Other parts of this series:
As I discussed in the first blog of this series, innovation is changing the way banks look at analytics and the processes that support them in identifying and investigating financial crime . For example, institutions are automating manual and duplicative processes and using technology and analytics to increase the efficiency and effectiveness of anti-money laundering (AML) and know your customer (KYC) activities. The next phase may well be to turn the focus to specific financial crime risks.
One of the most promising fields of innovation is in the use of advanced analytics and machine learning (ML) in AML transaction monitoring. Advanced techniques, approaches and technology are helping financial institutions obtain positive business transformation outcomes including:
- Driving down false positives and reducing headcount;
- Improving the detection of suspicious behaviour patterns;
- Prioritizing critical and higher-risk tasks;
- Reducing and consolidating cases for further investigation; and
- Enhancing data availability and recommending cases for validation by manual investigation.
The key move is from simple, rule-driven detection and analysis to the use of machine learning, big data and networks, progressing from “rule-led” to “intelligence-led” monitoring. At present, AML is driven by individuals and is both reactive and static. It relies upon structured, internally-generated data and rule-based models to describe what is happening. Soon, however, AML should be network-driven, proactive and incorporating continuous learning. It is expected to use both structured and unstructured data and rely upon artificial intelligence (AI) rather than rules-based models to predict, rather than describe, suspicious behaviour. AI should still require human experts – for example, to guide the AI as to which features to search within – making the move to AI a marriage of machine intelligence with human expertise and know-how.
The traditional approach to detecting money laundering is to codify, as rules, well understood techniques (known as typologies) used to launder money. Typically, these typologies utilise very few data points from the customer’s behaviour (such as the rapid movement of funds), and lack a contextual understanding for the transactions. This often leads to the generation of many false positive alerts. Furthermore, typologies are rapidly changing (as demonstrated by the advent of cryptocurrencies) and traditional rules-based systems are not set up to accommodate these changing behaviours.
The advances in ML and AI allow us to gather information from previously difficult-to-access sources, such as free text notes, images and video, from both internal and external data sources to support prioritisation and detection of AML risk. Utilising this vast bank of data on customers, and their links and relationships with others, allows the integration of customer risk assessment under a single umbrella, breaking down the traditional silos of investigating customer risk through KYC, adverse media, transaction monitoring and other methods and evaluating the customer in the broadest possible context. This also enables detection of new or emerging patterns of risky behaviour — without the need to revert to rules – so that anomalies in behaviour are flagged to allow expert human investigators to assess and evaluate.
In the next evolutionary phase, neural networks and machine learning can lead to outcomes based on predictive analytics. For example, we can make potential improvements in the quality of compliance by both reducing false positive alerts and improving detection in new areas of risk, and drawing on algorithms from the world of ML and AI that can holistically monitor financial crime risk across all customers.
Eventually, AI should learn to detect new, hidden patterns of client financial crime from vast sources of internal and external data, using these findings to monitor, improve and support expert investigators.
Innovation in the form of advanced analytics and machine learning is helping institutions improve the quality and reduce the volume of alerts, improve global compliance and increase the efficiency of operations teams. In the next blog in this series, I will look at how innovation is transforming the process of adverse media screening.
This blog is intended for general informational purposes only, does not take into account the reader’s specific circumstances, may not reflect the most current developments, and is not intended to provide advice on specific circumstances. Accenture disclaims, to the fullest extent permitted by applicable law, all liability for the accuracy and completeness of the information in this blog and for any acts or omissions made based on such information. Accenture does not provide legal, regulatory, audit or tax advice. Readers are responsible for obtaining such advice from their own legal counsel or other licensed professional.
Accenture is a leading global professional services company, providing a broad range of services and solutions in strategy, consulting, digital, technology and operations. Combining unmatched experience and specialized skills across more than 40 industries and all business functions—underpinned by the world’s largest delivery network—Accenture works at the intersection of business and technology to help clients improve their performance and create sustainable value for their stakeholders. With more than 442,000 people serving clients in more than 120 countries, Accenture drives innovation to improve the way the world works and lives. Its home page is www.accenture.com
Copyright © 2018 Accenture. All rights reserved.
Accenture, its logo, and High Performance Delivered are trademarks of Accenture. This document is produced by Accenture as general information on the subject. It is not intended to provide advice on your specific circumstances.
If you require advice or further details on any matters referred to, please contact your Accenture representative.