Understanding Artificial Intelligence to fight money laundering

19 December 2017, by Bachir El Nakib (CAMS) Senior Consultant Compliance Alert LLC

Within the financial services sector, Anti-Money Laundering (AML) is a significant challenge for many institutions, often consuming large numbers of people and effort to manage the process and comply with the regulations.  As a result, these same institutions are looking for new solutions to help them reduce the burden and increase the controls in this complex space.   The combination of artificial intelligence (AI) and, more specifically, machine learning (ML), are increasingly being considered as enablers of a better solution.

Despite its potential, however, adoption of AI and ML within Anti-Money Laundering has been relatively slow.  This is due, in part, to the limited understanding of how AI and ML could be applied within compliance programs, and to the fact that regulators and compliance officers are often concerned that AI and ML are “black boxes” whose inner workings are not clearly understood.  Regulators typically require compliance officers to understand and validate not just the outputs, but also how the outcomes from AML models are derived.  Despite some of the concerns, we already see movement and application of these technologies.

Machine learning has been shown to be particularly useful in conducting suspicious activity monitoring and transaction monitoring, two key AML activities. A common challenge in transaction monitoring, for example, is the generation of a vast number of alerts, which in turn requires operation teams to triage and process the alerts.  ML can teach computers to detect and recognize suspicious behavior and to classify alerts as being of high, medium or lower risk.  Applying rules to these alert classifications can facilitate the automatic closing of alerts, allowing humans to supervise the machines that triage these alerts rather than reviewing all of the alerts manually, and making better use of the time of these experts.

Institutions leveraging ML can reduce their dependency on human operators to perform routine tasks, reduce the total time it takes to triage alerts, and allow personnel to focus on more valuable and complex activities.  There will always be a need for human involvement in the AML process; in fact, hybrid human/AI models and processes are the direction we see the function moving towards and should enable AML transaction monitoring to take a step forward in both the efficiency and effectiveness of alert operations teams.

To implement ML as part of a transaction monitoring solution, firms need to get key elements in place.  These include:

• High quality data.   All monitoring systems and analytics, not just ML applications, depend upon high quality data.  Static files such as Know Your Customer data as well as dynamic data on customer transactions held by financial services firms frequently have low completeness ratios in areas such as payment information, along with high error rates.  Profile refreshes, conducted as part of sales and marketing exercises, can update data while increasing customer outreach and identifying cross-selling opportunities.

• A 360-degree view of the customer.  Currently, financial services firms do not have the global freedom to share information about their customers to build a comprehensive network, and they do not formally collaborate on AML initiatives.  Regulators are, however, increasingly leaning toward data sharing between banks.  Over time, as ownership and privacy concerns are addressed, large amounts of transactional data could become available on intrabank data clouds, making a 360-degree view of the customer more feasible.

• Expertise in financial services and ML.  Very few people are experts in both ML techniques and financial services.  As a result, there have been fewer applications targeting financial services problems from start-ups and established vendors, limiting acceptance of ML within the sector.  Firms hiring ML experts can provide the needed financial expertise, if they institute appropriate training and development programs.

• Straightforward systems and processes.  ML is a relatively new technology and there are few established, straightforward processes to follow to implement it.  Without knowing what to look for, teaching systems to detect certain types of financial crime can be tricky.  For example, how does one teach a system to recognize terrorist financing?  There are more established processes for managing fraud, but nothing as comprehensive for terrorist financing, other than name matching against terrorist lists.

Financial services firms are making progress in addressing these challenges and their appetite for automation is increasing rapidly.  Many banks have started implementing business process automation in the form of Robotic Process Automation (RPA). In fact, robotics and AI/ML solutions can exist independently of each other and each can support the other’s capabilities.  Robotics can be used to train AI/ML models and AI/ML models can be used to add decision-making or reading capabilities to robotics models.

In Anti-Money Laundering, as in so many other areas of compliance, operations, risk and finance, AI and ML could be important steps in financial services firms’ journey to greater efficiency and effectiveness. These improvements in compliance and resilience capabilities can help to benefit firms’ shareholders and customers, make regulators’ jobs easier, and strengthen the global financial system. 

Suspected money laundering enterprises are in the headlines again, ensnaring the state-controlled Industrial and Commercial Bank of China Madrid, Spain  as well as  .

ICBC divided the funds being transferred to China between multiple accounts to ensure they didn’t exceed the 50,000-euro limit above which it would have had to declare to authorities, according to the summary. The bank, it added, failed to report any suspicious transactions as required under Spanish law regarding 78 of its clients. Detention and bail rulings for some of these customers show that they were already under investigation for suspected involvement in a range of crimes, including money laundering, tax fraud, bribery, extortion, forgery and smuggling.

According to the confidential court filings, police allege members of these suspected criminal organisations used fake documents such as falsified passports to open accounts at ICBC’s Madrid branch. ICBC employees are suspected of accepting falsified invoices to justify the origin of money transferred to China, police say in the filings.

If the suspects from the bank and the alleged smuggling networks go on trial, it is possible some might deny they participated in the bugged conversations. They might also dispute the conclusions authorities drew from the wiretaps.

Yet for financial institutions that want to identify, track and stop money laundering and other financial crimes, the technological tools of banking’s yesteryear no longer cut it. A new wave of tech tools can help banks and other financial firms discover dispersed and complex webs of illegal activity.

Increasingly, financial firms are turning to artificial intelligence and sophisticated network analysis to detect such criminal enterprises.

\AI Helps Banks Catch Criminals

The legacy tools banks use to find suspicious dispersals of money often rely on logical “if-then” rules to spot criminal activity, as Wired reports. For example, seven-figure sums of money traveling between foreign capitals is bound to raise a monitoring system’s alarm bells and the eyebrows of financial security experts.

However, those tools are often of little use for countering groups like terrorist organizations, which deal with small payments dispersed across several continents — transactions that might not raise any red flags in anti-money-laundering (AML) systems.

Seeking an alternative, financial firms are turning to artificial intelligence and machine learning— a subset of AI that uses algorithms to detect patterns, predict outcomes and potentially operate autonomously — to mine bank data and find anomalies. Banks can use behavior-based analytics to spot abnormal activity based on a customer’s profile, software firm Verafin notes.

“It’s a surgical approach to finding a needle in a haystack,” Dan Stitt, a director of financial crime analysis for AI company QuantaVerse, told Wired. Stitt previously spent two decades in the financial crimes industry and served at the Drug Enforcement Agency and the U.S. Export-Import Bank.

QuantaVerse has “developed the AI technology some of the world’s biggest banks use to identify money laundering, terrorist funding, and other financial crimes,” Wired reports, and “the technology has already helped identify a Panamanian man the DEA called ‘one of the world’s most significant drug money launderers.’”

While the use of AI in AML systems is still in its early days, there are high hopes that it can make such investigations more efficient, because U.S. banks spend billions of dollars every year on such systems to comply with the Bank Secrecy Act of 1970, which requires financial institutions to help government agencies spot money laundering.

According to a PwC report released in April 2017, 30 percent of large financial institutions and 46 percent of large financial technology firms are investing in AI.

“Machines are able to take in multiple additional data points and analyze those data points in a way that may not seem obvious to human beings,” Kevin Petrasic, a partner at the law firm White & Case, who specializes in financial regulation, told Wired.

Other financial firms are also using AI to combat fraudsters. PayPal, which had 197 million active customers in 2016 and processed 6.1 billion payment transactions last yearcombines a homegrown AI engine with its own fraud detectives to spot suspicious activity.

PayPal uses deep learning software that relies on neural networks to mimic the human brain and detect patterns. As a result, PayPal has cut its fraud false-alarm rate in half, Hui Wang, the company’s senior director of global risk sciences, told American Banker.

Banking giant HSBC announced a partnership in June with Silicon Valley–based AI startup Ayasdi to automate AML investigations that have traditionally been managed by thousands of human workers, Reuters reports.

Banks might be concerned, though, about the investments needed to adopt AI tech. “One concern is, how much change will you need to make to your existing infrastructure to accommodate the new technology,” Arin Ray, a financial services analyst who has studied the use of AI in AML systems, told American Banker. “If it’s large-scale change, human nature would prevent you from getting onboard immediately because there are large costs involved and you're not sure of how it will play out.”

Network Analysis Can Help Spot Financial Crimes

AI is not the only tool banks can use to spot money laundering, tax embezzlement, corruption and other crimes. Elise Devaux, marketing project manager at visualization software company Linkurious, writing on the data science forum AnalyticBridge, notes that graph technologies“provide exhaustive overviews of the different entities and their connections” and “support complex data queries on large data-sets in a near-real time environment.”

Shell corporations, tax havens, dispersed payments, seemingly random activity and complex financial schemes can prevent identification or tracking of money flows, she says.

“To thwart such criminal strategies, finding information about a specific suspicious entity is not enough,” Devaux says. “Financial crime units have to investigate the connections between individuals, accounts, companies, locations, to trace complex transactions.”

Banks usually keep track of numerous information sources on their customers and their financial activities, she says. Network analysis and visualization technologies allow them to “index complex connected data and easily query them to find patterns.” Such systems then allow financial firms to “compile various information into a single data model.” That makes it easier to spot suspicious activity.

Large firms such as IBM also offer financial firms analytics and visualization tools to help top money laundering and other financial crimes. IBM relies on techniques like context computing, predictive analytics, Big Data, and IBM Watson to help stop fraud, “The IBM AML solution layers analytical techniques to provide customers with the tools they need to combat threats and the control to evolve their defenses as their needs change,” the company says, arguing that an “integrated, flexible platform reduces redundant systems, processes, errors and, ultimately, operating costs.”



Download File