What do you think?
Rate this book
Taming the Machine: Ethically Harness the Power of AI
AI promises to transform our world, supercharging productivity and driving new innovations. Taming the Machine uncovers how you can responsibly harness the power of AI with confidence.
AI has the potential to become a personal assistant, a creative partner, an editor and a research tool all at once. But it also represents a threat to your livelihood, data and privacy. Taming the Machine offers the practical insights and knowledge you need to work with AI with an ethical and responsible approach.
In this book, celebrated AI expert and ethicist Nell Watson offers practical insights on how you can ethically innovate with AI. It delves into the ethical issues of unbridled AI, highlighting the challenges that it will bring to society and business unless we fortify cybersecurity, safeguard our data, and understand the dangerous potential of artificial intelligence.
Step into the future and supercharge your performance safely by Taming the Machine .
AI has the potential to become a personal assistant, a creative partner, an editor and a research tool all at once. But it also represents a threat to your livelihood, data and privacy. Taming the Machine offers the practical insights and knowledge you need to work with AI with an ethical and responsible approach.
In this book, celebrated AI expert and ethicist Nell Watson offers practical insights on how you can ethically innovate with AI. It delves into the ethical issues of unbridled AI, highlighting the challenges that it will bring to society and business unless we fortify cybersecurity, safeguard our data, and understand the dangerous potential of artificial intelligence.
Step into the future and supercharge your performance safely by Taming the Machine .
- GenresArtificial Intelligence
336 pages, Paperback
Published May 28, 2024
8 people are currently reading
89 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
June 9, 2024
I found this an excellent read, which sums up the current state of development of AI computing and looks at some pathways and some possible outcomes. Each chapter has the references at the end so it's easy to follow, and there's a glossary but no index. The writing is nicely done and there are some clear tables that are very helpful.
I think readers will benefit from having read some other books before coming to this one. For instance I had read the mentioned books by Neal Stephenson, Nick Bostrom, Robin Hanson, and more not mentioned, which had covered the issue of Target sending vouchers for pregnancy products to a young teen girl, etc., how AI batches are trained to physical work, or the Go moves in the computer match against the master, rather better and at more length. But you can hunt these up if interested. The topic is AI and this author wants to stick to it. Familiarity with the topic also means you know terms like LLM, IOT and AGI, which are explained briefly, but you don't want to keep flicking pages when they come up again. Maslow's hierarchy of needs is another item which it's assumed you'll know.
We cover issues like hacking, pen testing, accidental data leak, and ethics of security. Also, legal frameworks and loopholes. We go from responsible users to third party chip suppliers to inbred datasets - training an AI on a dataset which is the output of a previous AI, with inherent issues. Bias in race, gender and other areas are looked at and suggestions are provided. Again, not in depth. The author is briefly summarising other papers and books.
Not much space is given to the races between the big name players in the market. Nor how trustworthy the general public finds their firms. From personal experience, I have been sent ads by Google onto my list of emails in my phone, for dating sites where I could meet fat women. I'm a woman, and have been an athlete most of my life, but obviously Google misinterpreted the keenness of my internet reading of science and computing papers and books, and my reading on health and good diet. Now suppose the same ads are sent to the phone of a married man, and his wife sees them. I reported the ads as offensive every time, and got automatically told they didn't contravene policies, until eventually Google got tired of serving me ads for services I was never going to buy. On the good side, other Google products such as Maps, Lens and Gmail are becoming more helpful.
A friend showed me his Facebook phone feed over a lunchtime meal, and it contained large ads for pints of beer of various brands. I said, "They know you're in a pub, they're geolocating you." He turned off the location on his phone. The ads were without regard for the fact that he might be a driver, or returning to work, and he does not drink beer, because he doesn't like it. Given the above story about me, I speculate that he might also have been a pregnant woman or under legal age.
A mere paragraph is given to the environmental cost of AI computing, and it's glossed over and we're told that the benefits outweigh the drawbacks. Do they really? Having spent pages counting the number of ways that an AGI might destroy the world, or destroy all humans, or demand ever more resources, or enslave all humans, etc. we're assured it'll be fine. That's after the list of numerous ways that AI is taking our jobs, leaving only gig work, removing training of young staff so they can't progress to senior posts, taking on creative outputs and peddling lies and deepfakes to targeted electorates. The author does not explain that an AI search as opposed to a standard internet search, uses 6 to 10 times as much energy. And we see that the results are prone to hallucinations, lies, sarcastic or bad answers from a single Reddit post presented as truth. The usage of REEs and precious metals, which could go into other items like MRI scanners and e-bikes, and trashpiles of chips, motherboards and GPUs from which metals are not recovered... this will make it necessary to mine the asteroids, not for us, but for AI demands.
Glossary p. 294 - 299 in my e-ARC. This book will be useful to those in the computer industry, tech journalists, students, and business people generally. A good point is that after each chapter we get a box on the bottom line, the big picture, and another on leadership actions. This might include providing opt in or opt out material to employees or users, regular checks that data is secure, collaborating with others and with universities. No photos - I thought a few photos of datacentres, AIs being trained on physical tasks, and some of the major names in the field, would be helpful to students. Largely names are absent and general ethics discussions are prominent. But it'll all work out fine. Maybe that's what an AI would say if it wrote this book. I can but speculate.
I read an e-ARC from Net Galley. This is an unbiased review by a live human.
I think readers will benefit from having read some other books before coming to this one. For instance I had read the mentioned books by Neal Stephenson, Nick Bostrom, Robin Hanson, and more not mentioned, which had covered the issue of Target sending vouchers for pregnancy products to a young teen girl, etc., how AI batches are trained to physical work, or the Go moves in the computer match against the master, rather better and at more length. But you can hunt these up if interested. The topic is AI and this author wants to stick to it. Familiarity with the topic also means you know terms like LLM, IOT and AGI, which are explained briefly, but you don't want to keep flicking pages when they come up again. Maslow's hierarchy of needs is another item which it's assumed you'll know.
We cover issues like hacking, pen testing, accidental data leak, and ethics of security. Also, legal frameworks and loopholes. We go from responsible users to third party chip suppliers to inbred datasets - training an AI on a dataset which is the output of a previous AI, with inherent issues. Bias in race, gender and other areas are looked at and suggestions are provided. Again, not in depth. The author is briefly summarising other papers and books.
Not much space is given to the races between the big name players in the market. Nor how trustworthy the general public finds their firms. From personal experience, I have been sent ads by Google onto my list of emails in my phone, for dating sites where I could meet fat women. I'm a woman, and have been an athlete most of my life, but obviously Google misinterpreted the keenness of my internet reading of science and computing papers and books, and my reading on health and good diet. Now suppose the same ads are sent to the phone of a married man, and his wife sees them. I reported the ads as offensive every time, and got automatically told they didn't contravene policies, until eventually Google got tired of serving me ads for services I was never going to buy. On the good side, other Google products such as Maps, Lens and Gmail are becoming more helpful.
A friend showed me his Facebook phone feed over a lunchtime meal, and it contained large ads for pints of beer of various brands. I said, "They know you're in a pub, they're geolocating you." He turned off the location on his phone. The ads were without regard for the fact that he might be a driver, or returning to work, and he does not drink beer, because he doesn't like it. Given the above story about me, I speculate that he might also have been a pregnant woman or under legal age.
A mere paragraph is given to the environmental cost of AI computing, and it's glossed over and we're told that the benefits outweigh the drawbacks. Do they really? Having spent pages counting the number of ways that an AGI might destroy the world, or destroy all humans, or demand ever more resources, or enslave all humans, etc. we're assured it'll be fine. That's after the list of numerous ways that AI is taking our jobs, leaving only gig work, removing training of young staff so they can't progress to senior posts, taking on creative outputs and peddling lies and deepfakes to targeted electorates. The author does not explain that an AI search as opposed to a standard internet search, uses 6 to 10 times as much energy. And we see that the results are prone to hallucinations, lies, sarcastic or bad answers from a single Reddit post presented as truth. The usage of REEs and precious metals, which could go into other items like MRI scanners and e-bikes, and trashpiles of chips, motherboards and GPUs from which metals are not recovered... this will make it necessary to mine the asteroids, not for us, but for AI demands.
Glossary p. 294 - 299 in my e-ARC. This book will be useful to those in the computer industry, tech journalists, students, and business people generally. A good point is that after each chapter we get a box on the bottom line, the big picture, and another on leadership actions. This might include providing opt in or opt out material to employees or users, regular checks that data is secure, collaborating with others and with universities. No photos - I thought a few photos of datacentres, AIs being trained on physical tasks, and some of the major names in the field, would be helpful to students. Largely names are absent and general ethics discussions are prominent. But it'll all work out fine. Maybe that's what an AI would say if it wrote this book. I can but speculate.
I read an e-ARC from Net Galley. This is an unbiased review by a live human.
May 6, 2025
"Taming the Machine" explores AI's potential and dangers, emphasising the need for ethical development. It takes a balanced view of the benefits as well as the precautions one needs to take. At the same time, it highlights AI's transformative power and suggests measures to reduce risks to privacy, security, and livelihoods.
The book stresses embedding ethics into AI through transparency, trustworthiness, and human governance, covering issues like bias, fairness, and accountability. The author provides practical insights for ethical AI implementation, with chapters ending in actionable summaries.
The book has been written in a clear and engaging style, and makes complex topics accessible to a broad audience, not just corporate leaders. It comprehensively addresses challenges like data quality, bias prevention, and privacy.
To conclude, the author calls for proactive measures, such as ethical guidelines, to steer AI development positively. While some reviewers wanted more implementation details, the book is considered a key guide for navigating the AI era responsibly.
The book stresses embedding ethics into AI through transparency, trustworthiness, and human governance, covering issues like bias, fairness, and accountability. The author provides practical insights for ethical AI implementation, with chapters ending in actionable summaries.
The book has been written in a clear and engaging style, and makes complex topics accessible to a broad audience, not just corporate leaders. It comprehensively addresses challenges like data quality, bias prevention, and privacy.
To conclude, the author calls for proactive measures, such as ethical guidelines, to steer AI development positively. While some reviewers wanted more implementation details, the book is considered a key guide for navigating the AI era responsibly.
May 19, 2024
In this book, Watson offers a comprehensive look at artificial intelligence from concept to creation and utilization along with the responsibilities we have for its ethical implementation in our everyday lives. I was particularly interested in the chapter about biases and the need for diversity in the data used to feed or teach AI. This is a wonderful book for the layperson who might be interested in learning more about the technology that is at our doorstep and quickly entering every aspect of our lives. *Advance copy provided by the publisher in exchange for my honest review.
July 27, 2024
Well detailed essay on all the angles of the AI, for the present and near future. The author explain his vision on the different AI applications, concerns & considerations (ethical, social), risks & uncertainties.
It does not go dive on specific business models; rather in the myriad of applications.
It does not go dive on specific business models; rather in the myriad of applications.
September 25, 2024
Well-written and organised, covers a broad range of challenges with AI and relevant research fields (without oversimplifying!). Had the pleasure of meeting Nell several times and she is as eloquent in real life as on paper!
November 20, 2024
Ai isn’t something to be scared of but it’s something to read about and form your own opinions. I use it for my work day and I have found it helps streamline my day but there are negatives too. This is a great book to gift- thoroughly recommend so that you can learn all about AI.
December 21, 2024
Nell does a phenomenal job breaking down extraordinarily complex subjects related to AI into manageable and understandable pieces, even for a layperson. Highly recommended for anyone trying to understand the importance of AI and how it may affect society.
Displaying 1 - 7 of 7 reviews