* Keynote presentation during the 3rd Symposium of the Global Forum on Illicit Financial Flows and Sustainable Development, June 4th and 5th, 2024
In today’s rapidly evolving digital landscape, data is the cornerstone of effective anti-money laundering and countering the financing of terrorism and proliferation (AML/CFT/CPF) strategies. We have access to an unprecedented volume of information, and our analytical capabilities have never been more powerful. However, the true potential of these capabilities can be realised through cooperation.
The importance of cooperation
Cooperation is the linchpin that connects disparate data points, systems, analytical tools, financial authorities and law enforcement agencies, creating a cohesive and effective response to financial crimes. We should also consider the relationship between supervisors and supervised entities. Traditionally, this has often been seen as an oversight function, with a clear delineation of roles and responsibilities. However, in a digitalised world, this dynamic is evolving. Supervisors and financial institutions have aligned incentives to work together more closely than ever before, sharing insights, data, and strategies to effectively combat financial crimes.
The scale of the problem
Financial crime is often a network-based event. Perpetrators use a variety of methods to conceal the source and destination of funds. This includes moving funds through different accounts, payment methods, payment systems and jurisdictions, creating complex and sophisticated transaction trails that span across industries and borders.
The scale of the global financial crime epidemic is immense. In the 2024 Global Financial Crime Report, Nasdaq Verafin estimated that in 2023 more than $3.1 trillion in illicit funds flowed through the global financial system.
Among the most prevalent crimes fuelling trillions of dollars in illicit flows and money laundering activity were a range of destructive crimes, including an estimated $782.9B in global drug trafficking/DTOs, $346.7B of illicit funds linked to human trafficking, $11.5B linked to terrorist financing. Additionally, in 2023, fraud scams and bank fraud schemes totaled $485.6 billion in losses globally. These crimes’ social and human costs are remarkable and hard to quantify.
The U.S. Federal Reserve believes traditional fraud models fail to flag as many as 95% of synthetic identities used to apply for new accounts. It regards synthetic identity fraud as the fastest-growing type of financial crime in the U.S., costing companies billions of dollars every year. Banks themselves suffered reported losses of $442 billion to payment, cheque, and credit card fraud in 2023 globally.
Rising compliance costs and technological challenges
As financial crimes become more sophisticated, the cost and complexity of compliance with numerous regulations continue to rise. The annual True Cost of Financial Crime Compliance Study, released by LexisNexis Risk Solutions, reveals that financial crime compliance costs have escalated for 99% of financial institutions, reaching a staggering of $61 billion in the U.S. and Canada alone. Over the past year, 79% of organizations have seen increased technology costs related to compliance and know-your-customer (KYC) software. This surge in expenses underscores the incentives for financial institutions to better identify and prevent money laundering activities, and to collaborate with financial authorities and enforcement agencies.
The study also sheds light on the emerging use of cryptocurrencies, digital payments and AI technologies as tools for illicit activities. In the past 12 months, 22% of companies reported significant increases in financial crime involving cryptocurrencies, while another 22% noted a heightened use of AI. This is corroborated by the INTERPOL Global Financial Fraud assessment, which found that technology is “enabling organised crime groups to better target victims around the world.” Criminals, the report notes, are increasingly using deception strategies like so-called “pig-butchering scams”, which often involve exploiting human relationship vulnerabilities as well as investment scheming to defraud victims, and are in part fuelling an increase in human trafficking for forced criminality in call centres. The scams usually involve cryptocurrencies, which not all law enforcement agencies have the capacity to trace.
These developments highlight the growing challenges faced by traditional supervisory techniques. Traditional methods often rely on static, rules-based approaches that struggle to keep pace with the dynamic and evolving nature of financial crimes. As a result, supervisors are finding it increasingly difficult to detect and prevent sophisticated fraudulent activities effectively. Conventional models often generate high false-positive rates and miss meaningful insights from the data collected, leading to inefficiencies and missed opportunities to catch real threats.
The need for more adaptive and advanced supervisory approaches is further underscored by the alarming growth of human trafficking fraud, particularly in the Americas. INTERPOL’s report states, “The use of Artificial Intelligence (AI), large language models and cryptocurrencies combined with phishing- and ransomware-as-a-service business models have resulted in more sophisticated and professional fraud campaigns without the need for advanced technical skills, and at relatively little cost,” .
In December 2023, for example, hundreds of victims across South and Central America were lured through messaging apps and social media platforms with the promise of rewarding jobs, only to be forced to work in call centres to commit financial fraud crimes. Such fraud is typically carried out by networks of individuals who are either highly coordinated or loosely connected, INTERPOL’s report states. In 2023, scammers stole over $1 trillion from victims, according to the Global Anti-Scam Alliance.
The World Economic Forum’s Global Risks Report 2024 includes adverse outcomes of frontier technologies, cyber insecurity and illicit economic activity as short- and long-term threats to the global economy.
These findings emphasise the need for a holistic and collaborative response to curb the mounting challenge of financial crime, to eliminate safe havens for the perpetrators, strengthening the intelligence of the authorities to detect and prevent financial crimes across different sectors and countries.
The opportunity of supervisory technology (suptech)
Having the right tools that augment the ability of the authorities to collect and analyse information is fundamental. Suptech presents a significant opportunity to address these challenges and shift towards more effective risk-based approaches. By adopting suptech solutions, supervisors can more away from rigid, rules-based frameworks and towards more adaptive models, better suited to the complexities of modern financial crimes.
There are several initiatives underway that leverage suptech to progress global AML/CFT efforts:
- In 2023, the World Economic Forum launched the Cybercrime Atlas initiative to better understand the cybercriminal ecosystem. The Atlas aims to squeeze the space in which cybercriminals operate and map the cybercrime landscape, covering criminal operations, structures and networks.
- New research published by the Bank for International Settlements (BIS) shows that graph or network-based machine learning analysis significantly improves the effectiveness and efficiency of anti-fraud and money laundering controls implemented by financial authorities. The BIS Innovation Hub’s Project Aurora explored the effectiveness and efficiency of different approaches to AML monitoring across three scenarios. The scenarios included analysis of synthetic transaction data at the individual financial institution level, national level and cross-border level. The effectiveness and efficiency measures refer to the ability to detect money laundering activity while keeping the number of false positives low. Under each scenario, graph-based machine learning models detect twice as many money launderers than the traditional, mostly rule-based supervisory frameworks. Graph-based machine learning models also outperform all other approaches researchers tested, such as isolation forest, logistic regression and artificial neural networks. Graph-based machine learning models also reduce the number of false positives compared to a rule-based approach at the individual financial institution level by between 40-85%.
- The Hong Kong Monetary Authority (HKMA) has started a collaborative pilot using network analytics to detect mule account networks and help disrupt movements of fraud proceeds. The HKMA system is designed to analyse transactional data, user behaviour, and network patterns in real-time. Leveraging advanced analytics and machine learning algorithms, the system identifies anomalies, suspicious patterns, and potential threats within the vast network of instant payment transactions.
- De Nederlandsche Bank (DNB) utilises advanced data analytics to assess AML/CFT risk and control frameworks of supervised entities comprehensively. DNB’s approach includes a network analysis tool to trace fund transfers to high-risk jurisdictions and a machine learning-based tool for scrutinising banks’ internal control frameworks, which enable the building of detailed risk profiles for financial institutions.
- The Financial Transactions and Reports Analysis Centre of Canada (FINTRAC) has developed a heuristic model that evaluates the effectiveness of financial institutions’ control frameworks. This model integrates natural language processing and a risk- scoring model with an adaptive learning approach. This multifaceted platform empowers financial authorities to detect and address money laundering and other related financial misconduct more efficiently.
- The Monetary Authority of Singapore (MAS) uses analytical tools to scrutinise unusual transactions, reduce manual data reviews, and employ predictive models for misconduct risk. Also, MAS is developing COSMIC, a digital platform for collaborative sharing of money laundering/terrorism financing (ML/ TF) information, in partnership with six major banks in Singapore. Focusing on risks like misuse of legal persons, illicit trade finance, and proliferation financing, COSMIC will allow financial institutions to share information on suspicious customers, enhancing their capacity to detect and deter financial crime.
These developments, which are explored in greater detail in our annual State of SupTech Report, underscore the transformative impact of suptech in AML/CFT/CPF supervision. More specifically, supervisors have placed significant emphasis on using suptech solutions across different supervisory areas, with AML/CFT/CPF supervision ranking as the third-highest priority, identified by 59% of supervisors. This strategic shift towards more interconnected and data-driven financial supervisory practices highlights the essential role of advanced technologies in enhancing effectiveness and efficiency of financial crime prevention.
Having the right tools that augment the ability of the authorities to collect and analyse information is fundamental. Suptech tools, with their advanced data analytics capabilities, play a crucial role in transforming raw data to actionable insights. Equally importantis the ability to share intelligence, data and code between financial supervisors and law enforcement agencies, demonstrating the power of effective partnerships in fighting financial crime. This requires a paradigm shift in how we view and manage data and code, moving from a mindset of ownership to one of stewardship. By viewing data and code as a shared resource, we can enhance our collective ability to detect, prevent, and respond to illicit financial activities.
The Cambridge SupTech Lab provides a suite of digital tools, capacity building, technical assistance, and research to develop and scale suptech solutions, to enable collaboration that yield tangible results, and to accelerate the digital transformation of financial authorities.
Together, we can make a meaningful impact in the fight against illicit financial flows.
