As data becomes increasingly complex, AI offers powerful tools to manage vast datasets, identify patterns and accelerate discovery. It’s reshaping how our scientists approach their work — and these are just some of the projects using AI to tackle major scientific challenges.

Haowen Zhao

Haowen Zhao at his desk with the laboratory he collaborates with behind him, taken by Nathan Pitt ©University of Cambridge.

Sormanni Group

Queens’ College

Haowen Zhao is a first-year PhD student in the Sormanni Group, where he develops AI models to design new antibodies to target diseases such as cancer and COVID-19.

He states: “Traditionally, antibodies were discovered by experimental methods, which are time-consuming and costly. I don’t do that – this is where AI comes in. It learns patterns quickly and generates examples of antibodies that could target a range of diseases.”

Haowen studied mathematics as an undergraduate, which gave him a strong foundation in computational skills. His work in the Sormanni Group is highly collaborative, relying on other members of the Group to test the antibody candidates proposed by his AI models.

Before Haowen joined the Group, other algorithms had already shown promise in targeting spike proteins on the COVID-19 virus. He hopes to have some potential drug candidates by the end of his PhD.

Niamh O’Neill

Niamh O'Neill with her salt simulation.

ICE (Interfaces: Catalytic and Environmental) Group 

Churchill College

Niamh O’Neill, a Gates Scholar and PhD student in the ICE Group, studies ions in water. One particular system she has studied is how salt (NaCl) dissolves in water as this everyday process is surprisingly not well understood at the atomic level. Niamh is using computationally efficient AI models trained to match accurate methods from quantum chemistry to simulate the dissolution process.

She explains: “Our simulations have shown that salt dissolves in two main stages. First, the ions gradually leave the crystal into the water. Then comes a crumbling point, where the crystal becomes unstable and breaks apart rapidly.”

Although dissolving salt may seem simple, it's a fundamental process involved in many crucial technologies, such as battery development. Conversely, removing salt from water is key to processes like purifying drinking water. The simulations Niamh is conducting have the potential to influence a wide range of essential areas — from energy storage to hygiene.

Samuel Brookes

Sam Brookes in front of his water simulation.

ICE (Interfaces: Catalytic and Environmental) Group / FAST (Frontiers in Atomistic Simulation Techniques) Group

Jesus College

Samual Brookes, a PhD student also in the ICE Group, trains AI to simulate environments of carbon dioxide (CO₂) and water molecules. 

For example, he has modelled how atmospheric CO₂ mixes with ocean water at the surface, revealing how it reacts to form molecules like carbonic acid. These insights help environmental scientists better understand how CO₂ affects ocean acidification and carbonate chemistry. 

Sam comments: “Our simulations are powered by AI models trained on complex and highly accurate data. These AI models can then run large and realistic simulations at a fraction of the time of traditional methods.”

Science thrives on collaboration and Sam works alongside other groups to supplement their practical findings. He recently submitted a paper titled ‘Unexpected oversolubility of CO₂ measured at electrode-electrolyte interfaces’ alongside the Forse Group looking at electrochemical CO₂ capture technology. The Forse Group utilises spectroscopic methods to study porous carbon materials for CO₂ capture After Sam has developed his AI models and run simulations, he can then give the Forse Group insights on their measurements and how the atoms in their materials are behaving. This back-and-forth enriches scientific research and AI plays an important role in speeding up the findings. 

From designing next-generation antibodies to simulating salt dissolution and understanding carbon capture, AI is proving to be a transformative force in scientific research. Haowen, Niamh and Sam each demonstrate how AI can accelerate discovery, uncover new insights, and support collaboration across disciplines. As these researchers continue to push the boundaries of what’s possible, their work shows how powerful the combination of human curiosity and machine learning can be in tackling some of today’s most pressing scientific challenges.