By Justin Newell

Money laundering and counter terrorist financing are pervasive crimes in Canada and worldwide. For its part, the Canadian parliament and regulatory agencies are continually looking for new ways to strengthen their financial crimes regulations. Reflecting this ongoing effort were the July 2019 amendments to the “Proceeds of Crime (Money Laundering) and Terrorist Financing Act” (PCMLTFA) and its related regulations. Among the amended requirements were those placed on reporting entities including financial institutions and service providers. While some of the new regulations went into effect immediately, most of them will be effective as of June 2021. These heightened regulations could not have come at a better time given that the COVID-19 pandemic has further emboldened criminals acting in these areas. This and the new regulations have prompted many regulated entities to assess their existing anti-money laundering (AML) and counter terrorist financing (CTF) processes and seek out new solutions for improving their monitoring. Specifically, they are turning to advanced technologies like hybrid artificial intelligence (AI), machine learning and fuzzy logic found in AML and CTF monitoring optimization software.

The response from government & industry
With the overall stewardship role for the Canadian economy, the Department of Finance Canada was the proponent for revamping the nation’s AML/CTF regulations. This came as a direct response to findings by the international standard-setting body, the Financial Action Task Force (FATF). It found that Canada’s AML/CTF regulatory regime had many vulnerabilities, particularly relating to transparency and law enforcement actions that were not commensurate with AML/CTF risks. The FATF further found that Canada’s primary anti-laundering agency, the Financial Transactions and Reports Analysis Centre of Canada (FINTRAC), provided authorities with financial intelligence that simply reflected disclosures responding to law enforcement personnel’s requests and nothing more.

Another significant problem was the length of time — generally several weeks — that it took for large Canadian banks to provide basic beneficial ownership information to law enforcement agencies. For example, records of transaction of potential proceeds from a crime often did not reach law enforcement for between 45 to 90 days, which sometimes contributed to this information becoming old and less useful for tracking down the proceeds of the crime and ultimately, prosecuting the crime. These findings contributed to the overhaul of the PCMLTFA.

For its part, Canada’s banking industry has been proactive in combatting money laundering and counter terrorist financing crimes. The industry was the first to voluntarily report suspicious financial transactions to the Royal Canadian Mounted Police and has been vocal in its support of the federal government’s AML/CTF measures. In addition, Canadian banks continue to allocate the necessary resources for reporting incidences/suspected incidences of financial fraud and working in collaboration with government and law enforcement to help detect, prevent, and prosecute related crimes. Still, banks and other financial service providers recognize that they must start deploying new tools and technologies that optimize their processes and better position them in their fight against financial crimes.

Hybrid AI, machine learning & fuzzy logic
To combat AML and CTF crimes, banks, insurance companies and payment processors are turning to holistic, enterprise-wide approaches essential for their compliance. Driven by advanced technologies such as hybrid Artificial Intelligence (AI), machine learning and fuzzy logic, these integrated solutions provide a holistic compliance process to optimally manage risk across all business lines and customer accounts. They arm businesses with an enterprise-wide view of customers and their portfolios, providing enhanced controls to fight financial crime in areas ranging from payment transactions, online and mobile banking, to credit card processing, ATM transactions, and mortgage processing, etc.

Integral to these advanced financial crime fighting platforms is their automated customer risk classification feature through which customers can be efficiently screened in real-time based on key indicators for estimating a specific customer’s potential risk to a financial organization. This on-boarding risk assessment is conducted in real-time against various watch lists compiled by regulatory bodies, as well as datasets prepared by commercial providers and internal lists. With this information, a customer can be assigned a risk level of high, medium, or low. To reduce the risk of money laundering and counter terrorist financing, the risk protection platform automatically reviews a customer’s behavior and will update his/her risk classification accordingly. Because watch lists continually change, it is essential that organizations remain vigilant and abreast of these list updates so that prohibited transactions can be quickly identified.

A prevailing problem for many financial institutions is the high levels of false positives. To combat this, advanced risk monitoring platforms automatically conduct customer screening against the various watch lists in real-time for new customers and periodically for existing customers and transaction counterparties. Built-in smart AI algorithms and false positive reduction mechanisms help keep the level of false positives low, while reducing related time and costs associated with investigations. When a positive match is identified, the platform creates an investigation alert which is assigned a score attesting to the strength of the match. The alerts are automatically displayed in real-time for analysis by another tool within the platform, the case management tool, which quickly generates required reports.

“Money mules & suspicious activity”
Today, a broad network of “money mules” are used to bring illicit funds into the system. Often recruited through social media, these individuals transfer money through various forms of money laundering crimes. Applying advanced AML/CTF solutions, their transactions can be evaluated in real-time. Through post-transactional monitoring, these transactions can be identified as suspicious based on the customer’s risk profile, and a profile of the customer’s relationships with other parties outside of the financial institution. The solutions also offer another layer of protection by providing out-of-box scenarios based on best practices designed to fight money laundering, terrorist financing and tax evasion crimes.

To further mitigate risks and improve their false positive rates, financial institutions are leveraging supervised and unsupervised machine learning. Supervised machine learning uses historical data to make predictions about a risk level associated with various red flag alerts and prioritize alerts for the next screening cycle. Unsupervised machine learning applies algorithms in AML Compliance to provide best practice classification model combinations that enhance an institution’s “Know Your Customer” and customer due diligence processes. This facilitates the detection of suspicious transactions that otherwise may have gone undetected.

Closing remarks
The United Nations Office on Drugs and Crime estimates that 2-5percent of the global GDP or $800 billion to $2 trillion in current U.S. dollars is money laundered each year. Clearly, it is the best interests of consumers, businesses, and the financial institutions and related companies serving them that money laundering and counter terrorist financing be fought with the most powerful arsenal available. There has been incredible progress in AML/CTF monitoring owing largely to the broader application of the most advanced technologies. By capitalizing on AI, Machine Learning, Fuzzy Logic, and proprietary algorithms incorporated into today’s most sophisticated software solutions, greater progress can be made in combatting financial crime.

Roy Prayikulam, Senior Vice President, Risk and Fraud at INFORM added, “The days for batch transaction analysis systems are numbered. Instant payments and PSD2 regulations require real-time decision engines in order to minimize friction. We are extremely pleased to work closely with EVO to help protect the organization and its merchants and to facilitate safer, faster payments using our real-time analysis technology.”

INFORM will begin the phased rollout of the RiskShield platform across EVO’s European operations later this year. Prayikulam concluded, “We are very enthusiastic about our recent selection by EVO Payments and look forward to building a strong partnership in which we can each learn from the other’s experiences on the international market.”

Justin Newell is Chief Operating Officer, INFORM Software, a leading global provider of optimization software for diverse industries

As financial networks move toward instant payments, so too must all ecosystem stakeholders and their transaction risk analysis platforms. Prior to choosing INFORM’s RiskShield, a risk assessment, fraud prevention and AML compliance monitoring solution, EVO Payments, Inc. (“EVO”), a leading global provider of payment technology integrations and acquiring solutions, administered a thorough search for a PSD2 ( a European regulation for electronic payment services) compliant, real-time transaction risk analysis (TRA) and monitoring platform.

“We conducted an extensive market analysis and concluded that RiskShield represented the best solution for EVO,” stated Darren Wilson, EVO’s President, International. “INFORM was responsive to our needs regarding PSD2 TRA exemptions, risk-based authentication, and instant payments. We are confident that INFORM’s ability to provide an agile solution will keep us prepared for any compliance or market changes that come within the rapidly evolving payments ecosystem. RiskShield’s intuitive front-end solution coupled with INFORM’s ability to rapidly deploy it also contributed to our final decision.”

Previous post

Bridging the EFT-ACH Chasm

Next post

Protecting Real-Time Payments