In an era when large language models can draft essays, pass bar exams, and hold conversations that feel startlingly human, one of the most pressing questions in the philosophy of technology is whether eighteenth-century critical philosophy still has anything to say. A newly published book review in the journal AI & Society argues that the answer is a resounding yes, pointing to philosopher Yuk Hui’s latest work, “Kant Machine: Critical Philosophy after AI,” released by Bloomsbury Academic in 2026 as a landmark attempt to reread Immanuel Kant through the lens of contemporary artificial intelligence. The review, written by Farman Ullah and Shihong Du of the College of International Studies at Southwest University in Chongqing, China, situates the book within a growing body of scholarship that treats machine intelligence not merely as an engineering achievement but as a philosophical event that demands a systematic conceptual response.
Hui, who holds positions in philosophy of technology and has emerged in recent years as one of the most influential voices in the field, is already known for a trilogy of ambitious works: “Recursivity and Contingency” (2019), “Art and Cosmotechnics” (2021), and “Machine and Sovereignty: For a Planetary Thinking” (2024). “Kant Machine” extends this project by returning to the founding figure of modern critical philosophy and asking what Kant’s architecture of judgment, reason, and understanding means for systems that appear to perform cognitive tasks without minds. The book, spanning 130 pages of dense argumentation, does not treat AI as a mere illustration of old philosophical problems. Instead, it inverts the relationship: machines become a test case that exposes hidden assumptions in Kant’s system and, conversely, Kantian categories become tools for diagnosing what contemporary AI genuinely is and what it merely simulates.
The title itself is deliberately provocative. It echoes a line of research stretching back decades, including Thomas Powers’ 2011 essay “Prospects for a Kantian machine,” which appeared in the landmark Cambridge University Press volume “Machine Ethics” edited by Michael Anderson and Susan Leigh Anderson. That tradition asked whether a machine could be built to act on universalizable maxims in the manner demanded by Kant’s categorical imperative. Hui’s project is different and, according to the review, considerably more radical. Rather than asking whether a machine can be Kantian, he asks whether Kant’s philosophy can survive contact with the machine, and whether critical philosophy as a method can be renewed, or perhaps completed, only after passing through the experience of artificial intelligence. The machine becomes a mirror in which the limits of human reason are redrawn.
A central conceptual anchor of the book, highlighted in the review’s detailed notes, is Kant’s notion of purposiveness, developed most fully in the “Critique of Judgement.” Kant argued that purposiveness can exist apart from any determinate purpose: we may not locate the causes of an organized form in a will, yet we can render its possibility intelligible to ourselves only by deriving it from a will. This distinction between purpose and purposiveness turns out to be philosophically explosive in the age of AI. Modern machine learning systems produce outputs, including text, images, and plans, that exhibit the appearance of teleology without any underlying intention in the human sense. Gradient descent does not will; a transformer model does not intend. Yet the forms it produces invite interpretation as if they were purposive. Hui’s book, the review suggests, uses precisely this gap to interrogate whether judgment, which for Kant mediates between understanding and reason, can be meaningfully attributed to artificial systems or whether it marks a categorical boundary that computation cannot cross.
The review emphasizes that Hui’s approach belongs to a broader critical reassessment of the intellectual origins of AI. Recent scholarship has begun to reexamine the philosophical commitments baked into the field at its founding, notably the program of John McCarthy, who coined the term “artificial intelligence.” Work such as M. M. Anderson’s 2024 article “AI as philosophical ideology: a critical look back at John McCarthy’s program,” published in Philosophy & Technology, argues that AI was never philosophically neutral but carried specific assumptions about mind, representation, and rationality from its inception. Hui’s rereading of Kant participates in this excavation, but with a distinctive twist grounded in his earlier work on cosmotechnics, the idea that every culture grounds its technical activity in a specific cosmology, a specific understanding of how the cosmos, morality, and technique hang together. Seen through this frame, AI is not a universal machine descending from nowhere but the expression of a particular technological tradition, one whose philosophical foundations deserve the same critical scrutiny Kant once applied to pure reason.
The book also engages a lineage that connects directly to the mathematical and engineering history of intelligent machines. Norbert Wiener’s “Cybernetics or Control and Communication in the Animal and the Machine,” first published by MIT Press in 1961, established the conceptual vocabulary of feedback, control, and communication that underlies much of modern computing. Hui’s earlier work on recursivity in “Recursivity and Contingency” traced how cybernetic recursion reshaped conceptions of nature and thought. In “Kant Machine,” the review indicates, this historical trajectory is brought face to face with the transcendental questions Kant raised: what are the conditions of possibility of knowledge, and can those conditions be instantiated in silicon? The convergence is not academic. As AI systems are increasingly trusted with judgment-like tasks in medicine, law, and governance, the question of whether they operate within anything like a critical framework, one that recognizes the limits of what can be known and willed, acquires practical urgency.
The reviewers situate Hui’s contribution within a contemporary debate about the capacities of large language models in particular. Recent work by R. Gubelmann, also in Philosophy & Technology, has argued from a Kantian-cum-pragmatist standpoint that speech acts, promising, commanding, asserting in the full normative sense, remain beyond the reach of large language models, because such acts presuppose forms of agency and normative accountability that statistical pattern completion cannot supply. This line of argument resonates strongly with Hui’s project. The review suggests that “Kant Machine” offers a systematic philosophical framework for such claims, moving beyond case-by-case assessments of model capabilities toward a deeper account of why certain capacities are tied to the structure of finite, embodied, judging beings. At the same time, the book reportedly avoids the easy route of declaring machines categorically deficient; instead it treats the machine as a genuine philosophical partner whose existence forces a reconstruction of critical philosophy itself.
Broader literature on the philosophy of artificial intelligence frames the stakes of this reconstruction. John Pollock’s influential 1990 article “Philosophy and artificial intelligence” in Philosophical Perspectives mapped how AI both draws on and challenges traditional epistemology, a dynamic that has only intensified in the decades since. More recently, V. Kondratenko and L. Slovianov’s 2026 article in Discovery Artificial Intelligence examined the interaction of artificial intelligence with the surrounding world from a philosophical standpoint, reflecting a growing recognition that AI is not a sealed computational artifact but an agent-like presence embedded in ecological, social, and epistemic environments. Hui’s planetary thinking, developed in “Machine and Sovereignty,” adds a geopolitical and ecological dimension to this picture: intelligent machines are entangled with questions of sovereignty, resource extraction, and the organization of the planet, questions that Kant’s cosmopolitanism, once radical, may need to be rethought to address.
The review by Ullah and Du, published on 03 September 2026, received on 01 May 2026 and accepted on 01 July 2026, presents the book as accessible to readers across philosophy, computer science, and science and technology studies, despite its demanding conceptual core. At xiv plus 130 pages, the volume is compact by academic standards, consistent with Hui’s practice of writing books that function less as surveys than as concentrated interventions. The reviewers note that the book’s argument rewards careful reading, particularly its engagement with primary Kantian texts, including the specific passages of the “Critique of Judgement” on which the discussion of purposiveness without purpose turns. For scholars of machine ethics, the book offers a counterweight to purely rule-based approaches to AI morality, suggesting instead that ethics must begin from an account of judgment that machines may formally approximate but substantively lack, or may possess in forms so alien that our inherited categories require revision.
For a general readership watching the AI debate oscillate between utopian and apocalyptic registers, “Kant Machine” arrives as a call for something more demanding: critical philosophy after AI, a philosophy that neither dismisses the machines as mere calculators nor capitulates to claims of machine minds. The review’s verdict, and the excitement it conveys, is that Hui has produced one of the first systematic attempts to carry the critical project into the computational age, treating the Kantian question, what can I know, what ought I to do, what may I hope, as newly urgent when the knower, the doer, and the hoper might no longer be only human. Whether machines have judgment, will, or purposiveness may turn out to be less important than the fact that asking these questions with the machine in view transforms philosophy itself. That, the reviewers suggest, is the book’s enduring contribution and the reason it is likely to be discussed across disciplines for years to come.
Subject of Research: A philosophical examination of Immanuel Kant’s critical philosophy, including judgment and purposiveness, in relation to contemporary artificial intelligence, as presented in Yuk Hui’s book “Kant Machine: Critical Philosophy after AI”
Subject of Research: Technology and Engineering
Article Title: Yuk Hui: Kant Machine: Critical Philosophy after AI
Article References: Ullah, F., & Du, S. (2026). Yuk Hui: Kant Machine: Critical Philosophy after AI. AI & SOCIETY. https://doi.org/10.1007/s00146-026-03235-y
Image Credits: AI Generated
DOI: 10.1007/s00146-026-03235-y
Keywords: Yuk Hui, Kant Machine, Immanuel Kant, artificial intelligence, critical philosophy, purposiveness, philosophy of technology, large language models, machine ethics, cosmotechnics, AI & Society, book review
Cite Scienmag News
APA MLA Chicago
Blake Davidson. (September 10, 2026). Philosopher Yuk Hui Rethinks Kant’s Critical Philosophy for the Age of AI. Scienmag. https://scienmag.com/philosopher-yuk-hui-rethinks-kants-critical-philosophy-for-the-age-of-ai/
Blake Davidson. “Philosopher Yuk Hui Rethinks Kant’s Critical Philosophy for the Age of AI.” Scienmag, 10 September 2026, https://scienmag.com/philosopher-yuk-hui-rethinks-kants-critical-philosophy-for-the-age-of-ai/. Accessed 10 September 2026.
Blake Davidson. “Philosopher Yuk Hui Rethinks Kant’s Critical Philosophy for the Age of AI.” Scienmag. September 10, 2026. https://scienmag.com/philosopher-yuk-hui-rethinks-kants-critical-philosophy-for-the-age-of-ai/
Copy citation Download RIS
Tags: AI and contemporary epistemologyBloomsbury Academic publications on AI philosophyemerging scholarship on AI andimpact of large language models on philosophyinfluence of Kantian thought on modern AI ethicsinterdisciplinary approaches to AI and philosophyKant critical philosophy in the age of AIKant critical philosophy in the age of artificial intelligenceKantian ethics and AImachine intelligence as philosophical eventphilosophical implications of AI-driven essay writing and legal testingphilosophical responses to AI advancementsphilosophy of AI and machine learningphilosophy of technology and AIphilosophy of technology and artificial intelligencephilosophy of technology and digital consciousnessplanetary thinking and sovereignty in AI discoursepost-Kantian philosophical responses to AIsystematic conceptual responses to AItechnological recursivity and cosmotechnicstechnological sovereignty and planetary thinkingYuk Hui’s reinterpretation of Kant
