What do you think?
Rate this book
The Ethics of Artificial Intelligence: Principles, Challenges, and Opportunities
The Ethics of Artificial Intelligence has two goals. The first goal is meta-theoretical and is fulfilled by Part One, which comprises the first three an interpretation of the past (Chapter 1), the present (Chapter 2), and the future of AI (Chapter 3). Part One develops the thesis that AI is an unprecedented divorce between agency and intelligence. On this basis, Part Two investigates the consequences of such a divorce, developing the thesis that AI as a new form of agency can be harnessed ethically and unethically. It begins (Chapter 4) by offering a unified perspective on the many principles that have been proposed to frame the ethics of AI. This leads to a discussion (Chapter 5) of the potential risks that may undermine the application of these principles, and then (Chapter 6) an analysis of the relation between ethical principles and legal norms, and a definition of soft ethics as post-compliance ethics. Part Two continues by analysing the ethical challenges caused
by the development and use of AI (Chapter 7), evil uses of AI (Chapter 8), and good practices when applying AI (Chapter 9). The last group of chapters focuses on the design, development, and deployment of AI for Social Good or AI4SG (Chapter 10); the positive and negative impacts of AI on the environment and how it can be a force for good in the fight against climate change-but not without risks and costs, which can and must be avoided or minimised (Chapter 11); and the possibility of using AI in support of the United Nations Sustainable Development Goals (Chapter 12). The book concludes (Chapter 13) by arguing in favour of a new marriage between the Green of all our habitats and the Blue of all our digital technologies and how this new marriage can support and develop a better society and a healthier biosphere.
by the development and use of AI (Chapter 7), evil uses of AI (Chapter 8), and good practices when applying AI (Chapter 9). The last group of chapters focuses on the design, development, and deployment of AI for Social Good or AI4SG (Chapter 10); the positive and negative impacts of AI on the environment and how it can be a force for good in the fight against climate change-but not without risks and costs, which can and must be avoided or minimised (Chapter 11); and the possibility of using AI in support of the United Nations Sustainable Development Goals (Chapter 12). The book concludes (Chapter 13) by arguing in favour of a new marriage between the Green of all our habitats and the Blue of all our digital technologies and how this new marriage can support and develop a better society and a healthier biosphere.
272 pages, Hardcover
First published January 1, 2022
37 people are currently reading
324 people want to read
About the author
Luciano Floridi
63 books144 followersLuciano Floridi is currently Professor of Philosophy and Ethics of Information at the University of Oxford, Oxford Internet Institute, Governing Body Fellow of St Cross College, Oxford, Senior Member of the Faculty of Philosophy, Research Associate and Fellow in Information Policy at the Department of Computer Science, University of Oxford, and Distinguished Research Fellow of the Oxford Uehiro Centre for Practical Ethics.
Floridi is best known for his work on two areas of philosophical research: the philosophy of information and information ethics.
Between 2008 and 2013, he held the Research Chair in philosophy of information and the UNESCO Chair in Information and Computer Ethics at the University of Hertfordshire. He was the founder and director of the IEG, an interdepartmental research group on the philosophy of information at the University of Oxford, and of the GPI, the research Group in Philosophy of Information at the University of Hertfordshire. He was the founder and director of the SWIF, the Italian e-journal of philosophy (1995–2008).
Floridi is best known for his work on two areas of philosophical research: the philosophy of information and information ethics.
Between 2008 and 2013, he held the Research Chair in philosophy of information and the UNESCO Chair in Information and Computer Ethics at the University of Hertfordshire. He was the founder and director of the IEG, an interdepartmental research group on the philosophy of information at the University of Oxford, and of the GPI, the research Group in Philosophy of Information at the University of Hertfordshire. He was the founder and director of the SWIF, the Italian e-journal of philosophy (1995–2008).
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 9 of 9 reviews
December 11, 2025
Non ho dubbi: si tratta di un volume molto acuto e interessante che avrei voluto scrivere io. L’ho consumato da quanto l’ho letto e sono certa che ancora qualcosa mi sfugga.
December 26, 2023
Libro attualissimo, anche se forse già parzialmente superato a causa della diffusione su larga scala dei modelli di OpenAI, avvenuta subito dopo la pubblicazione del libro e che ha cambiato un po' le prospettive dell'AI.
L'argomento è affrontato in maniera completa, dettagliata ed equilibrata. Ma d'altronde, dal prof. Floridi, non ci si poteva aspettare meno.
La lettura, contrariamente ad altre opere dell'autore, non richiede particolari conoscenze pregresse ed è quindi accessibile a chiunque sia interessato ad approfondire l'argomento.
L'argomento è affrontato in maniera completa, dettagliata ed equilibrata. Ma d'altronde, dal prof. Floridi, non ci si poteva aspettare meno.
La lettura, contrariamente ad altre opere dell'autore, non richiede particolari conoscenze pregresse ed è quindi accessibile a chiunque sia interessato ad approfondire l'argomento.
January 27, 2026
A Must-Read for Anyone Navigating the AI Revolution
Luciano Floridi has established himself as one of the most influential voices in the philosophy of information, and this book demonstrates precisely why. "Ethics of Artificial Intelligence" is a comprehensive, thought-provoking exploration of the moral challenges posed by AI technologies that are increasingly shaping every aspect of our lives.
What sets Floridi apart from other authors writing about AI ethics is his ability to construct a coherent philosophical framework rather than simply cataloguing concerns. His concept of the infosphere and the notion of onlife provide readers with powerful conceptual tools to understand our current technological condition. These are not mere buzzwords but carefully developed ideas that illuminate how profoundly digital technologies have transformed the human experience.
The book's greatest strength lies in its balanced approach. Floridi avoids both naive techno-optimism and paralyzing techno-pessimism. Instead, he offers a nuanced perspective that acknowledges AI's tremendous potential while remaining clear-eyed about its risks. His five ethical principles for AI—beneficence, non-maleficence, autonomy, justice, and explainability—provide a practical framework that can guide developers, policymakers, and citizens alike.
Particularly compelling is Floridi's analysis of AI as a "divorced" technology—one that separates the ability to act from genuine understanding. This insight alone is worth the price of admission, as it helps explain why even highly performing AI systems can fail in unexpected ways and why human oversight remains essential.
The discussion on algorithmic opacity and the right to explanation is especially timely, given the proliferation of black-box systems making consequential decisions about credit, employment, and criminal justice. Floridi makes a convincing case that explainability is not just a technical nice-to-have but a fundamental requirement for democratic accountability.
I also appreciated the emphasis on environmental sustainability—what Floridi calls "green digital." Too often, discussions of AI ethics focus exclusively on social impacts while ignoring the significant carbon footprint of training and deploying large-scale models. Floridi rightfully insists that ecological considerations must be part of any comprehensive ethical assessment.
Why Not Five Stars?
While the book is excellent, it occasionally suffers from the dense, academic prose that characterizes much philosophical writing. Some passages require multiple readings to fully grasp, which may frustrate readers looking for more accessible content. Additionally, while Floridi excels at theoretical analysis, the book could benefit from more concrete case studies and practical examples that illustrate how his principles apply in real-world scenarios.
The rapid pace of AI development also means that some specific examples and concerns discussed may feel slightly dated, though the underlying philosophical framework remains highly relevant.
The Verdict
If you are passionate about artificial intelligence—whether as a developer, researcher, policymaker, business leader, or simply an engaged citizen—this book is essential reading. Floridi provides the intellectual tools to think critically and systematically about the ethical dimensions of AI, moving beyond headlines and hype to engage with the deeper questions at stake.
In an era when AI is transforming everything from healthcare to warfare, from employment to democracy, we desperately need thoughtful frameworks for navigating these changes. Floridi delivers exactly that. This is not light beach reading, but for those willing to engage seriously with the material, the rewards are substantial.
Highly recommended for anyone who wants to understand not just how AI works, but how it should work—and what responsibilities we bear in shaping its future.
Luciano Floridi has established himself as one of the most influential voices in the philosophy of information, and this book demonstrates precisely why. "Ethics of Artificial Intelligence" is a comprehensive, thought-provoking exploration of the moral challenges posed by AI technologies that are increasingly shaping every aspect of our lives.
What sets Floridi apart from other authors writing about AI ethics is his ability to construct a coherent philosophical framework rather than simply cataloguing concerns. His concept of the infosphere and the notion of onlife provide readers with powerful conceptual tools to understand our current technological condition. These are not mere buzzwords but carefully developed ideas that illuminate how profoundly digital technologies have transformed the human experience.
The book's greatest strength lies in its balanced approach. Floridi avoids both naive techno-optimism and paralyzing techno-pessimism. Instead, he offers a nuanced perspective that acknowledges AI's tremendous potential while remaining clear-eyed about its risks. His five ethical principles for AI—beneficence, non-maleficence, autonomy, justice, and explainability—provide a practical framework that can guide developers, policymakers, and citizens alike.
Particularly compelling is Floridi's analysis of AI as a "divorced" technology—one that separates the ability to act from genuine understanding. This insight alone is worth the price of admission, as it helps explain why even highly performing AI systems can fail in unexpected ways and why human oversight remains essential.
The discussion on algorithmic opacity and the right to explanation is especially timely, given the proliferation of black-box systems making consequential decisions about credit, employment, and criminal justice. Floridi makes a convincing case that explainability is not just a technical nice-to-have but a fundamental requirement for democratic accountability.
I also appreciated the emphasis on environmental sustainability—what Floridi calls "green digital." Too often, discussions of AI ethics focus exclusively on social impacts while ignoring the significant carbon footprint of training and deploying large-scale models. Floridi rightfully insists that ecological considerations must be part of any comprehensive ethical assessment.
Why Not Five Stars?
While the book is excellent, it occasionally suffers from the dense, academic prose that characterizes much philosophical writing. Some passages require multiple readings to fully grasp, which may frustrate readers looking for more accessible content. Additionally, while Floridi excels at theoretical analysis, the book could benefit from more concrete case studies and practical examples that illustrate how his principles apply in real-world scenarios.
The rapid pace of AI development also means that some specific examples and concerns discussed may feel slightly dated, though the underlying philosophical framework remains highly relevant.
The Verdict
If you are passionate about artificial intelligence—whether as a developer, researcher, policymaker, business leader, or simply an engaged citizen—this book is essential reading. Floridi provides the intellectual tools to think critically and systematically about the ethical dimensions of AI, moving beyond headlines and hype to engage with the deeper questions at stake.
In an era when AI is transforming everything from healthcare to warfare, from employment to democracy, we desperately need thoughtful frameworks for navigating these changes. Floridi delivers exactly that. This is not light beach reading, but for those willing to engage seriously with the material, the rewards are substantial.
Highly recommended for anyone who wants to understand not just how AI works, but how it should work—and what responsibilities we bear in shaping its future.
December 24, 2024
This book is not a light read, nor is it a page-turner; it’s a dense and well-researched text that refers heavily to other works by the author and others, as well as relevant government initiatives, especially those in Europe. The author starts by challenging the popular notion that artificial intelligence is actually intelligent, arguing instead that AI is a form of agency that can carry out tasks that would otherwise require our intelligence. He paints this decoupling of task and intelligence as part of a broader decoupling typical of our digital era (consider how our physical presence is now divorced from our mental presence, as entire groups of people stare at their phones). He goes on to cover the essential topics of the ethics of AI, from biased automation to criminal uses to the impact on climate change and other social issues. I recommend this book, just be prepared to put in the effort.
October 13, 2025
L'ho letto per prepararmi ad una docenza sull'IA applicata al contesto della psicologia giuridica. Floridi è un caposaldo dell'etica ed è in grado, con il suo scritto, di trattare in modo semplice ma approfondito un argomento così tanto spesso oggetto di malinterpretazioni.
Consiglio di leggerlo, soprattutto se non si conosce molto dell'argomento e si ha qualche remora relativa all'IA, cosa sia ed il suo utilizzo.
Consiglio di leggerlo, soprattutto se non si conosce molto dell'argomento e si ha qualche remora relativa all'IA, cosa sia ed il suo utilizzo.
December 29, 2022
Floridi io ti odierò per sempre.
July 14, 2023
Uno tra i libri più interessanti che abbia mai letto nella mia vita. Penso che farà da timone per il futuro.
December 6, 2025
Interessante la prima parte, poi troppo specifico per filosofi ed esperti in generale, non ha nulla di divulgativo
June 28, 2025
Muy completo y accesible. Lo mejor que he leído de filosofía y ética de la IA. Aunque tampoco es que hubiera una bara muy alta, pero lo agradezco. Muy socialdemócrata y con un claro espíritu europeísta regulador. Pero al menos acepta la existencia del coste de oportunidad, que ya es mucho teniendo en cuenta el panorama.
Displaying 1 - 9 of 9 reviews