US‑China AI rivalry reveals competing governance models, power asymmetries, and the marginalisation of local voices in worldwide tech diffusion
Original framing: “US-China AI rivalry becomes a battle to export competing tech governance visions” — South China Morning Post
Empirical studies demonstrate that AI systems trained on data from the Global North perform poorly on African linguistic and socio‑economic contexts, reinforcing bias. Rigorous, open‑source benchmarking across diverse datasets is essential to avoid technocratic export of suboptimal models.
The US‑China AI contest is less a pure technological duel than a struggle over who sets the rules that shape data flows, labour markets, and environmental footprints worldwide.
Historical precedents of Cold War tech transfers, combined with Indigenous concepts of communal knowledge stewardship, reveal that current governance exports risk reproducing dependency unless they are co‑crafted with marginalized communities. Scientific evidence of bias, artistic‑spiritual critiques, and trickster inversions all point to the need for hybrid, culturally attuned frameworks. By institutionalising inclusive governance labs, transparent supply chains, and locally relevant capacity building, the power asymmetries can be rebalanced, turning AI from a tool of domination into a catalyst for equitable development.