IMDA and Microsoft Partner on AI Safety and Security Research

Singapore's Infocomm Media Development Authority (IMDA) and Microsoft have signed a Memorandum of Understanding to deepen collaboration on artificial intelligence safety and security — a move that positions Singapore at the forefront of responsible AI governance.

The agreement, signed on June 12 by IMDA Deputy Chief Executive Kiren Kumar and Microsoft Chief Responsible AI Officer Natasha Crampton, outlines three core areas of cooperation: technical research on AI safety, knowledge sharing on governance practices, and policy frameworks for responsible access to frontier AI models. The partnership will explore safety and security issues related to agentic AI, develop evaluation benchmarks and tools, and examine multilingual AI safety challenges.

This is not Singapore's first AI governance play — the city-state already operates the Singapore AI Safety Institute and has been an early adopter of AI testing frameworks through initiatives like Project Moonshot. But this MOU with Microsoft is notable for its specificity: the two organisations will jointly develop a white paper on how governments and infrastructure operators can access frontier AI models responsibly, addressing both demand-side requirements and supply-side policy considerations. The partnership also commits to sharing research findings with relevant ecosystem partners.

Why it matters for Singapore: As a small, open economy that bet early on becoming an AI hub, Singapore needs to balance rapid AI adoption with public trust. This MOU signals that IMDA is moving beyond high-level AI principles into the technical trenches — building actual benchmarks, tools, and evaluation methods alongside one of the world's largest AI model developers. For Singapore-based enterprises and AI practitioners, this means clearer guardrails and potentially faster access to frontier models under defined safety protocols.