arrow_back_ios Back View more articles
A tool that ensures AI models used by banks do not return biased funding decisions based on an applicant's gender, ethnicity, or age has scooped a £20,000 prize at a competition for female-led start-ups at the University of Bristol.

Tool that keeps AI bias out of banking decisions wins university start-ups prize

A tool that ensures AI models used by banks do not return biased funding decisions based on an applicant’s gender, ethnicity, or age has scooped a £20,000 prize at a competition for female-led start-ups at the University of Bristol.

Dr Xiyue Zhang, lecturer in the School of Computer Science, told the Research to Reward competition judging panel that financial firms like banks rely on AI for 90 per cent of their core business functions.

But there is evidence that important decisions like loan approvals can be biased, providing poorer conditions on the basis of an applicant’s gender, ethnicity or age, rather than their credit history.

Dr Zhang’s solution is a tool that certifies when these models are fair and balanced, giving them a sound ethical and regulatory footing.

Her idea won her the audience vote prize.

Dr Zhang said: “Winning this award is a great honour and a crucial milestone in my journey toward research commercialisation.

“The scheme, including the final event, was an incredible experience – from the funding programme introduction session to expert-led training workshops and the inspiring pitches from fellow finalists.

“It has been invaluable in broadening my horizon beyond foundational research to real-world applications.”

A second £20,000 prize, awarded by an all-female panel of business experts, was awarded to Dr Isabelle Cuthbert, research associate in the School of Physiology, Pharmacology & Neuroscience.

She has discovered a way to improve the use of super-powerful ‘CLEM’ microscopes, which allow scientists to look deep into the complex interior of cells.

Substances used to help highlight tiny parts of these cells – known as ‘probes’ – currently do not work well with CLEM microscopes, limiting their usefulness.

Dr Cuthbert’s innovation changes that, making CLEM microscopes and probes function far better together. It will allow scientists to study cells like never before, potentially helping to bring about breakthroughs in drugs for Alzheimer’s and cancer.

Research to Reward was organised by the University’s Commercialisation teams, in a bid to help academics turn their cutting-edge research into businesses that benefit the economy and society.

The event was hosted by professor Michele Barbour, associate pro vice-chancellor enterprise & innovation at the University of Bristol, who has herself spun out research into a successful startup.

“This event showcases the extraordinary talent and ambition of female researchers,” she said.

“From healthcare to AI ethics, these innovations have the power to change lives. Supporting women in research commercialisation is crucial, and it was inspiring to see such a strong lineup of ideas and pitches.”

UK’s most powerful supercomputer launches in Bristol

Read more

28.07.2025

Students’ AI football footage innovation wins share of £72,000 startup funding

Read more

10.04.2025

Reach for the stars – astrophysics student spins out business recycling tech

Read more

21.02.2025

University of Bristol reveals £1.13 billion contribution to regional economy

Read more

17.02.2025

Scientists’ pioneering technologies win them Women in Innovation Awards

Read more

05.02.2025

Bristol named 12th most sustainable university in the world

Read more

10.12.2024

Bristol’s £10 million supercomputer goes online

Read more

09.12.2024

Next-generation space materials from Bristol blast off for International Space Station tests

Read more

05.11.2024

Business Biscuit
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.