In the recent “Shaping Digital Finance” session at the 20th Milken Institute Global Conference, Corrie Elston, chief technology officer, financial services, Google Cloud Platform, said that a financial services institution may experience 80% accuracy in detecting fraud and 99.5% false positives. As mentioned earlier, only about half of money laundering or terrorist financing incidents were detected by system alerts, to begin with. In essence, this means firms aren’t doing a good enough job when it comes to detecting fraud accurately, let alone discovering potential cases of it.
Based on several prototypes Mr. Elston developed, he was able to take the identification of money laundering from 80% to 99.4% accuracy. To put this in perspective, that statistic suggests there are almost no false positives. Imagine the savings, from cost to personnel to branding, that machine learning can bring to financial services institutions. That said, as of the publication of this guide, none of the unique prototypes Mr. Elston developed are in production. He says it is because companies are afraid to have the conversation with regulators, even though there are no regulations that say you can’t use artificial intelligence or cognitive computing for such a use case.
Digital transformation challenges in banking have been well understood and the strategies to address them simple and clear. However, it is becoming increasingly apparent that the industry is reaching a tipping point in the digital transformation journey.
Simply put, financial services organizations can’t afford to lose out on the opportunity to be nimble and master the detection of suspicious activity when the ability to do so is already here. Being able to leverage machine learning to combat fraud involves having anti-money laundering (AML) policies, practices, procedures, and training systems in place. It involves having a robust customer identification program (CIP). It involves having a monitoring program for suspicious or unusual activity that covers a variety of transactions and monetary instruments.
The future is bright when it comes to using artificial intelligence and cognitive computing for combatting money laundering. According to WealthInsight, a source of high-quality intelligence on global high net worth and ultra-high net worth individuals, “Global spending on AML compliance is set to grow to more than $8 billion by 2017 (a compounded annual growth rate of almost 9%).”
HSBC, Credit Suisse, Wells Fargo, and most other major financial institutions continue implementing intelligent systems and processes that help mitigate financial crimes and the risks they present. In some cases, banks have publically stated that making improvements to Bank Secrecy Act (BSA) and AML programs is their highest priority.
Our hope (and expectation) is that in the coming years we’ll see many success stories in the use of machine learning for most compliance initiatives, especially those revolving around AML and KYC.
In our new guide, we explore the basic tenets of AML and KYC, as well as how you can leverage the power of machine learning technologies to enhance your firm’s compliance programs. You can download it here.