The global politics of artificial intelligence By MaurizioTinnirello, BocaRaton London, New York: CRC Press Tylor & Francis Group. 2022. ISBN: 978‐0‐429‐44667‐2 (hbk), ISBN: 978‐1‐138‐31457‐3 (pbk), ISBN: 978‐0‐429‐44672‐6 (ebk)
本书由多位学者合著，从全球政治视角探讨人工智能政策，涵盖美国战略、联合国角色、自主武器、疫情与AI、劳动力替代等议题，适合AI领域的学者、政策制定者和从业者阅读。
The book The Global Politics of Artificial Intelligence consists of 10 chapters and was written by 17 contributors from various universities in the United States, United Kingdom, Austria, Belgium and Germany. The editor, Maurizio Tinnirello, is an expert in the field of Global Governance and Artificial Intelligence. The primary contribution of this book is to explore important issues regarding AI policy, which has an impact on every aspect of human life, from a global political perspective. The first chapter in the book, written by Mona Sloane, provides an in-depth review of the US strategy on AI with 10 cues that can help policymakers develop a more holistic approach to framing the relationship between AI and society, particularly in the economic realm. The author positions this strategy in a dichotomous situation between encouraging AI innovation and mitigating the potential dangers of AI. In Chapter 2, Inga Ulnicane and her colleagues reflect on some emerging international AI policy trends and ideas and identify three main AI policy frameworks. In Chapter 3, Eugenio V. Garcia highlights how the United Nations as an international organisation can help bridge gaps at the global level and provide a legitimate, representative and policy-oriented venue for deeper international cooperation on AI issues in the years to come. In Chapter 4, Alfredo Toro Karnavali discusses the regulation of the autonomous use of AI-based weapon systems such as Autonomous Weapon Systems (AWSs). Technically he assesses AWS's ability to comply with the core principles of distinction, proportionality and prevention of attacks in armed conflict as codified in the rules of International Humanitarian Law (IHL). The focus of Chapter 5, written by Seth Baum and colleagues, is to present some insights from global AI risk studies in such domains as: biotechnology, nuclear weapons, global warming and asteroid collisions. In Chapter 6, Mark Coeckelbergh raises the issue of the Covid-19 pandemic and reveals many existential and political vulnerabilities associated with AI. For instance, he highlights issues that lie at the intersection of the pandemic and the use of AI and data science. He also highlights issues regarding freedom, justice and equality in a global context which are also related to issues of democracy. In Chapter 7, Mahendra Prasad emphasises political cases and problems of analog AI and digital AI, integrating decision theory and value alignment to solve AI problems more effectively. In Chapter 8, Vassilis Galanos reports on how he engaged in a long-term interview-based research on people working in AI-related fields. He examines AI research practices, conducted largely in the United Kingdom (sometimes compared with practices associated with the United States and EU), using the conceptual metaphor of imperial and nomadic science. In Chapter 9, Thanasis Apostolakoudis comprehensively discusses the issues associated with replacing labour with AI. He identifies fears and hopes around the use of AI in the workplace using value and benefit analysis. In the final chapter, Maurizio Tinnirello discusses the development of AI and in particular the possibility of artificial general intelligence (AGI) being able to match human intelligence. The central point of this chapter is the approach and ideological structure (capitalism, democracy) for developing AI/AGI to be safe, ethical, and friendly to humanity. It is important to understand that while the book offers a wide-ranging set of perspectives about AI, it lacks sufficient practical examples. Although it provides a general outline of AI issues, there is a missed opportunity to connect the discussions to real life situations. Readers would find the book more valuable if the authors had included more case studies and scenarios that describe AI from various political perspectives in the discourse. For instance, in Chapter 6, there is a great need for case studies and more philosophical research regarding the psychological-existential and political aspects of the pandemic and AI in a global context. We highly recommend that each chapter include questions that can serve the purpose of encouraging readers to think critically about the subject matter, increase their understanding and create meaningful dialogue. Furthermore, it is notable that while Chapter 10 provides a critical view of the use of AI/AGI from the perspective of the ideology of democracy and capitalism, other equally legitimate political perspectives are not considered. For instance, it would be interesting to consider other ideologies such as liberalism, socialism and monarchism. The main cultural focus of the book is the United States and the European states, thus primarily representing Western perspectives. We strongly recommend that future editions could benefit from inviting authors from various other countries, such as parts of the Asia-Pacific, Africa and South America, to offer their perspectives and experiences on global politics in relation to AI. This will increase the variety and diversity of content for the next edition and will showcase global political perspectives from a wider variety of contexts. Following this approach will make the book more contextualised to various situations and so useful on both a local and a global scale. Meanwhile, scandals such as the Cambridge Analytica case, should add urgency to policy action in the field of AI. The efforts of international assemblies such as the World Economic Forum, the United Nations and the Organisation for Economic Cooperation and Development have brought additional attention to AI policy and facilitated cross-country learning in this area. However, AI policy development is not evenly distributed throughout the world and is only concentrated in the most advanced regions. Overall, this book is highly recommended as critical reading and a useful tool for academics, practitioners, and policy makers in the field of AI. Additionally, the interdisciplinary approach of the contributors to this book adds to its broad appeal for anyone interested in AI issues from a global political perspective. Chapter 2 persuasively invites readers to consider not only positive expectations but also negative expectations about new technologies, and provides an illustration that the policies currently being designed are not just about investing more in AI as a promising technology, but also planning retraining programs to deal with job losses due to automation. Furthermore, Chapter 5 comprehensively answers important questions covering various global risks. This chapter is of particular interest because it not only explains the risks of AI but provides an alternative view. If AI is developed safely and responsibly, AI can bring a number of benefits, including reduction of other global risks, thereby contributing to a better understanding of the global politics of AI and arguably making the world a better place. Overall, the many viewpoints included in this book provide a comprehensive understanding of the issues related to the development and use of AI, thus contributing to a better understanding of the global politics of AI, to how we can ensure that AI is suitable for society for the benefit of all humanity, and not just function as a political tool.