The idea was announced by Foreign Affairs Minister Vivian Balakrishnan in Singapore’s national statement at the 81st UN General Assembly on 27 September.

Balakrishnan said AI poses three broad risks – autonomous systems that could act beyond human control, rogue actors who might use AI to create weapons of mass destruction, and the potential for wide‑scale economic, social and political upheaval.

He argued that only the UN, with its universal membership, can gather every country around common AI rules, because no other body has the same global reach.

A major obstacle, he warned, is the lack of trust between states as they compete strategically. Without confidence, any set of safeguards will be fragile.

To overcome this, Balakrishnan suggested concrete safeguards: rigorous testing and evaluation before deployment, clear limits on what autonomous systems may do, and built‑in mechanisms for human intervention when systems behave unexpectedly.

He added that human oversight must remain the final authority, ensuring accountability even as AI capabilities expand.