New research suggests that banks are squandering £2.7 billion (US$3.45 billion) a year by using outdated anti-money laundering (AML) systems that create ‘false positives’ – red flags against parties to or aspects of a transaction that turn out to be perfectly innocent.

The research by Fortytwo Data, which specialises in AML and sanction screening technology, suggests that this amount could be saved if banks upgraded to systems using big data and machine learning.

Research methodology

The research cited data from a WealthInsight/PWC study that reckoned spend by banks and firms in other regulated industries on AML compliance could hit £6.4 billion globally this year.

According to Fortytwo data, on average, 55 per cent of false positives and inefficiencies can be eradicated by the most modern systems. That equates to the £2.7 billion potential saving according to the research.

False positives

The research argues that the investigation and processing of false positives produced by legacy systems forces banks to retain many staff.

Fortytwo Data says it has analysed the likely impact machine learning and big data technology would have on the industry if it used first-hand intelligence on financial services clients and the success rate of its own platform in reducing false positives.

According to Fortytwo Data, it reduced one bank’s false positives by 97.4 per cent, though the technology firm reckons a reduction by a minimum of 20 per cent could be expected.

Trade finance

Regulatory technology, now frequently referred to as ‘regtech’ is gaining more attention from banks that see significant potential for new technologies to make compliance more efficient.

This is especially the case in the trade finance space, because of its paper-heavy processes and the ineffective use of transaction and counterparty data in risk identification.

Big data and machine learning platforms capable of analysing large data sets to reveal patterns, trends and associations can make the identification of red flags more reliable.