Morozov rightly calls for us to turn away from the seductive AI narrative of replicating human capabilities in autonomous machines, toward a rich older tradition of cybernetics and Deweyan pragmatism, which instead imagined a world where machines connect people to collaborate and self-govern more adroitly. He also draws on lost history to project this struggle onto the left-right political divide, looking, for example, to Latin American radicalism for inspiration. While my experience as Digital Minister of Taiwan lacks the romance of Salvador Allende’s experience in Chile, perhaps the pragmatism I sought to apply suggests a more consensual path toward Morozov’s ambitions.
My work was deeply grounded in the mainstream history of modern computing. After all, while cybernetics did inspire coup-subverted socialist experiments and hippie communes, it was also the primary influence on at least three of the most consequential and successful technological and management trends of the postwar era: personal computing, the Japanese manufacturing miracle, and the internet. Both leftist martyrs and Silicon Valley tycoons have fancied themselves the rebellious heroes of cybernetics. Yet its achievements arguably owe much more to the duller work of scholar-bureaucrats like J. C. R. Licklider (known as “Lick”) and W. Edwards Deming, who moved seamlessly across business, government, and the academy—a network Morozov would surely label the “military-industrial complex.”
So-called AI technologies may well come to shape all our lives, and we must do everything we can to put humanity’s hands on its steering wheel. Yet the tools that have thus far driven the digital age have much less of the logic of instrumental efficiency and human alienation that Morozov rightly critiques. In Lick’s words, personal computers offer “man-computer symbiosis” rather than artificial general intelligence. Meanwhile, the miracle of the Japanese kaizen method was built on Deming’s insight that empowering line workers to understand full production processes would both enable them to continuously improve quality and avoid replacing them with or transforming them into machines. As for the internet, its packet switching, hypertext, and Deweyan form of collaborative, standards-based governance offer a powerful substrate for a startling range of interactions without making us slaves to the premature optimization that computer science pioneer Tony Hoare identified as the “root of all evil.”
Of course, the reign of these contrasting paradigms may be in decline as the internet and personal computing have increasingly become cogs in the machine of AI-powered, centralized digital platforms. Yet when Lick foretold this tragic turn even at the occasion of the internet’s birth in his visionary 1979 essay, “Computers and Government,” he pinned the cause on precisely the sort of anti–military-industrial agitation that Morozov celebrates.
As program officer for the Information Processing Techniques Office, Lick had harnessed Department of Defense funding through the Advanced Research Projects Agency (ARPA) to jumpstart the funding of some of the earliest computer science departments, including Douglas Engelbart’s Augmentation Research Center at Stanford, and connected them through the ARPANET that grew into today’s internet. While he empathized with the anti–Vietnam War sentiments that fueled the Mansfield Amendments’ prohibition of defense-funded basic research, he saw clearly how the abandonment of networking by the U.S. government would allow corporate monopolies to dominate and stifle digital innovation.
Forced by these strictures to focus narrowly on the performance of weapons systems, ARPA turned, as Morozov bemoans, to a narrow logic of efficiency. This change is even symbolized in the rebranding of the organization to D(efense)ARPA. As Morozov recounts, this played into the hands of a funding-hungry AI community, which was, perhaps ironically, dominated by the work of pioneers like John McCarthy and Marvin Minsky—who, far from being neoliberals, were advocates of AI-powered utopias far beyond what Soviet planners thought possible. The turn we have seen in the West toward both AI and cryptocurrencies in the last decade and a half may be seen as the ripple effects of this change in priorities, just as the internet and personal computing revolutions were the ripple effects of Lick’s foundational investments.
In a recent book, ⿻ 數位 Plurality: The Future of Collaborative Technology and Democracy, E. Glen Weyl and I, in collaboration with dozens of leaders across industry, research, and government, insist that reports of the death of cybernetics have been greatly exaggerated. Its roots were planted much deeper in Asia than in the West. Deming brought participatory production to the core of the lives of millions of people in Japan, after all, and Dewey’s extensive travels in China between 1919 and 1921 made his pragmatic and democratic theory of education a foundation of Taiwanese land reform and education.
Thus, while AI and crypto, and the critiques like Morozov’s they inspire, dominate the tech discourse of the West, a more hopeful and consensual narrative is playing out half a world away. Japan’s Miraikan National Museum of Emerging Science and Innovation eschews the performative and apparently useless robot dogs of Boston’s Museum of Science in favor of playful and caring assistive technologies. In India, the publicly funded, openly interoperable Agri Stack—part of the broader India Stack initiative building up the country’s digital public infrastructure—has brought public services and payment systems to more than one hundred million unbanked farmers untouched by the cryptocurrency craze. And Taiwan’s burgeoning digital democracy has manifested Lick’s injunction from half a century ago: “Decisions about the development . . . of computer technology must be made . . . in the interest of giving the public itself the means to enter into the decision-making processes that will shape their future.”
This is a tradition worth learning from and building on. Yet it is not one that fits easily into the limited political narratives of our time. In Japan, India, and Taiwan—often celebrated by the U.S. political mainstream as strong allies—technological innovation is deeply integrated with a traditional, often religious, social fabric. This is foreign to Western narratives in which the secular eschatology of existential risks and social justice politics are the primary checks on corporate AI ambitions. Yet it may offer a path to more fully free us from the traps the current digital society has laid for us.
AI is not just threatening a disconcerting future; when applied to maximize engagement with polarizing content and thus advertising revenue, it is already warping our ability to see one another and the world around us. The most effective act of rebellion may thus be to transcend these incentives to reinforce existing divides by reaching out across cultures and ideologies to forge a future we want to inhabit together.
