NUS Secures Four Major AI-for-Science Projects Under S$120M National Initiative

The National University of Singapore has secured four of eight inaugural projects under Singapore's S$120 million AI-for-Science Initiative (AI4S), reinforcing its position as a global leader in AI-driven scientific research. The initiative, announced by Permanent Secretary for National Research and Development Professor Tan Chorh Chuan at the AI4X Accelerate Conference, aims to pair top AI researchers with domain experts to accelerate discovery across materials science, computing, genomics, and agriculture.

The four NUS-led projects span a remarkable range of scientific frontiers. The Materials Data Foundry, co-led by Nobel laureate Sir Konstantin Novoselov and Professor Alan Aspuru-Guzik, will create an open autonomous lab powered by AI and robotics to generate the world's largest dataset linking synthesis protocols to real-world material performance, targeting beyond-silicon electronics and clean energy catalysts. A second project on AI for Program Reasoning, led by Professor Abhik Roychoudhury, will develop specialised AI agents that can verify the correctness of code, including the vast amounts produced by other AIs — a critical capability as AI-generated code proliferates. The third project, Accelerating Genomic Research with AI, will build MultiOmicsFM, a unified AI foundation model that interprets DNA, RNA, and gene activity simultaneously, leveraging Singapore's unique multi-ethnic genomic datasets. The fourth applies Knowledge-Guided AI to create agricultural digital twins — virtual replicas of farmland that help farmers and policymakers optimise planting and resource use as climate change threatens Southeast Asia's food security.

Industry partners including Nvidia and VeChain are involved across multiple projects, alongside international collaborators from Imperial College London, MIT, ETH Zurich, the University of Toronto, and A*STAR. The initiative explicitly aims to nurture a new generation of 'bilingual' scientists fluent in both AI and their domain science, addressing a talent gap that has become a bottleneck for AI-enabled research globally.

Why it matters for Singapore: Winning half of the AI4S programme's inaugural projects validates NUS's strategy of embedding AI across traditional scientific disciplines. The initiative directly supports Singapore's ambition to become a hub for AI-driven discovery in high-value fields like advanced materials, precision medicine, and climate-resilient agriculture. For the broader ecosystem, these projects will produce foundational datasets, open-source tools, and commercially relevant intellectual property that strengthen Singapore's position in the global AI-for-science landscape.