Statement by Minister for Foreign Affairs Dr Vivian Balakrishnan at the UNSC Open Debate on AI and International Peace and Security, 25 September 2025

Mr. President,

Thank you, Mr. President, for this opportunity for Singapore. We are not a member of the Security Council obviously, but we are a tiny city state in the heart of Southeast Asia.

2        Artificial Intelligence is a very powerful platform which is generating a suite of tools. In the nature of powerful platforms, it can be used both for good as well as for harm.  Singapore believes that AI has multiple potential applications. First, in conflict prevention, it can help us with early warning systems. It provides tools to identify brewing crises before they erupt. It gives us, hopefully, precious time for preventive diplomacy. In peacekeeping, predictive analytics will help identify potential hotspots, optimise patrol routes, and enhance situational awareness, protecting both peacekeepers and civilians. For humanitarian operations, AI can certainly be deployed to optimise aid distribution, predict displacement patterns, and improve medical diagnostics in conflict zones.

3        But where conflict is concerned, there are two broad themes we need to be concerned with. First, we mentioned the advent of lethal autonomous weapons systems. Actually, it is no longer potential. If you think about the hot wars that are being fought right now, the truth is because of the urgency to identify multiple targets and to deal with them, human fingers, even today, are often not on the triggers. We have had to outsource it to AI systems. It is not something potential. It is something which is real.

4        That is at the tactical level, but there is also a strategic dimension to it. Deterrence and diplomacy, at least for the past eighty years, have depended on the fact that we assume that you are negotiating with another human intelligence with a brain and a heart. Concepts of deterrence are human inventions. If the strategic elements of diplomacy and deterrence are outsourced to AI systems, you are no longer dealing with another human being with inhibitions, with reservations, with an instinctive value for life. It will affect diplomacy, and it will, in the most egregious circumstances, affect deterrence and even mutually assured destruction from nuclear weapons amongst other things.

5        This is why Singapore believes it is vital to have guardrails in place.  AI has to be harnessed responsibly. In the case of Singapore, we have established national principles for AI in the military domain. These principles encompass responsibility, reliability, robustness, and safety, and guide our approach to developing, testing, and deploying AI-enabled military systems. Beyond national efforts, Singapore actively engages in international initiatives on AI governance, including the Responsible AI in the Military Domain (REAIM) process. We also recognise the importance of regional initiatives for inclusive and context-specific discussions on AI governance. We support the Association of Southeast Asian Nations (ASEAN) Responsible AI Roadmap and the ASEAN Defence Ministers Meeting Joint Statement on Cooperation in the Field of AI in the Defence Sector.

6        Today, as AI transforms the global security landscape, the question is not whether to embrace AI – it will be used, whether we like it or not – but how to harness it in a way that enhances international peace and security. This ultimately requires the Security Council and the General Assembly to work in partnership and in a complementary way, to build on the ongoing work in the First Committee on AI in the military domain.

7        We believe that the Security Council can play three important roles.

8        First, to actively explore possibilities for integrating AI capabilities into its own work and across the United Nations as a whole, including in conflict prevention, peacekeeping mandates and humanitarian operations.

9        Second, the Security Council should serve as a platform for confidence-building and the sharing of experiences and best practices through regular exchanges like today's open debate.

10      Third, the Security Council should take a broader view of what constitutes threats to international peace and security, and operate with the awareness that an AI-related incident could very easily create misunderstandings between states, miscalculations that can lead to escalations and outright conflict. We cannot rule out a role for the Security Council in coordinating responses, as part of its Charter-given responsibility for the maintenance of international peace and security in such circumstances, turbo-charged and catalysed by AI.

11      The international community stands at a critical juncture. As we saw in domains like cyber space and outer space, early, constructive and inclusive multilateral engagement on these emerging platform technologies is absolutely crucial. We must act decisively to ensure AI strengthens rather than undermines international peace and security.

12      Singapore commits to work with all member states towards a future that is safer and a future where AI serves our collective security and is not used as a tool for division or an accelerator for war.

13      Thank you.

.     .     .     .     .