By PAUL O’DONOGHUE, Senior Correspondent
HARNESSING AI and behavioral analytics are crucial to building an effective real-time AML (anti-money laundering) program, according to a new report from IMTF.
The company, which develops anti-financial crime RegTech solutions, said the development is crucial amid the shift to real-time payments .
“As criminals exploit instant transactions and digital channels, traditional AML approaches are no longer enough,” the organization said in a newly-published report.
IMTF identified a real-time AML program as a system which can “detect and prevent suspicious activity before the damage is done”.
The report noted that real-time payments are projected to grow globally by 30% “year over year” to 2030.
It said that financial institutions must be able to develop an AML system which effectively identifies possible criminal activities in real time – not after a crime has occurred.
IMTF identified five key steps to develop such a system:
- Establish event-triggered KYC processes. This is to keep risk-assessments and monitoring aligned with the real-world risks.
- Shift transaction sanctions screening from batch processing to real-time screening.
- Implement real-time transaction screening programs. This is to prevent high risk transactions from being executed.
- Leverage Artificial Intelligence (AI) and machine learning. This is to detect known illicit transactions and patterns in real-time.
- Detect unknown suspicious and changing customer behavior.
Challenges
The report said that a ‘real-time’ approach to transaction monitoring presents two major challenges for compliance.
“From a technical perspective, financial institutions must have the infrastructure and systems in place,” IMTF said. This is to connect ‘data silos’ and better inform AI-models. However, it said the reality in many financial institutions is “far from ideal”.
“Customer data is scattered across siloed systems,” it said. The report added that this makes it difficult for financial institutions to “fully leverage the insights they already possess about a client”.
The other key challenge identified is the constant nature of real-time monitoring.
“When transactions are processed in real-time, compliance teams must also operate in real-time, meaning 24/7,” IMTF said.
However, it noted there are practical difficulties with this. These include labour laws which limit or prohibit regular weekend and night shifts.
Or challenges finding qualified staff in tight labor markets willing to work odd hours.
“One way to ease this operational challenge is to use Artificial Intelligence and automation,” it said.
“Machine learning models can be trained to replicate past decisions on alerts. [This] significantly reduces the volume of real-time alerts. [It] leaves only exceptional cases for manual investigation and intervention.”
The full report is available to read [HERE].
We hope you enjoyed reading this article
If you would like unlimited access to AML Intelligence premium articles, newsletter delivered twice a week, access to our Global Bank Fines and Penalties database, free access to Boardroom Series events and much more, select one of our subscription options and become a subscriber!
