What do you think?
Rate this book
A Citizen's Guide to Artificial Intelligence
A concise but informative overview of AI ethics and policy.
Artificial intelligence, or AI for short, has generated a staggering amount of hype in the past several years. Is it the game-changer it's been cracked up to be? If so, how is it changing the game? How is it likely to affect us as customers, tenants, aspiring home-owners, students, educators, patients, clients, prison inmates, members of ethnic and sexual minorities, voters in liberal democracies? This book offers a concise overview of moral, political, legal and economic implications of AI. It covers the basics of AI's latest permutation, machine learning, and considers issues including transparency, bias, liability, privacy, and regulation.
Artificial intelligence, or AI for short, has generated a staggering amount of hype in the past several years. Is it the game-changer it's been cracked up to be? If so, how is it changing the game? How is it likely to affect us as customers, tenants, aspiring home-owners, students, educators, patients, clients, prison inmates, members of ethnic and sexual minorities, voters in liberal democracies? This book offers a concise overview of moral, political, legal and economic implications of AI. It covers the basics of AI's latest permutation, machine learning, and considers issues including transparency, bias, liability, privacy, and regulation.
232 pages, Hardcover
Published February 23, 2021
16 people are currently reading
198 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 7 of 7 reviews
July 15, 2024
Jon Zerilli et. al. have offered up a primer for the coming AI Future that misses none of the nuance. This book takes you through the various topic areas related to Artificial Intelligence, the pros, the cons, and the gray area in between. It does so in a way that builds easily from one section to the next or allows you to tackle them independently. The sum of the parts is a complete and compelling philosophical discussion on everything machine learning and artificial intelligence.
The crux of their approach with this book is to dispel the myth and hype around Artificial Intelligence. While also being cautiously optimistic about what we can achieve with AI if we avoid the pitfalls. Being philosophically grounded, the book is as informative about humanity as it is about Artificial Intelligence. Our authors show how humans create the models, training sets, and other components; and in doing so we naturally infuse it with our biases and societal complexities. The best gem they uncover is that in our approach to create thinking machines, we have to learn to think differently about how the social contract and what it means to be human.
As packed as that agenda is, the authors put this book together using a clear, consistent, and concise voice. That was a smart move that makes the book a sticky page turner despite the heady topics. It maintains an artfully written and intelligent prose that is accessible to all despite the myriad of expert opinions that comprise it. Artificial Intelligence is complex by nature and the conversations around it leave many people behind. Our authors have bridged that gap and perhaps added some knowledge that should be part of the national cannon. I highly recommend it to all.
The crux of their approach with this book is to dispel the myth and hype around Artificial Intelligence. While also being cautiously optimistic about what we can achieve with AI if we avoid the pitfalls. Being philosophically grounded, the book is as informative about humanity as it is about Artificial Intelligence. Our authors show how humans create the models, training sets, and other components; and in doing so we naturally infuse it with our biases and societal complexities. The best gem they uncover is that in our approach to create thinking machines, we have to learn to think differently about how the social contract and what it means to be human.
As packed as that agenda is, the authors put this book together using a clear, consistent, and concise voice. That was a smart move that makes the book a sticky page turner despite the heady topics. It maintains an artfully written and intelligent prose that is accessible to all despite the myriad of expert opinions that comprise it. Artificial Intelligence is complex by nature and the conversations around it leave many people behind. Our authors have bridged that gap and perhaps added some knowledge that should be part of the national cannon. I highly recommend it to all.
July 3, 2022
A good introduction to AI technologies and their ethical, legal, and social implications, even if verging on techno-optimism at times.
May 6, 2024
It’s a decent book and gives one a good feel for the interaction between predictive machine learning technologies (this is primarily what he means by ‘AI’) and human flourishing (politically and individually), a field that has started to call itself AI Ethics. Of particular concern to him, and to many people, are concerns about bias, transparency, and fairness, and these are discussed in a fair amount of detail. He spends a lot of time dealing with hype, providing conceptual clarifications, and motivating the philosophical/legal details through case studies.
A running example in this field is that of predictive policing: police using machine learning technologies to best allocate their resources, and to figure out where crime is. The idea behind this is admirable, in that it is efficient, saving resources, and supposedly unbiased, in that no humans are allowed to involve themselves in the allocation process. But the problem of course is that the process isn’t exactly bias free, even if no protected traits are fed into the algorithm: there are proxies, like postcode (a major problem in cities with historical segregation), which still track those protected traits, and this can lead to disproportionate policing of African American and Latinx communities. This, in turn, can reinforce police bias, through a kind of self-fulfilling and self-reinforcing process, in which biases are reinforced through the discovery of disproportionate amounts of crime in the target area. The problem then, is that a lot of these technologies, though they provide efficiency and quick fixes, do not address systemic issues. And I guess this was my main takeaway from the book: to keep an eye out for these dangerous quick fixes, to keep systemic issues in mind (and this of course requires the participation of marginalised communities in addition to lawmakers and technologists), and to stay clear of the hype, opting for books and papers instead of eye-grabbing news headlines. I should also add that he addresses issues that go beyond what I have discussed in this review so if I've piqued your interest, have a look at the table of contents.
At the same time, the book won’t give you too much insight into artificial general intelligence and agent-based systems (ChatGPT hadn't achieved anything of substance at the time of its writing, but this can easily be remedied by reading Bostrom’s Superintelligence or Stuart’s Human Compatible.
A running example in this field is that of predictive policing: police using machine learning technologies to best allocate their resources, and to figure out where crime is. The idea behind this is admirable, in that it is efficient, saving resources, and supposedly unbiased, in that no humans are allowed to involve themselves in the allocation process. But the problem of course is that the process isn’t exactly bias free, even if no protected traits are fed into the algorithm: there are proxies, like postcode (a major problem in cities with historical segregation), which still track those protected traits, and this can lead to disproportionate policing of African American and Latinx communities. This, in turn, can reinforce police bias, through a kind of self-fulfilling and self-reinforcing process, in which biases are reinforced through the discovery of disproportionate amounts of crime in the target area. The problem then, is that a lot of these technologies, though they provide efficiency and quick fixes, do not address systemic issues. And I guess this was my main takeaway from the book: to keep an eye out for these dangerous quick fixes, to keep systemic issues in mind (and this of course requires the participation of marginalised communities in addition to lawmakers and technologists), and to stay clear of the hype, opting for books and papers instead of eye-grabbing news headlines. I should also add that he addresses issues that go beyond what I have discussed in this review so if I've piqued your interest, have a look at the table of contents.
At the same time, the book won’t give you too much insight into artificial general intelligence and agent-based systems (ChatGPT hadn't achieved anything of substance at the time of its writing, but this can easily be remedied by reading Bostrom’s Superintelligence or Stuart’s Human Compatible.
December 12, 2025
Written in 2019, A Citizen's Guide to AI is one of the more approachable and level-headed books I’ve read to date on the subject of regulating AI. Instead of addressing the typical doomsday scenarios, the authors focus on where AI already shows up in our lives - i.e., risk scores, policing tools, decision systems in government and business - and explain why things like transparency, bias, and accountability matter so much to us all.
What I personally appreciated the most was the steady insistence that AI needs meaningful oversight. Not because machines are about to take over, but because our institutions (and people) can easily overtrust automated systems, or use them in ways that "quietly" erode fairness and privacy. The examples - especially around policing, child welfare, and judicial tools - make the point well but without feeling alarmist.
There's nothing remotely technical about it, and some people may want more on how these systems operate under the hood. But as a civic guide to AI, it obviously hits the mark. So if you are focused on this sort of thing in your work, then this book is well worth a look. It's clear, thoughtful, and practical.
What I personally appreciated the most was the steady insistence that AI needs meaningful oversight. Not because machines are about to take over, but because our institutions (and people) can easily overtrust automated systems, or use them in ways that "quietly" erode fairness and privacy. The examples - especially around policing, child welfare, and judicial tools - make the point well but without feeling alarmist.
There's nothing remotely technical about it, and some people may want more on how these systems operate under the hood. But as a civic guide to AI, it obviously hits the mark. So if you are focused on this sort of thing in your work, then this book is well worth a look. It's clear, thoughtful, and practical.
October 31, 2021
Vital for getting an understanding of the most important aspects of AI today and about its future, all explained and written in a comprehensible manner.
June 23, 2022
3.5, read for class, purr
March 3, 2025
Overall, a good overview of issues surrounding AI.
Displaying 1 - 7 of 7 reviews