AML solution of the year: SAS

The fight against money-laundering seems like an uphill battle. Money-launderers and terrorists are becoming more creative, often using technology to disguise funds obtained from illegal activities, such as drug-trafficking, corruption, tax evasion and fraud.

Adding to the pressure of sniffing out the bad eggs, financial institutions have to comply with increasingly stringent and growing anti-money laundering (AML) rules and regulations. While AML laws cover a relatively limited range of transactions and criminal behaviours, the implications can be far-reaching, often requiring financial institutions to evaluate their risks continuously.

Banks are increasingly relying on analytical and statistical methodologies to help meet these demands to reduce false-positive alerts and increase monitoring coverage, all the while trying to keep costs down.

Effective AML transaction monitoring strategies typically include segmenting the customer base by analysing customer activity and risk characteristics to monitor dubious activities more effectively.

Through its AML solutions, SAS has helped financial institutions achieve more than 90% model accuracy, reduce false positives by up to 80% and improve the suspicious activity report (SAR) conversion rate fourfold.

The solution leverages SAS Viya, an elastic and scalable cloud platform for private and public clouds. The platform allows firms to process huge data sets faster and operationalise complex analytics models, including artificial intelligence, machine learning and deep learning.

The solution also allows elastic processing, enabling quick experimentation with different scenarios, and the application of more sophisticated approaches to increasing amounts and speeds of incoming data.

SAS aims to relieve the burden for financial firms in terms of cost, resources and efficiency. Its solution helps financial institutions access and integrate any data regardless of source, size or speed, to provide firms with insights into suspicious activity detection.

Keith Swanson, director of pre-sales for Asia-Pacific, global security intelligence practice for SAS, says that, in the last few years, the financial services industry and regulators have been focusing on using advanced analytics, artificial intelligence and machine learning. As a result, SAS has worked to simplify the access and interoperation of these capabilities with its solutions.

SAS’s continued investment of 25% of its revenue into research and development has led to releases over the past five years, including the cloud-native SAS Viya, SAS Visual Investigator, and the extension of real-time monitoring and screening functionality for AML use cases.

In the latest version of its AML solution, SAS has applied advanced analytics and machine learning to offer a more intuitive and configurable investigation architecture. With improved alerts and case-management capabilities, it covers all aspects of the analytics life-cycle, from data to discovery to deployment.

It also added real-time screening capabilities, allowing firms to process thousands of transactions per second using streaming analytics to spot red flags. The solution can detect sanctioned entities, beneficial owners and their linkages in real time.

SAS also applies robotic process automation (RPA) to investigations, reducing case review times by 20–30%, and minimising manual errors.

One Asian bank turned to SAS to enhance its transaction monitoring system. The bank has also subscribed to SAS’s fraud detection solution and scenario analytics platform to tune scenario parameters in the AML transaction monitoring system. It is currently exploring further collaboration with SAS to use artificial intelligence and machine learning to reduce false-positive alerts and enhance operational efficiency.

Another bank in the region has been using the SAS AML solution since 2008. As the bank has increased its regional footprint, the SAS AML solution has provided it with the flexibility to cover the bank’s new products and transaction channels. Looking ahead this bank, too, will work with the SAS team to apply machine learning to identify suspicious activities better, prioritise alerts and reduce false positives.

Moving forward, SAS will further progress microservices and containerisation capabilities within SAS Viya, and embrace open-source capabilities.

“The use of these technologies will enhance our existing focus on data accessibility, data orchestration, user experience and ‘in-the-box’ advanced analytics that reduce time to adopt analytic methodologies for the best outcomes and customer experience while combating financial crimes,” Swanson says.

SAS AML, fraud and financial crimes compliance solutions are already present across multiple firm types and used by customers worldwide, ranging across financial services, communications, gaming, retail and government, among others.

Swanson adds that SAS is using its experience in helping firms mature their AML capabilities, reduce risks, including helping certain segments that have applied greater focus in remediating their financial crime risk in trade finance, insurance, securities and investments, and financial advice.

“SAS is continually focused on identifying how our experience working with such a diverse customer base can be leveraged across our customer base, as well as working with and monitoring focus areas from leading industry and regulatory bodies,” he says.

An Asia Risk judge says SAS’s submission demonstrates that it continues to apply its deep IP on analytics to tough industry problems such as AML. “Of particular note in this submission was their application of [artificial intelligence], [machine learning], RPA to tackle the problem of false positives and therefore reduce the human capital wastage on investigation time,” the judge says.