Global South think tank maps AI divides at UN high-level week

The South Centre’s brief is a useful snapshot of how AI politics look from the Global South. While US and European debates oscillate between innovation and existential risk, many developing countries at the UN high‑level week framed AI as a tool for catching up economically, provided they are not shut out of compute, data and standards. At the same time, they echoed growing concern about frontier systems’ systemic risks and the fact that their governments have little say in how these models are trained and deployed.([southcentre.int](https://www.southcentre.int/policy-brief-163-5-october-2026/?utm_source=openai))

This tension matters for the AGI race because global governance will not succeed if it is seen as a club of a few powerful labs and regulators. The brief argues that unilateral export controls and voluntary lab pacts are insufficient, and calls for structures that give Global South states real voice in setting safety thresholds and sharing the economic upside. If that pressure crystallizes into new UN processes or South‑South alliances, it could reshape where frontier labs site infrastructure and how they structure access to powerful models.

In practical terms, the document is another data point that AI diplomacy is now a specialized domain. Labs that ignore these multilateral dynamics risk facing fragmented rules, data localization demands and political pushback when they try to enter emerging markets.