The ten finalists of the Manchester Prize for AI breakthroughs have been announced. The winner will receive £1 million of funding in spring 2026 and will have to demonstrate a roadmap for adoption, scale and impact by 2030.
The prize rewards AI innovations made by UK-led teams which are designed to serve the public in three key areas. These are: delivering economic growth, upgrading public services and supporting an ethical transition to Net Zero. This year's prize is focused on projects which help public bodies meet clean energy goals.
Secretary of State for Science, Innovation and Technology, Peter Kyle emphasised: "AI is opening up transformative new ways to tackle climate change and support the UK’s ambition to become a clean energy superpower."
Energy Secretary Ed Miliband said that "clean power is the economic opportunity of the 21st century and these projects will help households and businesses take advantage of lower bills, in a smarter and faster way than ever before."
How it works
Funded by the Department of Science, Innovation and Technology and delivered by prize runners Challenge Works, the Manchester Prize is in its second year. The inaugural winners were Polaron who harnessed AI to develop materials used in wind turbines and electric batteries.
Each of the ten finalists received £100,000 of seed funding and £60,000 of compute credits to train and scale AI models. The finalists also benefit from investor readiness and a network of AI professionals.
Julia King, Baroness Brown of Cambridge, chair of the Manchester Prize judging panel highlighted: "the rapid advancement of AI means we have tools like never before to achieve the goal of decarbonising the economy while supporting individuals, communities and businesses to thrive."
Energy Secretary Ed Miliband detailed that "from specially designed radiator walls to a smart power grid that flicks on and off as we need, AI has the potential to help every home in Britain to feel the benefits of warmer homes and homegrown clean energy."
The projects
Agent Net Zero was developed by Sheffield University's Advanced Manufacturing Research Centre and analyses a comapany's environmental impacts 24/7 in real time. The tool identifies hotspots and automatically suggests improvements, providing actionable insights into reducing carbon footprints.
Biofuel AI automates biofuel production from feedstock recipes to strategies for long term acquisition boosting the efficiency of clean energy delivery. This project was developed by the University of Surrey.
Carbon Re was developed by its eponymous company counterpart as a joint project between University College London and the University of Cambridge. The project uses AI to control the processes in cement production, cutting costs and lowering carbon emissions.
Cavolo was developed by Kale AI and uses AI to make urban logistics greener. It helps businesses switch to Light Electric Vehicles and optimises delivery routes to make city deliveries more eco-friendly.
Deep.Optimiser-PhyX tackles carbon emissions in the steel industry using an AI-powered Digital Twin, a digital replica of the production process, which can more accurately predict temperatures and improve scheduling, increasing energy efficiency. This project was developed by Deep.Meta.
DRIVE or Deep Re-enforcement learning for Intelligent Vehicle and Energy optimisation was developed by Flexible Power Systems. It uses advanced AI Deep Reinforcement Learning to help large fleets of vans, buses and trucks to switch to electric by managing things like charging schedules.
EnergyWall by Underheat is a collaboration with the University of Salford. The project provides a smarter and more scalable way to decarbonise heating by using AI to analyse building designs and plan the installation of EnergyWall's brick technology, which turns bricks into radiators.
Green Loops uses AI to analyse photovoltaic (PV) cells from solar panels and assesses them for recycling into new panels to avoid e-waste. The project was developed by the University of Wolverhampton in conjunction with ABCircular GmbH Berlin.
Grid Stability was developed by the University of Manchester uses AI to monitor power grid stability to increase the reliability of the grid whilst clean energy solutions are upscaled. This is vital as there must be a balance between the energy going into and out of the grid.
Rapid Thermal Performance Assessment algorithms (RaThPAs) have been developed by Kestrix creates maps of the heat loss from buildings. AI and thermal drones are used to map heat loss across neighbourhoods allowing for energy upgrade planning with fewer costly site visits.