Machines Behaving Badly: The Morality of AI

by Toby Walsh

Can we build moral machines?
Artificial intelligence is an essential part of our lives – for better or worse. It can be used to influence what we buy, who gets shortlisted for a job and even how we vote. Without AI, medical technology wouldn't have come so far, we'd still be getting lost in our GPS-free cars, and smartphones wouldn't be so, well, smart. But as we continue to build more intelligent and autonomous machines, what impact will this have on humanity and the planet?
Professor Toby Walsh, a world-leading researcher in the field of artificial intelligence, explores the ethical considerations and unexpected consequences AI poses. Can AI be racist? Can robots have rights? What happens if a self-driving car kills someone? What limitations should we put on the use of facial recognition? Machines Behaving Badly is a thought-provoking look at the increasing human reliance on robotics and the decisions that need to be made now to ensure the future of AI is a force for good, not evil.

Genres: Nonfiction, Technology, Science, Artificial Intelligence

262 pages, Kindle Edition

Published May 26, 2022

About the author

Toby Walsh is one of the world’s leading researchers in Artificial Intelligence. He is a Professor of Artificial Intelligence at the University of New South Wales and leads a research group at Data61, Australia’s Centre of Excellence for ICT Research. He has been elected a fellow of the Association for the Advancement of AI for his contributions to AI research, and has won the prestigious Humboldt research award. He has previously held research positions in England, Scotland, France, Germany, Italy, Ireland and Sweden.

Reviews

[Indy Scarletti (paperindy) · Rating 4 out of 5]

This was a super readable introduction to the complexity of AI ethics. Though a lot of the issues covered I’ve come across before, I did get a few great new examples to look at and thought this was a fascinating read coming from an AI researcher himself. Not overly complex and very accessible, if you want a good starting point for this topic this would be a great book to pick-up.

[Majdy · Rating 3 out of 5]

A nice read

The book is certainly a good read but on a very surface levels. It touches on the problems of ai but if you are tech savvy and have read into the subject before. It doesnt really offer anything new or expand your knowledge

[Danielle · Rating 4 out of 5]

A conversational-style book that addresses how AI has its limitations, particularly in robotics, ethics, and emotional range. I thought it was a good read and quite uncanny to see how the author's predictions played out or not, given that we have come a long way since ChatGPT came out. I do think the tone of the book can be a bit doomsayer-y, which I don't disagree with, but considering I am already a pessimist maybe it might not be as balanced as I think.

[Graham Jenkins · Rating 5 out of 5]

You've probably all heard a variant of what's known as a "trolley problem", where we have a runaway trolley heading down a railway track. Five men are working on this track and are certain to die when the trolley reaches them.

However, it's possible for you to switch the trolley to an alternative spur on the track, saving all five lives. Unfortunately, one man is working on this spur, and he will be killed if the switch is made. Surveys apparently show that 90 percent of people will go ahead and switch the trolley.

There's a variant of this problem, where the errant trolley is about to go under a footbridge, and there's a fat man on the bridge who you can push off and into the path the trolley; he will surely stop the trolley, but he will die. Surveys show that only 10 percent of people will push him; go figure!

If humans exhibit such apparently inexplicit behaviour patterns, how can we expect a self-driving car to decide whether it should kill the child crossing the road ahead of it, or crash into a wall and kill its driver?

The author doesn't have a solution for this, but he does present some fascinating scenarios that will help you understand the benefits and the dangers that come with artificial intelligence.

[Aneesha · Rating 4 out of 5]

Did not expect to enjoy reading this quite as much as I did. The book asks what I think are important questions for these strange times that we live in.

[Enie X · Rating 4 out of 5]

An easy read for someone who wants to learn about ethical concerns of AI application.

[Jana · Rating 4 out of 5]

3.5

[Pharlap · Rating 4 out of 5]

I was quite suspicious about the title of this book.
Are machines really behaving? Can they be judged by human standards?
In my opinion, the author very early in the book gives a negative answer to these questions.
Machines behave as they were programmed by the people.
Here is the heart of the matter.
Who are the people who stand behind the machines?
Author introduces them as geeks from the Crazy Valley.
Here most phantastic ideas are conceived, some of them are noticed.
Noticed by whom?
The answer is even more disturbing than for the first question - biggest and most influential companies in the world - Amazon, Google, Microsoft, Facebook.
Each of them has a long record of tax avoidance and abuse of privacy.
Do we have any other option, any choice?
Answer is No.
One important point - author mentions some basic human behaviours and reactions , the most obvious is pain.
The point is that pain is biology and chemistry while the machines react only to physics. Therefor the logical conclusion is that AI should be translated as ALIEN Intelligence - fully agree.
From this point the book contains a long list of problems we can encounter when using SI and many attempts of National and International organisations to set up some uniform rules and standards. Looks like all of them are destinied for failure.

[Karen ⊰✿ · Rating 4 out of 5]

If we think back to the major technological changes over the last 20 years, social media comes to mind. Particularly how different it was when we first logged in to myspace or facebook, wide eyed and innocent. Thinking we were sharing with friends and having a bit of fun. Versus the massive privacy and security concerns that have emerged over the last almost 20 years.
Perhaps that collective experience has given us the ability to now reflect critically and call for change and regulation with AI, something that took probably closer to 10 years to happen more broadly with social media.
It seems that just putting our head in the ground on AI is not really an option and so I appreciate this book and others like it that are aiming to educate us non-techies in simple language on what AI is and what important moral questions we need to consider.

[Nestor · Rating 4 out of 5]

An open-eyes book on how AI ethics is distorted by governments, organizations, and companies. The book has its highs and lows, but it provides a good summary of ethics rules and the violations and contradictions that have been said.

Don't expect a high philosophy book on the subject, is more down to the earth, but is a good start for people interested in the subject and starting to learn. As I always say, the answer, forecasting or the solutions for humankind's future is already in the Sci-Fi literature like or not, and that's why I love Sci-Fi.

His proposal that we pay $30/year to Google so they don't sell our data doesn't fit really, since we did that for Cable TV and now Streaming services and they now not only charge us for the services but they sell Ads, so this is not the solution.

[Thomas Goddard · Rating 3 out of 5]

Reasonably quick read this one. The chief concern of the age... I feel foolish ending that sentence there. How about... AI, the chief concern of our age along with war, disease, obseity, climate change, political and religious extremism, loss of biodiversity, water scarcity, food security, misinfomration, loss of privacy.... that's enough.

AI is hailed by the tech bros worldwide as being the solution to a lot of those issues, though not of course to itself. We've a lot to contend with in the next few decades. Didn't enjoy this one as much as I enjoyed Superintelligence by Nick Bostrom.

It isn't a bad book. But is a basic introduction to the topic. If you know next to nothing about AI, then it is a great starting point. I wanted more.

[Julian Walker · Rating 4 out of 5]

A highly thought-provoking and stimulating read, packed with observations and historical perspectives, the author adds his own views on many of the more outlandish AI propositions and promotions, and a healthy dose of realism against some of the claims.

One early and fascinating insight is around Facebook's involvement with Cambridge Analytica and the historical precedent to the initiative (from around seventy years earlier), debunking many of the stories and statements made in the more recent episode's aftermath.

Great thinking, well presented - and a very satisfying read.

[Carol Griffiths · Rating 4 out of 5]

A fascinating look at the world of Artificial Intelligence - both the potential threats and potential benefits. Walsh suggests that the world needs to work quickly to minimise the threats and maximise the benefits. His reporting of the behaviour of some of the Big Techs is quite worrying, but he does end on a positive note.

[Oz Book · Rating 5 out of 5]

There's a lot being written about AI these days. Most of it by people who don't have much of an understanding of what it is they're writing about.

It's therefore refreshing to come across this book from one of the experts in the field.

It's a well rounded and thoughtful look at some of the most pressing problems facing tech today. Well worth a read.

[Craig Martin · Rating 4 out of 5]

A timely refresher on the opportunities and threats that AI bring. Toby Walsh writes well and has a generally balanced view. This is a good introductory text, and is particularly relevant given the current debate about ‘pausing training’ of large scale models given the rapid advances of GPTs.

[Sue · Rating 4 out of 5]

An excellent, engaging book talking about the ethics and morality of robots and AI. Defnitely a book to read right now as AI is where the world wants to head but at what cost?

[Suzie Leadabrand · Rating 4 out of 5]

Fascinating and easy to read and an eye opener on burgeoning technology companies.

Segued into Infinity Gate, fiction, which made reading interesting after T Walsh’s book.