In recent years, the world has seen a proliferation of European laws and regulations in the banking and insurance sectors. This has, in turn, focused attention on money laundering, pushing financial intermediaries towards strengthening their safeguards.
The Financial Action Task Force (Fatf-GAFI) intergovernmental organization combats money laundering. Its work underlines the importance of data analysis through artificial intelligence (AI).
The use of data analysis tools is also becoming popular among anti-money laundering authorities. These bodies are using tools that increase their ability to detect networks of transactions, identify anomalous behavior and transform data into useful information.
The scenario’s complexity is amplified by new products, such sa cryptocurrencies, and new digital sales channels.
Machine learning detects suspicious transactions that occur both according to known patterns and according to new, apparently hidden ones. It can also automate the collection of “unstructured” external data and can, potentially, improve customer segmentation by identifying unusual customer behavior.
By applying intelligent classification criteria, segmentation can become more effective. This is because the monitoring system becomes active not only if certain thresholds are exceeded, but also manages the information processed by dynamic algorithms that define different risk thresholds.
More sophisticated algorithms allow the operator to give priority to the evaluation of alerts with a greater degree of probability of being linked to a suspicious operation.
Anti-money laundering control uses new tools based on AI and other technologies to verify, for example, the presence of subjects on sensitive lists, such as being subject to sanctions, terrorists, politically exposed persons or criminals. It can also activate continuous monitoring processes of suspicious transactions.
Among the most relevant tools supporting anti-money laundering analysis are deterministic models for calculating the risk profile by weighing subjective and objective risk. Subjective risk comes from the analysis of the characteristics of the subject. Meanwhile, objective risk is based on the analysis of customer operations. These are classified into various macro-areas and calculated on a vast series of risk factors.
There are also MTMs, transaction monitoring engines integrated with AI logic that reduce the false positives from the first analyses.
Meanwhile, network analysis allows people to see, in a clear and immediate way, the relationships between subjects. It does this by analyzing customer activity and drawing on external sources to retrieve further information, like company shareholdings.
In short, new technology can speed up analysis times, allow precise and timely identification of phenomena at risk of money laundering by reducing false positives and also highlighting complex phenomena that are difficult to detect, bring important benefits in terms of effectiveness and efficiency, and optimize the internal processes of the entities responsible for the financial transaction.