perhaps even a short tutorial-style walkthrough to help demystify the mathematics and intuition behind these algorithms.
February 9, 2025
I was hoping for a lot more from this book (No Starch Press has so many good books), but it suffers from two flaws:
1) It doesn't do a great job at introducing machine learning to beginners. It's okay, but it suffers from the same flaw that bedevils most "intro" books covering very complicated topics in that it's really, really difficult to convey complex information in the a thin volume.
2) Chapters 7, on LLMs, and 8, on the future of AI, are ridiculous. The majority of chapter 7 is spent on cherry-picked examples that attempt to convey the idea that LLMs are borderline AGI, which they most definitely are not. There is very little information on how LLMs work, but lots of starry-eyed anthropomorphizing enthusiasm that veers into apologia when discussing LLM failures. It's also annoying in chapter 8 when the author waxes positively on the future of AI while paying little to no attention to obvious ethical concerns like social disruption, deepfakes, and the use of copyrighted works as ML training data.
The book does have useful insights, though, even if you know quite a bit about machine learning. Maybe just rip out or delete chapters 7 and 8 from your copy.
1) It doesn't do a great job at introducing machine learning to beginners. It's okay, but it suffers from the same flaw that bedevils most "intro" books covering very complicated topics in that it's really, really difficult to convey complex information in the a thin volume.
2) Chapters 7, on LLMs, and 8, on the future of AI, are ridiculous. The majority of chapter 7 is spent on cherry-picked examples that attempt to convey the idea that LLMs are borderline AGI, which they most definitely are not. There is very little information on how LLMs work, but lots of starry-eyed anthropomorphizing enthusiasm that veers into apologia when discussing LLM failures. It's also annoying in chapter 8 when the author waxes positively on the future of AI while paying little to no attention to obvious ethical concerns like social disruption, deepfakes, and the use of copyrighted works as ML training data.
The book does have useful insights, though, even if you know quite a bit about machine learning. Maybe just rip out or delete chapters 7 and 8 from your copy.
July 7, 2024
"How AI Works: From Sorcery to Science" by Ronald T. Kneusel is a great book for anyone curious about artificial intelligence. These days, we use AI tools like ChatGPT all the time, and it's easy to forget how amazing they are. This book helps bring back that feeling of wonder by explaining how AI actually works.
The best thing about this book is how easy it is to understand. Kneusel explains AI step by step, starting with simple ideas and slowly moving to more complex ones. He doesn't use complicated words, which makes it feel like you're just chatting with a friend who knows a lot about AI.
Reading this book made me really impressed with what humans have created. It's amazing to learn about all the clever ideas behind AI. It made me appreciate these tools even more.
If you want to learn about AI but don't know where to start, this book is perfect. It's a fun way to understand the technology we use every day. I definitely recommend it to anyone who wants to know more about how AI works!
The best thing about this book is how easy it is to understand. Kneusel explains AI step by step, starting with simple ideas and slowly moving to more complex ones. He doesn't use complicated words, which makes it feel like you're just chatting with a friend who knows a lot about AI.
Reading this book made me really impressed with what humans have created. It's amazing to learn about all the clever ideas behind AI. It made me appreciate these tools even more.
If you want to learn about AI but don't know where to start, this book is perfect. It's a fun way to understand the technology we use every day. I definitely recommend it to anyone who wants to know more about how AI works!
December 13, 2024
*How AI Works* by Ron Kneusel is a concise and highly readable introduction to AI, offering clear explanations of core concepts like neural networks and machine learning. Despite largely being text-only, the book does an excellent job of simplifying complex topics, making it accessible to a wide audience. Under 200 pages, it’s a short read that doesn’t sacrifice depth where it matters.
However, certain topics like GANs and Gradient Descent felt challenging to fully grasp with the text-only approach, requiring a few re-reads to piece things together. The lack of visuals or code examples, while understandable given the book's scope, may leave some readers wishing for a bit more support in these areas.
Overall, it’s an impressive effort, particularly given the constraints of a text-only format. If you’re looking for a short, well-written introduction to AI without getting bogged down in technical details, this book is a good choice. Easily a 4/5.
However, certain topics like GANs and Gradient Descent felt challenging to fully grasp with the text-only approach, requiring a few re-reads to piece things together. The lack of visuals or code examples, while understandable given the book's scope, may leave some readers wishing for a bit more support in these areas.
Overall, it’s an impressive effort, particularly given the constraints of a text-only format. If you’re looking for a short, well-written introduction to AI without getting bogged down in technical details, this book is a good choice. Easily a 4/5.
November 23, 2024
As I learn more about applying LLMs (large language models) this year in Salesforce, I wanted a broader perspective on AI. This book - How AI Works: From Sorcery to Science by Ronald T. Kneusel -
delivered exactly that view.
The author gives an overview of AI's development which has been underway for decades. There are chapters on machine learning, neural networks and LLMs. My favorite chapters were right at the end of the book, which focused on LLMs (i.e., Chat GPT and similar) and musings about the future of AI. I would have liked to see more exploration of applied AI to solve problems.
Even though the author describes the book as non-technical, I found several sections of it to be quite challenging. Several pages - especially some of the theoretical explanations - felt incomprehensible. Notwithstanding these confusing passages, I found it a helpful book to further understand AI.
delivered exactly that view.
The author gives an overview of AI's development which has been underway for decades. There are chapters on machine learning, neural networks and LLMs. My favorite chapters were right at the end of the book, which focused on LLMs (i.e., Chat GPT and similar) and musings about the future of AI. I would have liked to see more exploration of applied AI to solve problems.
Even though the author describes the book as non-technical, I found several sections of it to be quite challenging. Several pages - especially some of the theoretical explanations - felt incomprehensible. Notwithstanding these confusing passages, I found it a helpful book to further understand AI.
April 16, 2025
It seems that the author was writing a book about neural networks in 2022 and ChatGPT happened in the middle of his writing, as the last 2 chapters are more a dithyramb to it.
He just assumes that hallucinations would be solved, etc.
First 6 chapters are more thorough written, but I’m not sure about the audience. Yes, he avoided any formula, but his casual usage of words like “manifold” does not assume a completely unprepared audience. And in many cases a formula or two would actually make it easier to understand.
He just assumes that hallucinations would be solved, etc.
First 6 chapters are more thorough written, but I’m not sure about the audience. Yes, he avoided any formula, but his casual usage of words like “manifold” does not assume a completely unprepared audience. And in many cases a formula or two would actually make it easier to understand.
March 26, 2024
The first chapters are an amazing explanation on the basics of AI models, very easy to read if you’re STEM trained.
The last couple of chapters, however, were a great leap from the factual to the speculative. The author’s enthusiasm was contagious, but I think was getting the better of him in terms of providing facts instead of opinionated commentary, largely through rose tinted glasses, about the capabilities of LLMs.
The last couple of chapters, however, were a great leap from the factual to the speculative. The author’s enthusiasm was contagious, but I think was getting the better of him in terms of providing facts instead of opinionated commentary, largely through rose tinted glasses, about the capabilities of LLMs.
March 16, 2024
Really good on explaining the "how" of how AI works, but missing out on the "why" in a lot of chapters. May be this was intentional, but still would have liked a little bit more explanations on cause and effect. I was thinking of giving this three stars but the last chapter on "talking" with AI was fun.
May 9, 2024
Another really well written and very consumable by technical and non technical people. If you are curious about AI, I really recommend reading this book, though Machine Learning for Absolutely Beginners was a slightly better starter, which this being a fast follower, it's a really good combination.
August 24, 2025
Solid, approachable overview of how modern AI actually works. Covers everything from classic models to CNNs, GANs, diffusion, and LLMs, all in plain English. If you’re already using or building with AI tools, you probably won’t find much new, but it’s a great refresher and fills in context gaps. Good for curious users or technical folks who want a no-math, big-picture guide.
January 19, 2024
I really enjoyed this book and found it really helpful for learning the names of various AI models. I am trying to get started on AI programming and found this a great first step. #Goodreadsgiveaways
September 16, 2024
Una buena introducción al mundo de la IA. Da nociones históricas básicas y una explicación solida de la diferencia entre IA, machine learning y deep learning. El capítulo sobre LLMs es particularmente actual.
May 19, 2025
A Helpful Overview For The Naive Student Of AI
I find the underlying math challenging, while this readable book gave me a general insight as to how the math is used to create, and train, AI systems
I find the underlying math challenging, while this readable book gave me a general insight as to how the math is used to create, and train, AI systems
October 7, 2023
Very basic intro into AI.
January 15, 2024
Great explanation to technology. This was a Goodreads giveaway winner.
January 27, 2025
great !!
May 3, 2025
Found this to be very readable and a great explanation.
January 2, 2026
(Original SG Rating: 4.25)
January 7, 2026
Nearly everything I was looking for - super helpful.
Although I say nearly because I wish there was more attention given to transformers and LLMs, but hey, I bet I can find a book for that.
Although I say nearly because I wish there was more attention given to transformers and LLMs, but hey, I bet I can find a book for that.
November 4, 2024
AI is fully based on neural networks. The author makes it a point to discuss neural networks in details: NN, CNN, GAN (for Generative AI), RNN, GPT. We see some examples of each of these and how they fare on different data. Finally, ChatGPT and LLM's are discussed and the probability of how and when AI will replace human tasks and take away more jobs from us, are elaborated on. Read this book if you want concrete examples on how NN and ChatGPT perform on sample problems.
It's a good sidebook to read.
It's a good sidebook to read.
April 1, 2025
Quite technical for a lay person like me. If you understand what the books say then it's likely that you already have a good idea about how AI works.
Displaying 1 - 24 of 24 reviews