Department for Science, Innovation and Technology: Manchester Prize

Overview

Up to 10 finalist teams will initially receive £100,000 each, a generous package of non-financial support, and access to free compute to develop their innovation. One finalist will win the £1 million grand prize in early 2025.

The challenge statement

The first Manchester Prize will be awarded to the most innovative and impactful AI solution which demonstrates social benefit by overcoming challenges in the fields of energy, environment and infrastructure.

Solutions could include:

• reducing energy costs for consumers by using AI to model household energy use and identify targeted interventions, such as retrofitting and replacement
• supporting emergency service response by bringing together a range of spatial data about the road and built environment to improve last mile routing
• improving the response to extreme weather conditions by using AI and earth observation data to predict areas vulnerable to flooding, or to support better real-time spatial data of events such as wildfires and flash floods
• reducing disruption to public services through predictive modelling of infrastructure resilience, with automated scheduling of maintenance, such as deploying teams to fix potholes or other traffic obstructions
• enhancing food security by using earth observation and soil data to monitor and improve farming productivity and crop yield
• improving efficiency and reducing resource consumption in manufacturing by using AI to optimise or automate energy-intensive processes
• We encourage solutions that demonstrate advances in technical capabilities such as generalisation, uncertainty quantification, interpretability, data-efficient AI and physics-based AI.

Winners will be judged according to the criteria of innovation, impact, long-term viability, feasibility, and safety and ethics.

Eligibility Criteria

Challenge prizes are designed to be open to a wide range of organisations and individuals. What matters is the quality of your solution – not your profile or track record.

For your team to be eligible to enter the Manchester Prize, you must meet these requirements:

• Eligible entrants: Entries can come from individuals, companies and other types of organisation (e.g. non profits, charities, research and technology organisations). In the case of individuals, the lead entrant must be at least 18 years of age.
• Consortia entries: Teams may enter as consortia of any combination of the above, but must nominate one individual or organisation as the lead entrant. The lead entrant will enter into contracts and receive funding from the Manchester Prize. Organisations and individuals other than the lead entrant can be based outside of the UK.
• Geographical scope: The lead entrant must be based in the United Kingdom, and be able to receive funds into a UK bank account in the name of the lead entrant.

In addition to these requirements, teams will be subject to due diligence checks, and teams which are made up of individuals will need to have registered a legal UK entity and have a business bank account in place prior to receiving any finalist award.

You may submit as many entries as you wish, but only one entry per lead organisation or individual will be selected as a finalist.

Judging criteria

Up to ten finalists will be selected at the end of the entry phase, and one winner of the grand prize will be selected at the end of the finalist phase.

At both points, the successful teams will be selected based on five judging criteria. These are equally weighted.

The judging criteria are:

• Innovation: Teams should demonstrate how their solution is an innovation in artificial intelligence (AI), compared to the current state of the art. This may be innovation in the underlying AI, or in a novel application of an existing AI approach, or both (innovation in business model, marketing or service design is not taken into account).
• Impact: Teams should clearly define what impact they aim to have and why they expect this to arise from their solution. The impact must relate to the challenge statement, but it does not need to align with any of the example use cases set out.
• Long term viability: Teams should articulate why there is a credible path to adoption (commercial or non-commercial) for their solution, and what their plan is to pursue it.
• Feasibility: The team should show how their scientific and technical approach is appropriate and how the team has the capacity to deliver a working prototype by the start of 2025.
• Safety and ethics: The team should show they are taking action to showcase best practice in responsible AI.