This week on Sinica, in a special episode recorded at the Davos On Air booth at the World Economic Forum's Annual Meeting of the New Champions in Dalian, I sat down with Angela Huyue Zhang, professor of law at USC's Gould School of Law and author of High Wire: How China Regulates Big Tech and Governs Its Economy, and S. Alex Yang, professor of management science and operations at London Business School. Angela and Alex — who are also married, and who arrived at this collaboration from opposite ends of the academy — have developed what I think is one of the most useful new mental models for understanding China's political economy: the platform state. Their argument is that we should think of the Chinese state less as a central planner or owner and more as a platform company like NVIDIA or Apple, one that builds architecture, sets standards, and governs an ecosystem within which fiercely competitive private firms fight it out. Value accrues at the system level rather than as firm profit, and the payoff may decide who wins the race to put AI to work across an entire economy. 5:37 – Three puzzles the framework is built to solve: profitless dominance in solar, EVs, and batteries; why the "grabbing hand" hasn't strangled Chinese innovation; and how China is attempting both zero-to-one invention and one-to-hundred scaling at once 7:57 – The platform state thesis: why the Chinese government behaves like a platform company, how nurturing an ecosystem of private firms solves the information deficit that cripples command-and-control, and why over-entry, involution, and consolidation are a repeated pattern — from EVs to the 140-plus humanoid robot companies operating today 16:13 – The aha moment: how a paper on the legal infrastructure of physical AI became the platform state idea over the Zhang-Yang dinner table, and whether this is a new species of political economy or the East Asian developmental state in new clothes 20:44 – State conditions: why state capacity and domestic scale are the two preconditions for the model, and why an ambitious Vietnam — which has the top-down capacity — may still find the Chinese playbook impossible to replicate 23:39 – Profitless dominance by design: harvesting versus extracting, the Uber analogy, overshooting as a control-theory strategy for nudging sectors, and how the anti-involution campaign and the 60-day supplier payment mandate show the state moderating the very competition it engineered 30:33 – Organized chaos: from the bike-sharing graveyards of the O2O wars to today's disciplined market, the exit of more than 400 EV makers since 2018, and why the survivors of China's "Premier League" of competition are now turning profitable 32:59 – The 3Gs playbook: growing markets by solving the cold-start problem, from Liuzhou's EV test drives to Beijing's green license plates, and how subsidy is only one lever among many 38:36 – Governing the ecosystem like Apple runs its App Store: why Beijing regulates generative AI with a light touch but physical AI is a different species entirely, law as the sixth layer of the AI stack, why robotaxis scale faster in China than in the U.S., and the state-convened standard-setting that's driving down humanoid robot costs 46:54 – Two flywheels: the familiar data-and-cost flywheel and the deeper state capacity flywheel, and how the National AI Fund's small but voting stake in DeepSeek aligns a complementor with the domestic stack — tilting the ecosystem toward Chinese chips 53:28 – Guarding the moat: automotive data rules and Tesla's stalled FSD ambitions, the unwound Manus sale, China's own small yard and high fence, and the closing provocation — that America could build the smartest frontier models and still lose the diffusion race to "artificial good-enough intelligence." Plus: the case for coopetition, and what policymakers should (and shouldn't) borrow from the platform state Links from the episode Angela's paper on law as the sixth layer of China's AI stack: * The Sixth Layer: The Legal Infrastructure for Physical Artificial Intelligence in China (SSRN) [https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6300241] Three Project Syndicate op-eds on the platform state idea: * The Rise of the Chinese Platform State [https://www.project-syndicate.org/commentary/china-innovation-model-could-give-it-an-advantage-in-ai-race-by-s-alex-yang-and-angela-huyue-zhang-2026-03] by S. Alex Yang & Angela Huyue Zhang * Are Government Stakes the Key to AI Sovereignty? [https://www.project-syndicate.org/commentary/government-stake-in-ai-us-and-chinese-approaches-by-angela-huyue-zhang-2026-06] by Angela Huyue Zhang * Overcapacity Is China's Biggest AI Advantage [https://www.project-syndicate.org/commentary/china-overcapacity-may-help-it-win-global-ai-race-by-angela-huyue-zhang-2026-01] by Angela Huyue Zhang Two Management Science papers on commercial platforms: * Crowd-Judging on Two-Sided Platforms: An Analysis of In-Group Bias [https://pubsonline.informs.org/doi/10.1287/mnsc.2023.4818] * Improving Dispute Resolution in Two-Sided Platforms: The Case of Review Blackmail [https://pubsonline.informs.org/doi/10.1287/mnsc.2022.4655] See Privacy Policy at https://art19.com/privacy [https://art19.com/privacy] and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info [https://art19.com/privacy#do-not-sell-my-info].