Banks face calls for ban on ‘dangerous’ AI

Durham University research calls for a dedicated rulebook for AI in finance, with the highest-risk uses banned and tougher safeguards for systems that could affect customers’ rights, security and access to financial services

Banks could be banned from using the most dangerous forms of artificial intelligence under proposals designed to protect customers from biased decisions, data leaks and misleading information.

New research from Durham University Business School calls for dedicated rules governing the use of AI by banks and other finance firms, with restrictions determined by the potential risk to customers.

Under the framework, the most harmful uses of AI would be banned, while banks would face tougher rules when the technology could have a serious impact on customers. They would also have to make clear when certain content had been generated by AI.

Everyday technology posing little risk, such as AI-powered spam filters, would escape the stricter rules.

Professor Habib Ahmed, of Durham University Business School, proposed the system amid concerns that existing AI laws are too fragmented to deal adequately with the particular risks posed by its use in finance.

Professor Ahmed said: “AI laws are currently sparse and fragmented – but constantly evolving.

“While some countries have implemented AI laws – like the EU and China – others have taken a lighter touch, like the UK and the US.”

Under his proposed framework, which draws on the European Union’s AI Act, banned uses could include AI systems that manipulate or exploit people’s decisions, including the scraping of facial imagery.

Systems considered high risk would remain permitted but face stricter regulation because of their potential impact on people’s rights, security or access to vital services and data.

Lower-risk uses would carry transparency requirements, including telling customers when they were interacting with AI-generated content, text or video.

The study says stronger safeguards are needed because financial firms hold sensitive personal information and make decisions that can affect people’s access to money, credit and essential services.

Poorly controlled AI could expose customer data, reinforce bias in lending or risk assessments, create cyber security weaknesses or leave firms too dependent on outside technology providers.

The risks could also extend beyond individual customers, with the research warning that increasing use of AI could have implications for financial stability, consumer protection and financial integrity.

Professor Ahmed added: “The use of AI in finance is going to increase in the future and its impact is set to be extremely high.

“Finance affects everyone – from the day-to-day consumer to the global markets – and if we are risking data leaks, biases in data or misleading information due to AI, we are drastically affecting all aspects of financial life, both at the micro and macro levels.”

READ MORE: NVIDIA teams up with Wall Street giants in $500bn push to fund AI boom. Chipmaker joins forces with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR to finance huge new AI computing centres.

Do you have news to share or expertise to contribute? The European welcomes insights from business leaders and sector specialists. Get in touch with our editorial team to find out more.

_{Main image: Leeloo The First via Pexels}