Machine Decision is Not Final: China and the History and Future of AI

by Benjamin H. Bratton (Editor), Bogna Konior (Editor), Anna Greenspan (Editor)

Contributors: Blaise Agüera y Arcas, Bo An, Benjamin Bratton, Chen Quifan, Gabriele de Seta, Shuang Frost, Vincent Garton, Steve Goodman, Yvette Granata, Anna Greenspan, Amy Ireland, Bogna Konior, Vincent Le, Lawrence Lek, Lukáš Likavčan, Suzanne Livingston, Iris Long, Meng Bingchun, Reza Negarestani, Wang Hongzhe, Wang Xin, Xia Jia, You Mi.

An interdisciplinary, cross-cultural collection that decenters familiar narratives to provide a fresh perspective on what artificial intelligence is today, and what it might become.

Historians, media theorists, science-fiction writers, philosophers, and artists from China and elsewhere re-examine the nation’s intense engagement with AI, moving beyond the clichés that still dominate the contemporary debate.

Visions of the contested future of AI veer between common planetary goals and a new Cold War as culturally-specific models of intelligence, speculative traditions, and thought experiments come up against the emergence of novel forms of cognition that cannot be reduced to any historical cultural tradition.

This uniquely-positioned volume provides expert insight into this tension, using China as a touchstone for rethinking ‘artificiality’ and ‘intelligence’ as sites of difference in a way that is already present in the difficulty of precisely translating the Chinese term 人工智能.

Tracking the history of Chinese AI from the pre-Cultural Revolution to the post-Deng Xiaoping eras right up to contemporary debates surrounding facial recognition, the writers in this collection draw on a mixture of speculative thought experiments and cutting-edge use cases to offer singular views on topics including AI and Chinese philosophy, AI ethics and policymaking, the development of computational models in early Chinese cybernetics and the aesthetics of Sinofuturism.

Spanning borders between different worlds, histories, futures, and foundational models, Machine Decision is Not Final is not only a timely reappraisal of the stakes of AI development, but a tool for constructing more global imaginaries for the future of AI.

Genres: China, Philosophy, Nonfiction, Technology, Artificial Intelligence, History

272 pages, Unknown Binding

Published August 1, 2022

About the author

Benjamin H. Bratton is a theorist whose work spans philosophy, computer science, and design. He is Associate Professor of Visual Arts and Director of the Center for Design and Geopolitics at the University of California, San Diego. He is also Visiting Professor of Critical Studies at SCI-Arc (the Southern California Institute of Architecture) and Professor of Digital Design at the European Graduate School in Saas-Fee, Switzerland.

Reviews

[Farzad · Rating 3 out of 5]

MDINF is at its strongest when it sticks to things it can genuinely illuminate: the cultural weather around Chinese AI, the long shadows of Qian Xuesen and cybernetics; and the ways SF writers, artists, and internet users metabolize automation. The chapters on linguistic framing, early technoscience, and even the Guizhou cloud build-out provide a much needed assessment of Chinese “vibes” on ai at a deeper cultural level. I believe these were the more fundamental aims of the authors. It can however sometimes read as if ancient philosophy, Legalist metaphysics, or Daoist non-action are quietly steering how Chinese engineers build models today—I don’t think the labs at Tsinghua, Baidu, Alibaba, or SenseTime are subtly incorporating the Daodejing into their conception of AI while debugging distributed training. I wish it spent more time covering cultural navigation of forces shaping Sino-AI like compute shortages, provincial subsidy games, enterprise demand, and U.S. export controls; but I suppose that's not really the aim anyway. This is a beautifully written & culturally rich map of China’s imagination of AI, just not the map you’d want if you were trying to better understand how AI is currently being developed or governed there.

[Karson · Rating 2 out of 5]

many of these authors could probably produce interesting, insightful, novel work on AI if they actually read arXiv papers (or at least followed AI twitter), but unfortunately, it’s quite clear that they don’t. eg one chapter philosophizes its way into reinventing sandbagging and deceptive alignment. another extracts the idea of instrumental convergence (and kinda mesa-optimizers, or a conflation of the two) from Nietzsche while somehow also citing Bostrom’s discussion of instrumental convergence from Superintelligence and not realizing that this is the same thing. some of the reasoning is quite poor, and much of the writing is of the typical horrendously (unnecessarily) obscurantist critical theory type.
a few of the chapters have some real 干货 about eg Qian Xuesen and some current developments on the ground. would have been great to have a lot more of this.

[Adrian · Rating 4 out of 5]

A fascinating read that illuminates the differences Western and Chinese philosophies of AI, while highlighting some of the inherent contradictions in the discourse of the former. There are many thought-provoking chapters here - I particularly enjoyed the ones focused on language and the Dark Forest theory - although some may not be as accessible to readers not well-versed in Kantian philosophy. I also think that despite its broad scope, it does very little to address some of the issues that have been exemplified by the current rise of LLMs, as in the value of human work and creativity; instead it focuses more on questions of what intelligence really is and our relation to it. Therefore, it provides impulses to the current debate, but (perhaps purposefully) distances itself from it.