Financial institutions have for years banked on rules-based transaction monitoring systems (TMS) to root out money laundering and other financial crimes, only to be served up copious false positives that result in paralyzing inefficiencies, runaway investigation costs, and unseen false negatives that represent serious risk to the institution.
As TMS frequently flags the normal transactions of legitimate clients, false positive alerts trigger time-consuming and expensive human investigations. The industry estimates that approximately 95 percent of the alerts generated by TMS are false positives. Additionally, per industry estimates, up to 70 percent of financial crimes go unalerted and are therefore uninvestigated and unreported. These false negatives have the potential to cause significant risks to both the organization and their accountable employees.
QuantaVerse recently began working with a leading commercial bank looking to accomplish the following:
- Control costs while improving ability to drive suspicious activities out of its operations
- Determine what its TMS has missed in order to reduce future risk
- Improve its investigative processes by decreasing false positives and facilitating faster reporting
- Reduce risk without having to build out a huge, costly compliance investigation team
- Better identify and avoid bad actors so that it can take on more business without increasing risk
- Re-instate correspondent banking services it had de-risked years earlier
In working with the bank, the QuantaVerse AI-powered financial crime system was fed and then analyzed three years of entity transactional data as well as one year’s worth of transaction monitoring system (TMS) alerts. In that year, the bank’s TMS flagged hundreds of thousands of transactions. Investigators ultimately identified 560 alerted entities (in a total of 9,700 individual alerts) that they deemed suspicious.
While a majority of the bank’s TMS alerts were based on account activity (the number of transactions and the total dollar amount of the transactions) and jurisdiction, the QuantaVerse Alert Investigator solution analyzed the historical transaction patterns of entities to determine whether or not anomalous activity occurred.
The QuantaVerse solution was used to conduct a lookback analysis that examined transactions for indication of risk observables (such as related entities, economic purpose, adverse media, transaction beneficiaries and more). After a complete analysis, the solution was able to clear a staggering 70 percent of false positive alerts.
By automating the investigative process, the bank is confidently clearing alerts without time-consuming investigative efforts. With respect to the remaining 30 percent of false positive alerts, QuantaVerse is able to replicate much of the human investigative process that analyzes entities, transactions, and the intention of those transactions to reduce the bank’s AML case investigation time by an additional 75 percent. This is accomplished by enabling the bank to adjudicate the alert through the QuantaVerse Financial Crime Report (FCR). This AI-generated document frees up the time and talent of investigators so they can focus on the most complex financial crime risks.
QuantaVerse Financial Crime Reports include reputation, monitoring, and intent observables that the bank is using as a point of departure to accelerate their investigative activities. For each alert, an FCR helps the bank adjudicate a case in just 15 minutes whereas it might take many hours to complete an investigation without one. QuantaVerse FCRs are making it easier and faster for the bank to determine what cases needs to be investigated furthered and suspicious activity reports filed.
Editor’s Note: The availability of human investigators was one driver of the bank’s decision to automate their AML investigative process. Now that COVID-19 has suddenly made investigative resources scarce, its decision to automate is almost prophetic. To that end, QuantaVerse has adapted its technology for financial crime teams that are finding themselves short of investigators during COVID-related stay-at-home orders.
Serving almost 20 million customers, the bank was concerned about the risks associated with false negatives that its current AML compliance technology was missing. Intent on driving financial crime out of its operation, the bank began searching for a solution that could enhance its existing rules-based transaction monitoring system (TMS) and minimize the risk related to undiscovered financial crime.
While all agree the promise of AI for AML is great, taking the steps to select and implement AI within an institution is relatively new. So, what is the best approach for implementing AI and machine learning into your efforts to combat financial crime?
Over the course of the last two years, we’ve seen regulators and financial institutions make great strides towards thwarting more global crime through the identification of money laundering.