AI for Science: NUS leads cutting-edge research with 4 major AI-based projects to fast-track science and technology

An error occurred while preparing your download

NUS has secured four major projects under Singapore's S$120 million AI-for-Science Initiative (AI4S), reinforcing its position as a global leader in AI-driven scientific research. This achievement underscores the University's unique strengths in bridging advanced AI capabilities with world-class expertise across multiple scientific disciplines, such as advanced materials, computing, genomics and agriculture.

On 16 June 2026, Singapore officially launched eight inaugural projects under AI4S, a landmark national initiative spearheaded by the National Research Foundation to harness the power of Artificial Intelligence (AI) to revolutionise scientific discovery. Announced by Professor Tan Chorh Chuan, Singapore's Permanent Secretary (National Research and Development) at the AI4X Accelerate Conference 2026, these strategic research projects pair top AI researchers with domain experts from leading local and international institutions to spur research innovation in areas of interest to Singapore. This ambitious effort aims to nurture a new generation of "bilingual" scientists, fluent in both AI and fields like life science, materials science, and quantum science, so as to accelerate the pace of innovation.

Here, we outline the four NUS projects that are among the eight selected under the AI4S initiative.

Materials Data Foundry: Accelerating Synthesis of Complex Materials for Future Applications

The Materials Data Foundry (MDF) is a joint project co-led by Professor Sir Konstantin Novoselov from the NUS Institute for Functional Intelligent Materials (I-FIM) and Professor Alán Aspuru-Guzik from the University of Toronto's Acceleration Consortium to address the dearth of high-quality data in materials science. Using an open autonomous lab powered by AI and robotics, the MDF will create the world's largest dataset linking synthesis protocols to real-world material performance.

The lab will apply its platform to three testbeds: beyond-silicon and quantum-topological materials, durable oxygen-evolution electrocatalysts and corrosion-resistant high-entropy alloy coatings. The project also includes industrial partners like Nvidia and VeChain to tap on the cutting-edge digital solutions on the market. The dataset developed will fuel AI models to accelerate the discovery of new materials for electronics, clean energy, and sustainable infrastructure, bridging the gap from idea to industrial use.

Read more here.

AI for Program Reasoning

Co-led by Professor Abhik Roychoudhury from the Department of Computer Science in NUS School of Computing and Professor Cristian Cadar from the Imperial College London, and in collaboration with leading experts from the Singapore Management University, Massachusetts Institute of Technology, and ETH Zürich, this project addresses the urgent need to ensure software correctness and security as AI-generated code becomes increasingly prevalent.

The project will build advanced AI tools to automatically analyse, verify, and prove the correctness of computer programs to ensure that they are safe, secure, and work as intended. Employing formal reasoning, which involves proving using mathematical precision, and informal reasoning to understand the behaviour of undocumented code, the project will test its tools on critical systems like network protocols and components of the Linux operating system kernel. Ultimately, the goal is to create specialised AI agents that can help developers catch errors and reliably audit the vast amounts of code produced by other AIs.

Read more here.

Accelerating Genomic Research with Artificial Intelligence: From Data to Discovery

A joint project led by Professor Cheng Ching-Yu from the NUS Yong Loo Lin School of Medicine and his collaborators at A*STAR Research Entities (ARES), this project addresses the challenge of analysing vast and complex genomic data by developing MultiOmicsFM, a unified AI foundation model. Unlike existing AI tools that examine DNA, RNA, and gene activity in isolation, MultiOmicsFM will be designed to interpret them in unison, creating an integrated picture of an individual's genetic makeup. By leveraging Singapore's unique multi-ethnic genomic datasets, the project aims to expedite discoveries in disease risk prediction and mRNA therapy optimisation, positioning Singapore as a global leader in AI-driven precision medicine.

Read more here.

KGAI4Ag: Advancing Knowledge-Guided AI to Develop Agricultural Digital Twins for Singapore's Climate Resilience

Professor Roman Carrasco from the Department of Biological Sciences at the NUS Faculty of Science and his collaborators at the Illinois Advanced Research Center at Singapore Ltd. (Illinois ARCS) will be working together to tackle the threat of climate change on Southeast Asia's food security by building agricultural digital twins. These digital twins are virtual replicas of farmland, powered by Knowledge-Guided AI (KGAI), which uniquely combines data with established scientific principles of crop growth to create more reliable and interpretable simulations. The platform will deliver practical forecasting and decision-support tools to help farmers and policymakers optimise planting strategies, resource use, and supply chains, positioning Singapore as a regional hub for climate-resilient agricultural innovation.

Read more here.