Something we said? Don’t leave just yet!

For more information about latest events, news and insights, leave us your email address below.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form

AI holds the key to successful AML

AI holds the key to successful AML but its success rests on considered implementation

Napier AI
December 8, 2020

Artificial Intelligence has huge potential benefits for transaction monitoring, but how we introduce new systems is as important as the new systems themselves.

The advantages of AI technology can only be felt if it is implemented successfully. There are a lot of challenges to overcome, the key rests in a gradual, well thought out introduction, which allows for systems and staff to adapt and evolve together.

Old systems cannot be replaced overnight, and well-trained staff who are comfortable reading AI data is essential. There will be a lot of work to do, but the end result will be a far more effective and efficient transaction monitoring system.

AI and transaction monitoring: the capability gap

Financial institutions must identify what they wish to achieve by introducing AI into the transaction monitoring system. There is always going to be a gap between what the organisation wants to achieve through automation and what can be achieved from an internal, and often legacy, perspective as well as a regulatory perspective.

The Money Laundering Reporting Officer may wish to reduce the level of automation if they do not fully understand, or feel comfortable with, the technology. Lack of understanding was found to be one of the biggest concerns for regulated bodies to adopt AI technology.

Gradually introduce AI for transaction monitoring

One way to introduce AI and machine learning is to run an existing rules-based transaction monitoring system alongside a new AI-powered system. Doing this allows the old system to assist in training the new. It also enables financial institutions to challenge the traditional system, and simultaneously improve the performance of both.

Before a system that uses machine-learning can be fully deployed, it will need human input to fine-tune its initial ‘self-taught’ learning. The biggest challenge is that learning is not a ‘tick and forget’ exercise. Criminals are constantly changing their tactics in an attempt to remain under the radar.

For these reasons, most of a machine’s learning will occur when the system is live, as a result of humans reviewing the alerts generated by the system. When analysts feed back their knowledge into the machine, the machine is able to improve the accuracy and relevance of its alerts to reduce false positives and help avoid false negatives.

Getting the basics right for effective transaction monitoring

If there is one theme that keeps cropping up, it is the need to get the basics right before implementing AI. This process can take a significant amount of time and effort – but it will ensure the benefits of machine learning are maximised and in turn, provide a difficult to replicate competitive advantage.

Aside from organising data and training staff on how to use the system, much of the preparation for AI and machine learning rests in laying a solid foundation with transaction monitoring rules. It is really important to get the rules right as AI should initially complement rather than replace traditional systems.

Give compliance specialists superpowers

Advanced AI analytics to aid transaction monitoring

With the introduction of AI, the nature of a compliance specialist’s role will increasingly change from mundane data processing to highly investigative work as the AI automates the repetitive tasks involved in previously manual processes.

The introduction of AI-enhanced analytics will also give compliance specialists far deeper insights into customer behaviour and transactions. These analytics give the compliance specialist a view of the past, present and predicted behaviour of a customer, allowing the analyst to more easily identify unusual and suspicious activity. Having this intelligence available in easy-to-view graphics reduces the risk of illicit activity going undetected and simultaneously offers the compliance specialist even more control.

Use network analytics in transaction monitoring for deeper intelligence

The introduction of network analytics into transaction monitoring systems will help compliance specialists map a customer’s network producing a global view of their relationships including immediate and extended networks.

This will provide the compliance specialist with the ability to identify relationships and jurisdictions which may be prohibited or require further investigation in a way that legacy technology is unable to do.

Dynamic risk profiles enhance transaction monitoring

Determining a customer risk profile has been a long-established requirement by the regulators. However, a generic approach by definition is not effective in transaction monitoring as money launderers very quickly change their behaviours and activity once onboarded.

An AI-enhanced transaction monitoring system can produce dynamic customer risk profiles which are automatically adjusted based on the customer’s behaviour, transactions, and other intelligence.

Leveraging third party intelligence to complement transaction monitoring

Good intelligence plays an increasing role in effective transaction monitoring systems. By introducing intelligence data from third parties, analysts have access to a wealth of contextual intelligence to help make rapid decisions about changes in customer behaviour to identify suspicious activity.

Look to the future as AI advisors support compliance officers

As technology becomes more sophisticated, compliance specialists will have an AI Advisor by their side that continuously analyses patterns and learns to indicate unusual or complex insights and recommends areas of focus or action.

Pave the way to AI success with careful planning

AI technology offers many benefits to AML, however there are a lot of challenges that need to be acknowledged and overcome. Introducing AI successfully rests on gradual implementation, ensuring that old systems are slowly phased out and that staff are comfortable with the new technology. A realistic and sensible plan, executed with advice from highly trained experts, is the only way to overcome these hurdles and successfully implement AI. It is a large-scale project for any company to undertake, but the benefits of AI in transaction monitoring are undeniable, and in the long run will prove more cost-effective and time efficient than any legacy technology.

Transform your AML compliance with award winning technology

This article is an extract from our larger paper on Transaction Monitoring.

If you would like to read more about how cutting-edge technology could benefit you, download Successful Transaction Monitoring: Overcoming Challenges Through Collaboration co-authored with Lysis, or feel free to get in touch to see how we could help you with your AML needs.

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyse site usage, and assist in our marketing efforts. View our Privacy Policy for more information.