UK financial services firms blame a lack of regulatory clarity and legacy systems for constraining deeper deployment of machine learning (ML) technology, according to a survey conducted by the Bank of England.
Brekaing it down, almost half of the 71 firms polled said there are Prudential Regulation Authority and/or Financial Conduct Authority regulations that hinder ML deployment.
A quarter of firms or 25% said this is due to a “lack of clarity” within existing regulation.
Aside from regulatory restrictions, legacy systems was top of the list as the greatest barrier to ML adoption and deployment.
Despite the challenges, almost three quarters of firms are using or developing ML applications, which are becoming increasingly widespread across more business areas.
This trend looks set to continue and firms expect the overall median number of ML applications to increase by 3.5 times over the next three years.
The largest expected increase in absolute terms is in the insurance sector, followed by banking.
The study also found that ML applications are now more advanced and increasingly embedded in day-to-day operations.
Around 79% of ML applications are in the latter stages of development – either deployed across a considerable share of business areas and/or critical to some business areas.
In addition, financial services firms are thinking about ML strategically. The majority of respondents that use ML have a strategy for the development, deployment, monitoring and use of the technology.
The study said that 80% of respondents that use ML say their applications have data governance frameworks in place, with model risk management and operational risk frameworks also commonplace for 67%.
As for benefits, firms point to enhanced data and analytics capabilities, increased operational efficiency, and improved detection of fraud and money laundering.
The study noted that respondents do not see ML, as currently used, as high risk.
The top risks identified for consumers relate to data bias and representativeness, while the top risks for firms are considered to be the lack of explainability and interpretability of ML applications.
“One of the reasons ML has not yet taken off at the trading level is because all the information banks need is stored across multiple systems and in different formats,” said Daniel Carpenter, CEO of Meritsoft (a Cognizant company).
He added, “Linking this information together to predict things such as trade fail rates is nigh on impossible without digitising and normalising the data and automating to ensure that data persists through settlement process.
Only once this has happened can banks even being to start thinking about predicting where the future costs could lie using ML.”
 ©Markets Media Europe 2022
