News

AI could pose new risks to UK financial stability, Bank of England warns

The Bank of England is reviewing how artificial intelligence (AI) and machine learning is being used in financial services (Aaron Chown/PA)
The Bank of England is reviewing how artificial intelligence (AI) and machine learning is being used in financial services (Aaron Chown/PA) The Bank of England is reviewing how artificial intelligence (AI) and machine learning is being used in financial services (Aaron Chown/PA)

The Bank of England is reviewing how artificial intelligence (AI) and machine learning is being used in financial services, as governor Andrew Bailey admitted the technology must be embraced “with our eyes open”.

The Bank’s Financial Policy Committee (FPC) revealed it is looking more closely into whether the wider adoption of AI poses risks to financial stability in the UK.

AI and machine learning has been used by financial firms for at least a decade, such as to detect fraud and money laundering.

But adoption of the technology has become more widespread and its usage more varied.

This could pose “system-wide financial stability risks”, such as increasing cyber-risks or leading to more “herding” mentality, meaning when financial decisions are influenced by group behaviour, the FPC found.

“Given the rapid pace of innovation and potentially widespread use cases, the impact of AI and machine learning on financial stability needed careful monitoring and consideration,” the committee said.

The FPC is set to delve deeper into the possible risks to UK financial stability next year.

Governor of the Bank of England, Andrew Bailey, said it has to embrace AI “with our eyes open”.

“It is something that I think we have to embrace. It is a very important development, it has quite profound implications potentially for economic growth, productivity, and how economies are shaped going forward.

“The moral of the story is if you’re a firm using AI you have to understand the tool you are using, that is a critical thing.”

Firms are commonly using machine learning but are more in the “exploring stage” with generative AI, which refers to complex data models which can create something completely new.

“AI has tremendous potential as well, we mustn’t describe it entirely as a bag of risks,” Mr Bailey added.

The Bank, alongside financial regulators, are due to publish a consultation paper this month on the risks associated with “critical third parties”, referring to firms that provide data and AI models.