AIchemy Frontier Fund awards �ԹϺ��� researcher funding to advance AI-assisted materials discovery

AIchemy Frontier Fund supports research at �ԹϺ�� to accelerate AI-assisted discovery of next-generation materials

21 May 2026

Researchers from �ԹϺ�� and the University of Cambridge have secured £700,000 from the AIchemy Frontier Fund for a project using generative AI to identify new materials that can be tested in the laboratory.

The project titled: Alignment of Generative AI for Materials Discovery via Experimental Feedback brings together Professor from the Department of Materials at �ԹϺ�� and Dr from the University of Cambridge to develop AI tools that are directly connected to laboratory testing.

The collaboration will focus on improving how new materials are identified and validated, with an emphasis on ensuring that computational predictions can be realised experimentally.

Closing the gap between simulation and experiment

The project addresses a key limitation in current materials discovery, where most generative AI models are evaluated only using computer simulations. This means many predicted materials are never tested in the laboratory, limiting progress towards identifying materials that are both high-performing and practically synthesizable.

To address this, the team will connect AI models with experimental work, allowing results from the lab to directly inform and improve future predictions and helping to focus effort on materials that are more likely to be successfully realised.

Developing a more realistic approach to materials design

The research will focus on optoelectronic materials, using automated synthesis and measurement to generate experimental data at scale. By incorporating factors such as processing conditions and atomic-scale disorder into the models, the team aims to improve how accurately materials can be predicted and realised.

By combining modelling with experimental feedback, the project aims to identify materials that are both theoretically valid and experimentally achievable, helping to improve how efficiently new materials can be developed.

All data and outputs from the project will be shared openly in line with FAIR principles, supporting reproducibility and further research across the community.