In a recently published Viewpoint titled Artificial Intelligence in Corporate Banking, we discussed the increasing usage of machine learning in certain areas of corporate processes, including AP and AR to improve document match rates and so forth. So this particular news release is within that same sphere, and points to a specific use case around fraud management. One of the earliest deployed scenarios for machine learning is for both preventing fraud and effectively reducing false positives.
“We apply sophisticated analytical techniques to vast amounts of payments data to build models which identify suspicious activity. Every time a business pays an invoice, a behavioural signature is left behind. By analysing these signatures, and the signatures of historical frauds, we are able to identify and flag suspected incidents of fraud.”
The pre-requisite for successful use of machine learning is an ability to manage large data sets, which means implementing digital systems and processes. In this particular announcement, the bank seems to be directing the service more towards SMEs, which have historically been burdened under the weight of paper, exacerbating ongoing cash flow concerns.
“Detecting invoice redirection fraud is akin to finding a needle in a haystack, as there are tens of millions of legitimate non-real time payments every day. While the volume of fraud is relatively low, the values are typically large amounts, so the business impact of this type of fraud can be crippling.”
Overview by Steve Murphy, Director, Commercial and Enterprise Payments Advisory Service at Mercator Advisory Group
Read the quoted story here