Singapore proposes UN framework for AI safety rules

Singapore proposed on September 27 at the UN General Assembly in New York the exploration of a UN framework convention for AI safety rules. Foreign Affairs Minister Vivian Balakrishnan stated that common rules are necessary as governments confront risks from autonomous systems, AI-enabled biological weapons or other weapons of mass destruction or disruption, and broader economic, social, and political upheaval.

Balakrishnan asserted that the UN's universal membership makes it the best-placed body to bring countries together on AI regulations. He suggested that governments could begin with scientific cooperation, shared testing and evaluation methods, and cross-border reporting of serious AI incidents, while ensuring all countries have a meaningful role.

Singapore supports continued investment in AI alongside safeguards such as rigorous testing before deployment, limits on autonomous systems, and mechanisms for human intervention. Balakrishnan added that people must remain in control and accountable for AI systems.

The proposal would align with an existing UN track on autonomous weapons, where talks have previously drawn calls for formal negotiations. In 2025, 156 states backed a UN General Assembly resolution on the matter.