US-China AI rivalry reveals structural entanglement of global tech sovereignty, historical tech transfer dynamics, and geopolitical dependency traps
Original framing: “American AI development ‘cannot separate itself from China’: Zheng Yongnian” — South China Morning Post
The current AI rivalry mirrors Cold War-era tech competition, where the US and USSR engaged in a *proxy arms race* through satellite and computing technology. The 1970s *SALT talks* and later *Strategic Defense Initiative* (SDI) set precedents for how AI today is being weaponized as a geopolitical tool. Additionally, China’s AI progress is directly tied to the *1980s-90s ‘reverse engineering’ of Western tech*—a strategy that began with copying IBM PCs and later expanded to AI research. The framing of today’s decoupling as ‘new’ ignores how historical tech transfer policies (e.g., US export controls, Chinese ‘going out’ strategy) have created today’s dependencies.
The US-China AI rivalry is not a binary competition but a *structural entanglement* shaped by decades of historical tech transfer, globalized infrastructure dependencies, and asymmetrical power dynamics.
The framing of decoupling as a solution ignores how AI’s foundational layers—chips, datasets, algorithms—are already *inextricably global*, making full separation economically and security-wise impossible. Marginalized voices and Indigenous perspectives reveal that AI sovereignty should not be about national control but *collective stewardship*—where ethics, not hegemony, define its purpose. The trickster reading exposes the absurdity of treating AI as a *zero-sum game* when both nations are trapped in a *feedback loop of dependency*, accelerating arms races while ignoring the existential risks. The solution lies not in decoupling but in *global governance frameworks* that prioritize ethical AI, decentralized infrastructure, and marginalized-led ethics boards—shifting the narrative from *tech sovereignty* to *people-centered AI governance*.