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

Three ways AI can take on money launderers more effectively

Money laundering in the UK is out of control. An estimated £100bn plus flows through the UK each year.

Julian Dixon
May 21, 2019

£100bn is laundered through the UK each year. AI can stop it.

Money laundering in the UK is out of control. An estimated £100bn plus flows through the UK each year. Yet recent reports show there hasn’t been a single prosecution for money laundering since new regulations were introduced in 2017. How is this possible?

Highly-fragmented efforts

The Treasury Committee chaired by Nicky Morgan concluded that the country’s AML efforts are ‘highly fragmentary’. It starts with a convoluted approach to AML. Different professions are overseen by different regulators, which causes confusion and potential duplication of effort. For example, there are multiple cases being investigated by both the FCA and NCA. It’s clear the left hand doesn’t know what the right is doing.

Everyone is responsible

With the regulators and government seemingly powerless to effectively take on the money launderers, it’s the responsibility of businesses and software providers to step up. Without concerted, focused and smart thinking, the criminals will continue to win.

Financial activity is hard to track and even harder to assess. There are many reasons for this but one of the key factors is found in AML software and the way rules are configured for transaction monitoring and screening. The challenge is to calibrate these rules to make sure you don’t end up with more false positives than false negatives.

Test and learn strategy

To make calibrating rules more effective, US regulators recently called a tech amnesty to encourage tests of AI-based AML platforms. This gives companies a chance to test AI analysis against historical data sets without fear of prosecution if the new data uncovers historical wrong-doing.

If the UK is to clampdown more effectively on criminal activity it needs to take a similar approach.Many big banks have already set up research labs to test AI-based technology.

To drive real change, AI needs to become the go-to solution.

With this in mind, here are three ways I believe a rapid shift to AI-based solutions, supported by a tech amnesty, would see Britain take the fight to financial criminals more effectively.

1. Reduce false positives

Legacy systems have been great at providing compliance officers with mountains of paperwork. It is their job to sift through all the suspicious transactions coughed up by computers.

The problem is that most of the transactions are perfectly innocent but every single red flag — and there can be hundreds of thousands produced in a matter of hours — has to be looked at individually by a real, live person. It’s not an impossible task. But it certainly is time consuming. Worse still, the sheer workload means that a true threat could go unnoticed for days if not weeks.

The reality is that the vast majority of the red flags produced by legacy systems should never have been raised in the first place – they are often merely a product of a system that is falsely matching data to incorrect criteria.

This needs to change. We know there are better ways to analyse transactions to detect suspicious behaviours. Modern technology such as AI is vastly more sophisticated and, if employed correctly, can reduce volumes of false positives dramatically.

Doesn’t that not just beg the question, but practically shout it? If modern technology can dramatically reduce the volumes of false positives; and throw up more accurate threat identification, surely it follows that not only will a larger proportion of real alerts be reported but also, it won’t take as long to review them all.

2. Put compliance in the control

Everyone and their dog has been laying claim to using AI. Recruitment firms, investment platforms, even dishwashers (not kidding).

AI of the kind we use isn’t so much a set of rules or the ability to create rules. It goes much further than that. AI-based platforms can sift through far more data and adapt without the need for further human instruction, which only slows things down.

Methods of laundering money change constantly. Launderers stay ahead of the authorities in every way – scrupulously adjusting their behaviours to fall within known search parameters to avoid being caught. Compliance officers are often in the dark about exactly what it is they are looking for; and with the vast and increasing numbers of transactions passing through bank accounts daily, they are on a clock to detect patterns of illicit transactions.

But with AI, searching for the ‘unknown unknowns’ becomes achievable. Machine learning algorithms can change the way compliance officers find patterns of suspicious activity.  And AI programmes can do this at the drop of a hat, testing and adapting thousands of times quicker than any human ever could.

Ultimately this puts the compliance officer in the driving seat with a chance to pre-empt a money launderer’s next move.

3. Improve Intelligence

The challenge doesn’t stop with legacy systems that currently spew out too many false positives and/or false negatives.

What happens after that is also somewhat worrying. Intelligence gathered from accurate analysis of transaction monitoring is right at the front-end in the fight against financial crime.

However, law enforcement agencies are showered with more reports of suspicious activity (SARs), many of these low quality reports, than they could ever deal with. The 2018 National Crime Agency report on SARs shows that, between April 2017 and March 2018, the NCA received 463,938 SARs alone.

The explosion in SARs could be driven by a desire by those who must comply with AML regulations to show they are trying detect wrongdoing. Add to this the fact that ineffective transaction monitoring systems tend to generate too many alerts (as we’ve already described), and the picture of over-reporting becomes clearer as multitudes of (false) alerts are investigated and consequently reported.

Although SARs are key to helping law enforcement fight financial crime, over-reporting hampers efforts all round. The current defensive reporting style is a significant burden on workloads and budgets for organisations and law enforcement alike.

AI will allow companies to strip back the redundant SARs with confidence, allowing analysts to direct their attention to the transactions more worthy of investigation.

It follows that fewer alerts with allow time for more thorough investigation. And consequently more accurate intelligence will be passed on which will put law enforcement agencies in a stronger position to take on the battle against financial crime.

It’s a virtuous circle of accuracy, efficiency and intelligence, meaning time is well spent for both companies and law enforcement agencies.


If you’d like to know more about how we can integrate AI into your AML systems and controls, please contact us. Alternatively, you can learn more if you book a demo.

Julian has more than 20 years of financial services experience gained at major investment banks including Deutsche Bank, JP Morgan and Commerzbank. His roles have ranged from front-office sales leadership to private equity. Julian has extensive knowledge of financial services processes and technology.
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.