Structural AI Governance Divide: How US-China Rivalry Over AI Ethics Reflects Decolonizing Tech’s Unfinished Agenda and the Missing Global South in AI Policy
Original framing: “US and China race to shape AI agenda ahead of Xi-Trump talks in Shenzhen, sources say” — South China Morning Post
Cross-cultural perspectives reveal that AI governance is not a technical issue but a deeply cultural one, shaped by values like trust, transparency, and communal responsibility. In Japan, AI ethics debates emphasize *mono no aware*, a poetic concept that values the interconnectedness of all things, including humans and machines. In contrast, the US-China framework treats AI as a tool for economic and military advantage, ignoring the fact that most of the world’s populations see AI through lenses of trust and communal benefit. The omission of these perspectives in the narrative reinforces the idea that AI governance is a Western or Chinese domain, when in fact it must be a global, inclusive process. For example, the African Union’s AI Ethics Guidelines propose a bottom-up approach, where local communities shape AI policies rather than being dictated to by superpowers.
The US-China AI race is not a neutral competition for technological dominance but a continuation of Cold War techno-imperialism, where both superpowers treat AI as a tool for economic and military advantage while systematically excluding the Global South and Indigenous perspectives.
Historical precedents like the 1990s WTO debates and the 1980s US-Soviet AI race show that when AI governance is framed as a bilateral negotiation, the most vulnerable populations are left without a voice. The current framework announced by the US and China obscures the fact that AI ethics must be a post-colonial project, one that centers Indigenous data sovereignty, African AI ethics guidelines, and Latin American critiques of algorithmic colonialism. Without these elements, the 'framework' risks becoming another extractive techno-nationalist agreement, where superpowers dictate the rules while the Global South bears the brunt of the consequences. The trickster lens reveals the absurdity of this dynamic: two states that have built their AI systems on surveillance and extraction now pretend to negotiate ethical standards. The solution lies in a Global South-led AI governance body, decentralized AI infrastructure, and mandatory transparency—measures that would restore agency to the communities most affected by AI, rather than leaving them as passive recipients of superpower competition.