TDB: Financial crime remains an ongoing concern of risk managers across many jurisdictions. What are some key trends in financial crime that you are seeing, which merits closer attention and why?
Joel Lange: Advancements in technology and increased cross-border transactions have made the task of regulatory compliance even more demanding. The rise of digital transactions, global connectivity, and modern financial practices create new challenges that call for an enterprising approach to enhanced vigilance and adopting advanced strategies to effectively combat financial crime.
More than ever, companies today are beholden to escalating and ever-changing demands, and need to constantly adapt, improve, and update their processes and workflows to protect their businesses from risk.
Compliance teams play a key role in conducting traditional compliance checks as a good first line of defence against money laundering and terrorist financing. Negative news screening, for example, can quickly assess clients and prospects or monitor overseas business during onboarding (i.e. Russian connections that place increased sanctions and compliance risks on businesses).
Singapore’s recent money laundering case, valued at S$2.8 billion in assets, highlighted the critical importance of negative news screening. It is also extremely important for companies to identify ultimate beneficial ownership as criminals tend to set up shell companies with complex corporate structures for money laundering and predicate crimes.
In India, companies are looking to fortify their businesses and are actively pursuing technological advancements in compliance solutions with the approaching FATF audit in November.
Hong Kong’s progressive approach to develop the virtual assets sector by aligning with regulatory principles such as KYC, AML/CTF rules or customer due diligence is seen as a step in the right direction when the city’s Securities and Futures Commission (SFC) became one of the first and few regulatory authorities to introduce a comprehensive regime regulating a wide range of virtual asset-related activities.
Against this backdrop, I anticipate that we’ll start to see more and more organisations increasingly embrace AI as a tool to combat financial crime and shape their risk management strategies.
TDB: The RiskCenter Advanced Screening and Monitoring (ASAM) solution augments capabilities of financial institutions in combatting financial crime and mitigating third-party risks. How is ASAM differentiated from other AI based platforms that also seek to strengthen institutional AML efforts and ensure regulatory compliance?
Joel Lange: As compared to other AI-based platforms aimed at strengthening institutional AML efforts and ensuring regulatory compliance, ASAM stands out by going beyond simple name matching, resolving real-world identities in multiple scripts through breakthrough Natural Language Processing (NLP) and identity resolution technology. The tool minimises false positives when matching against client records by considering observed attributes like age, gender, nationality, and other biographical details. Advanced machine learning extracts valuable insights from both structured and unstructured data sources, reducing processing time and enhancing accuracy.
ASAM’s versatility shines through its intuitive and flexible design, making it a universal solution suitable for implementation by a wide range of organisations, from small and midsize enterprises (SMEs) and startups to global banks and corporations. It offers quick deployment as a Software-as-a-Service (SaaS) solution with support for multiple APAC languages and script capabilities. Additionally, it seamlessly integrates via an API into other platforms, enabling inclusion in broader automated workflows.
More importantly, ASAM brings together our best data assets, not just our proprietary structured risk data, but also our unstructured text from thousands of licensed news sources on the Factiva business intelligence platform, which provides an unrivalled selection of global news, data and information from 200 countries and 32 languages.
TDB: What is the significance of leveraging Ripjar’s machine learning technology in the development of ASAM and how is it uniquely contributing to efficiency in risk control?
Joel Lange: After an extensive review of technology on the market Dow Jones chose Ripjar for its unique ability to screen and monitor at volume against both structured and unstructured data. Leveraging Ripjar’s cutting-edge machine learning technology in the development of ASAM holds immense significance for enhancing efficiency in risk control. Ripjar’s Labyrinth Screening machine learning algorithms are a key asset in this regard, as they have the ability to process an extensive volume of media articles in over 20 languages every day. Their primary role is to discern and highlight the adverse media content that is relevant to your organisation’s risk assessment.
These Machine learning data classifiers excel in extracting essential entities from articles and determining the significance of each entity in relation to the article’s risk profile. The unique strength lies in the synergy between Ripjar’s top-tier technology and high-quality risk data. This synergy resolves the common dilemma faced by compliance teams, where they often have to choose between speed, quality, or coverage in their risk control efforts.
ASAM’s integration of cutting-edge machine learning techniques also empowers it to extract meaning and value from diverse structured and unstructured data sources. This combination of advanced technology and high-quality data is a game-changer in risk control as it not only reduces the time required for comprehensive risk assessment but also significantly enhances the accuracy of the results.
TDB: Generative AI is receiving considerable attention particularly for fraud detection and risk modelling and although it has gained wider adoption, doubts remain as to its efficacy and reliability. How can this be addressed and is there a case for Generative AI’s broader use in financial services or should it be deployed more selectively?
Joel Lange: Since the emergence of ChatGPT, the transformative power of generative AI has garnered significant attention in the financial services sector, particularly when it comes to fraud detection and risk modelling. However, in such a highly regulated sector, governments and organisations still have lingering doubts regarding its reliability – which calls for a cautious approach to its integration.
To tackle these concerns effectively, prioritising transparency and explainability is crucial. Generative AI models often operate as opaque “black boxes”, highlighting the need for methods that provide users with insights into their decision-making processes. Therefore, organisations operating in the sector should allocate resources to research aimed at enhancing the interpretability of these models.
Additionally, we need to ensure data quality and mitigate biases when using generative AI. The reliability of the technology hinges on the quality of its source data, necessitating a strong focus on the need for data quality assurance and the implementation of mechanisms to identify and rectify biases.
Ultimately, the decision regarding the broader deployment or selective use of generative AI in financial services lies in the industry’s ability to effectively address these concerns. Here at Dow Jones, we believe that striking a balance between innovation and risk management is important in fully realising the potential of generative AI, especially within the financial sector.
