OU gets £106k from UKRI for a project to assess AI-enhanced grant peer review

The Open University (OU) has secured £106,078 in funding to lead a new research project exploring how AI and large language models (LLMs) can transform the peer review process for research grants. The project, titled “LLMs Supporting Grant Peer Review,” will run for 12 months from October 2025 and aims to enhance the fairness and efficiency of funding decisions.

Peer review is essential to maintaining the integrity of research funding, but the system faces growing challenges due to rising application volumes and reviewer workload. This project will investigate how AI-driven tools using LLMs can support reviewers by providing accurate, efficient, and transparent assessments, without replacing human judgment.

The project was developed in the context of UKRI’s Metascience programme. The UK Metascience Unit was established in late 2023 as a joint initiative between the UK’s Department for Science, Innovation and Technology (DSIT) and UK Research and Innovation (UKRI). It was created in response to the Nurse Landscape Review with the aim to apply scientific rigour to how science itself is funded and conducted - improving efficiency, research quality, and policy outcomes. Building on OU’s strengths in AI, open science, and digital research systems, the proposal was successfully funded through the UK Metascience Unit. 

This is a three-university collaboration led by the University of Sheffield, with The Open University as the largest partner, contributing around 50% of the resources. The University of Salford completes the partnership, combining expertise in AI technology, research ethics, and funding policy. The OU’s team, led by Prof Petr Knoth and Dr Francesco Osborne, will focus on developing and testing LLM-based tools, while partners will explore governance, evaluation, and policy integration.

The project aims to show how LLMs can assist peer reviewers in reducing workload, improving turnaround times, while maintaining high standards of quality and fairness. By the end of the 12 months, the team hopes to provide a blueprint for AI-assisted peer review systems that could be adopted across UK funding agencies and beyond.

More specifically, the project will experiment with and assess the following scenarios: 

• Using AI to desk-reject low quality proposals, 
• using AI as a third reviewer, and 
• using AI as a meta-reviewer. 

The proposal will also explore agentic AI approaches where specific AI agents evaluate proposals with respect to specific quality aspects, such as rigour or originality. 

Success will mean demonstrating that AI tools can enhance, not replace, human peer review, ensuring scalability while safeguarding research integrity. If successful, this work could shape the future of funding evaluation worldwide.

The project is uniquely positioned to explore this potential, leveraging rare access to both funded and unfunded UKRI applications. This will allow us to rigorously simulate and analyse how different peer review scenarios affect funding outcomes.

Prof Petr Knoth commented, “Peer review in funding faces major challenges due to the time-intensive nature of thorough evaluation. Individual reviewers typically read only a subset of applications, meaning different proposals are judged by different reviewer groups - some stricter, some more lenient - raising concerns about consistency and fairness. AI presents a promising opportunity to support peer reviewers by providing tools that enhance decision-making while reducing the burden on human assessors.”

Dr Francesco Osborne said: “Peer review is essential for scientific progress, yet current systems face significant sustainability challenges. Our research will explore how AI can assist human reviewers in maintaining a process that is robust, fair, and scalable for the future. Crucially, we will focus on developing solutions that are both transparent and ethical, where AI serves as a supportive tool rather than a substitute for human judgment.”

This major funding underscores The Open University’s leadership in AI-driven research innovation, reinforcing its role in shaping the future of scholarly communication and research funding.