nach oben

2023 | Buch

Kapitel lesen Erstes Kapitel lesen

Artificial Intelligence Applications in Banking and Financial Services

Anti Money Laundering and Compliance

verfasst von: Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Verlag: Springer Nature Singapore

Buchreihe : Future of Business and Finance

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book discusses all aspects of money laundering, starting from traditional approach to financial crimes to artificial intelligence-enabled solutions. It also discusses the regulators approach to curb financial crimes and how syndication among financial institutions can create a robust ecosystem for monitoring and managing financial crimes. It opens with an introduction to financial crimes for a financial institution, the context of financial crimes, and its various participants. Various types of money laundering, terrorist financing, and dealing with watch list entities are also part of the discussion. Through its twelve chapters, the book provides an overview of ways in which financial institutions deal with financial crimes; various IT solutions for monitoring and managing financial crimes; data organization and governance in the financial crimes context; machine learning and artificial intelligence (AI) in financial crimes; customer-level transaction monitoring system; machine learning-driven alert optimization; AML investigation; bias and ethical pitfalls in machine learning; and enterprise-level AI-driven Financial Crime Investigation (FCI) unit. There is also an Appendix which contains a detailed review of various data sciences approaches that are popular among practitioners.

The book discusses each topic through real-life experiences. It also leverages the experience of Chief Compliance Officers of some large organizations to showcase real challenges that heads of large organizations face while dealing with this sensitive topic. It thus delivers a hands-on guide for setting up, managing, and transforming into a best-in-class financial crimes management unit. It is thus an invaluable resource for researchers, students, corporates, and industry watchers alike.

Inhaltsverzeichnis

Frontmatter

Chapter 1. Overview of Money Laundering

Abstract

This chapter provides a context to the financial crimes for the readers who are new to the topic. It explains the money laundering concept and motivation of money laundering for various participants. The chapter explains various mechanisms of money laundering like structuring, muling, and concepts like integration. The chapter delves deeper into the regulations that are introduced by major economies like the USA and the Eurozone. It discusses in detail the evolution of various regulations and amendments that are introduced by the regulators as they were exposed to newer ways and mechanisms of evolving financial crime tools and vehicles. In the next section of the chapter, authors have aimed their attention to the financial institutions. It discusses the inherent conflict of interest that a financial institution has in combatting financial crimes. The chapter eventually ended with the discussion on impact created by regulators putting more pressure on the financial institutions.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 2. Financial Crimes Management and Control in Financial Institutions

Abstract

This chapter provides a glimpse of a typical financial crime organization in a financial institution. It discusses the governance structure that comprises two important elements—organization design for managing financial crimes and a set of policies and procedures. Content of the policies for any financial crimes management is typically prescriptive across countries and is generally in line with FATF recommendations. Procedures vary depending on the scope and scale of business, however, the essential elements of ensuring that the customer and beneficiary they are dealing with are not watchlist and that the participants in the transactions are not trying to subvert the system to make illegal money seem legal. Internal audit acts as the third line of defense for the compliance. The other aspect of governance is the compliance organization. While there exist different configurations for the compliance organization, common elements are watchlist screening, transaction monitoring, customer risk assessment, legal, and reporting to the central bank. Authors have also discussed an ideal organization structure and other best practices to govern the financial crime monitoring organization. Authors provided examples of a few rules or scenarios that monitor money laundering in banks and insurance companies are also presented. Lastly, the management reporting that provides performance management overview to the senior stakeholders is discussed. It can be used as a basis to design various dashboards that the compliance organizations may want to implement.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 3. Overview of Technology Solutions

Abstract

The chapter provides an overview of the technology solution suite that is required for managing and combatting financial crimes. It provides functionality of KYC, onboarding risk solution, financial transaction monitoring solution, the case investigation, and the reporting module. The chapter provides key functionalities that are required of the technology solution. It also provides overview of the technical requirements. Any financial institution shall be able to review them and can create their specifications both functional and technical through this chapter. With the digitization theme shaping in various financial crime control units, the chapter discuss key themes within AI and digitization that are shaping up the AML market. The next section focus on overview of the AML market along with the key features that the customers look for, when they search for such a product in the market. It also highlights key product and service elements which make a product win or lose customer preference in the market.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 4. Data Organization for an FCC Unit

Abstract

Data is the foundation on which any robust AML platform resides. This chapter starts with focusing on important aspects of data quality like data integrity, fill rates, presence of outliers, and missing value treatment. Chapter provides explanation of these along with relevant examples of those from AML context. It also discusses ways to deal with data issues. The chapter then focuses on various dimensions of data that is leveraged to develop a robust AI-enabled AML solution. It captures dimensions from raw data fields and mechanism of converting the raw data into a mart that can be leveraged for analyzing 360-degree customer views. Data challenges related to cross-border data sharing, creation of synthetic data, and pooling for handling less volume of data for generating meaningful insights are discussed. It provides hands-on approach to assess the underlying distributions and then generate synthetic data. In the context of data privacy, the chapter also deals with the GDPR-related restrictions and challenges that EU is facing in that context. Lastly, the chapter provides some of the areas of improvements that any head should focus on and shall they intend to develop a robust AI or digital AML organization.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 5. Planning for AI in Financial Crimes

Abstract

The intent of this chapter is to create a business case and assist the practitioners in developing a roadmap for setting up an AI organization to combat financial crimes. It starts with the trends and forces that are shaping the need for AI enablement in FCC organizations. It covers all types of evolutions from products to regulators’ focus coupled with the complexity of transactions. The chapter shifts its focus on key success factors that a planner or head should keep in mind while designing the AI organization. Authors have used their experiences to highlight typical pitfalls that a lot of planners miss while setting up such an organization. The chapter then focuses on specific use cases and opportunities that the head or the planner can leverage to define its own end-state vision for the AI organization for their FCC unit. Eventually, the reader should be able to create a vision for its organization while keeping some of the building blocks and caveats in mind for such an initiative.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 6. Applying Machine Learning for Effective Customer Risk Assessment

Abstract

This chapter focuses on three specific topics from a customer risk perspective—KYC process, watchlist screening, and customer risk assessment. While a lot of financial institutions are already working on automating the KYC process, authors have tried to provide an end-to-end process of digitization using application of OCR, computer vision, and basic machine learning-driven algorithms. Watchlist screening is a slightly more complicated topic requiring a robust machine learning-driven name-matching algorithm, where additional information can further reduce false positives. A detailed process of name matching is discussed that starts from preprocessing of names, transliteration and standardization of names and eventually name-matching algorithms. Authors have also provided an overview of a few prominent name-matching algorithms that focus on phonetics and edit distance. Eventually, the process of ensembling and machine learning was discussed that provide the best possible results. The last section in the chapter discusses customer due diligence process. It explains risk assessment scorecards. Authors have provided overview of expert-based and machine learning-driven scorecard. The chapter provides detailed explanation of machine learning-driven modeling process going upto discussion on the factors that are typically used in the development of such a scorecard.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 7. Artificial Intelligence-Driven Effective Financial Transaction Monitoring

Abstract

Financial transactions monitoring is one of the most voluminous aspect in financial crime monitoring for any financial institutions. There are millions of transactions that every financial institution executes. They need to ensure that the transactions that have been authorized are not facilitating any form of financial crime. The chapter provides overview of real-time transaction monitoring and the necessity of ensuring that the counterparty to the transaction is not a watchlisted entity. Another important aspect is monitoring the financial transactions in non-real-time basis through scenarios. Concept of scenario finetuning through thresholds is discussed. The chapter also delves into definition of dynamic segments which are behavioral as compared to demographics that can further help finetuning the thresholds as the thresholds can be adjusted based on the past behavior of the segment, rather than painting all members of a demography-based segment with the same color. Threshold finetuning through ATL and BTL testing is also discussed. Last portion of the chapter explains possible overlaps that exists across scenario. A mathematical approach to optimize the overlaps is also discussed.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 8. Machine Learning-Driven Alert Optimization

Abstract

This chapter discusses machine learning approach for the financial transaction monitoring. One common theme that we would recommend is to start with simplistic approach containing readily available data of good quality. Build slowly additional information like more variables or factors, network-related information, and then eventually work on transaction-level monitoring. However, to start with it could be alert scoring as it seamlessly integrates into the existing transaction monitoring for the organization. There are multiple methodologies proposed by different researchers. Even we have proposed an approach. Practitioners can adopt one of the proposed approaches or create a new one. Our proposed approach creates a customer view as it provides multidimensional view of the risk that the customer undertakes and then percolates that down to scenario level, given the type of transaction risks a particular scenario provides. A detailed overview of the model development process is explained in the appendix. With the right applicability of the framework, a user with non-technical background should also be able to work with the data scientists to lead the machine learning-driven alert scoring or transaction scoring.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 9. Applying Artificial Intelligence on Investigation

Abstract

Investigation of alerted cases is the real operational cost for the business. Every alerted case has to be investigated and then a judgmental decision is taken to either proceed with the case or close it. A big portion of time is spent in collating information across different channels, transaction types, parties, and KYC to assess, whether it warrants deeper investigation. This chapter discusses various analytical insights, consolidation of 360-degree view, automation of adverse media, and network as a few tools that can assist the investigator in reducing the turnaround time significantly, thereby reducing the cost of investigation. The chapter also discusses opportunities of automation through digitization, machine learning like automated triggers for risky network, and unstructured text mining for adverse media, NLG for the automation of narration creation as some of the tools for the business.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 10. Ethical Challenges for AI-Based Applications

Abstract

One of the important aspects of leveraging machine learning applications is the biasness or privacy of individual. Machine learning algorithms are generally a black box. It becomes difficult for the users to further infer the implications, unless they have transparency in terms of whole model development lifecycle including various decisions made as a part of the development process. This chapter first deals with the misuse of personal information. This to a major extent is handled through a robust data governance process. However, bias in decisioning and eventually machine learning stays an important aspect that needs deliberations. Bias can be introduced because of human bias or data-driven bias. Even the ways to handle the bias come from reviewing the whole model lifecycle starting from sampling approach, ways to ensure representation. Whether key segments are rightly represented. If the organization has objective to introduce diversity, then ensuring few specific elements of diversity are not unnaturally discriminated against. Lastly, the model review process is explained that takes care of potential bias. Mechanism to handle human biases is also discussed. A reader will be able to understand various ethical issues and a framework to detect and mitigate those ethical challenges.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Chapter 11. Setting up a Best-In-Class AI-Driven Financial Crime Control Unit (FCCU)

Abstract

This book kept discussing best practices of the financial crime management. Starting from governance, going upto data organization and then digitization and artificial intelligence applications. All key aspects are covered. This chapter crafts them together to showcase the best-in-class organization. The character of an organization is defined by its leadership, which forms the core of any FCC unit. It is followed by the people who are tasked with furthering the objective of the FCC unit. An ideal organization design and key skill set for running a successful organization are discussed. Next in line is the data organization which feeds into all the intelligence that the organization generates and leverage for decision-making. It is followed by capabilities in terms of machine learning and the technological sophistication which is required for precise insights generation, prediction, forecasting, or optimization and then automating the process through artificial intelligent productization. All of these elements playing in sync create a best-in-class organization.

Abhishek Gupta, Dwijendra Nath Dwivedi, Jigar Shah

Backmatter

Titel: Artificial Intelligence Applications in Banking and Financial Services
verfasst von: Abhishek Gupta
Dwijendra Nath Dwivedi
Jigar Shah
Verlag: Springer Nature Singapore
Electronic ISBN: 978-981-9925-71-1
Print ISBN: 978-981-9925-70-4
DOI: https://doi.org/10.1007/978-981-99-2571-1