The Alignment Problem

Machine Learning and Human Values

By: Brian Christian

Narrated by: Brian Christian

A jaw-dropping exploration of everything that goes wrong when we build AI systems and the movement to fix them.

Today’s “machine-learning” systems, trained by data, are so effective that we’ve invited them to see and hear for us - and to make decisions on our behalf. But alarm bells are ringing. Recent years have seen an eruption of concern as the field of machine learning advances. When the systems we attempt to teach will not, in the end, do what we want or what we expect, ethical and potentially existential risks emerge. Researchers call this the alignment problem.

Systems cull résumés until, years later, we discover that they have inherent gender biases. Algorithms decide bail and parole - and appear to assess Black and White defendants differently. We can no longer assume that our mortgage application, or even our medical tests, will be seen by human eyes. And as autonomous vehicles share our streets, we are increasingly putting our lives in their hands.

The mathematical and computational models driving these changes range in complexity from something that can fit on a spreadsheet to a complex system that might credibly be called “artificial intelligence.” They are steadily replacing both human judgment and explicitly programmed software.

In best-selling author Brian Christian’s riveting account, we meet the alignment problem’s “first-responders,” and learn their ambitious plan to solve it before our hands are completely off the wheel. In a masterful blend of history and on-the-ground reporting, Christian traces the explosive growth in the field of machine learning and surveys its current, sprawling frontier. Listeners encounter a discipline finding its legs amid exhilarating and sometimes terrifying progress. Whether they - and we - succeed or fail in solving the alignment problem will be a defining human story.

The Alignment Problem offers an unflinching reckoning with humanity’s biases and blind spots, our own unstated assumptions and often contradictory goals. A dazzlingly interdisciplinary work, it takes a hard look not only at our technology but at our culture - and finds a story by turns harrowing and hopeful.

©2020 Brian Christian (P)2020 Brilliance Publishing, Inc., all rights reserved.

A must read to every data scientist👩🏼‍💻

This book is most definitely the best creation by Brian Christian. I’m speechless, and listened to the whole book in 6 days. This is a brilliant niche in AI, up to date, and fascinating.
I cannot emphasize in words how much I enjoyed it (perhaps in a word2vec it’s the infinity vector?) And I can’t wait for Brian’s next audiobook.

Best overview of the topic I’ve read

This is an excellent overview of the field of machine learning - its history, the problems it faces as it is applied in the real world, and the potential routes for its (and our) future. I’ve read many layperson books on the topic in the past, and I learned a ton from The Alignment Problem. I now have a more complete understanding of the structure of the field and the technical framework it’s built on, as well as the what the technology can and so far cannot do for us. Equally of interest are the questions about human society that the book raises. Can we succeed at aligning AI’s goals with our own if we don’t really understand what our own goals are? Christian describes the current ideas for how this may be possible, as well as the ways in which AI has so far been highlighting moral questions that we didn’t even realize we were confused about.

The amount of research that went into this book is astounding - Christian draws from interviews with the top figures in every aspect of the field. It’s written clearly and succinctly, in a way that should be accessible to any reader, regardless of their prior knowledge of the topic. It’s also narrated by the author in a clear and incisive voice. An overall fantastic experience.

Good history of the problem

The author provides a detailed history of the problem in relation to machine, learning, and human values. It was an interesting read, although dry in some spots was overall worth the read.

This book was very enlightening

This book does a great job of telling the history and current limits of AI. It also does a good job of shining a light on or own shortcomings and how we should be careful not to imprint them in our AI. We should be careful but we can learn a lot from each other.

The narration was great and the story was very interesting.

Hands Down: the Best AI Book

As a tech professional who has made it a point to listen to a number of these AI books on Audible, I recommend this book. It is truly a must-read.
The quality of the book is solid, the scope is complete and the explanations are approachable. What sets this book apart is that this book talks at length about how the folks that are on the cutting edge of AI are attempting to make machines care about what humans care about. This is the "Alignment Problem".

Phenomenal

This is extremely good. If you want to understand the real dangers of AI, read this book. Most of the recent work is either overblown or catastrophizing, but the author here takes a nuanced approach that looks at the actual research and problems arising from trying to implement human values in AI.

Every other sentence is thought provoking and spurred wholly new insights for me. I cannot recommend this one enough.

Excellent and broad introduction

A very good and broad introduction to the various topics and issues surrounding AI and machine learning.

Enthralling

This was such an incredible portrayal of the advancements in machine learning by tying it to the ways humans learn.
The author, in a very relatable way, ties the development of better learning systems to our collective understanding of how humans grow and learn.
Incredibly fascinating all the way through!

Very enlightening

I learned a great deal about artificial intelligence reading this book including current big ideas, and clearly presented ethical and practical challenges in designing and implementing AI. The author’s narration was engaging and clear.

I strongly recommend this book and audiobook.

Solid book

Great overview of machine learning as it relates to the difficulties implementing human preferences, but little on the threat of ai as an existential issue