How a Multinational Finserv Slashed International Due Diligence Time With AI

Globally, businesses lose a staggering $42 billion to fraud each year, and every one in two companies has experienced fraud. Financial services companies are increasingly being challenged by customers, competitors and regulators to detect and manage risk meticulously. Concurrently, the fines imposed for non-compliance have also escalated to record levels.

According to PwC’s 2020 Global Economic Crime and Fraud Survey, 25% of businesses are using artificial intelligence to detect changes in transactional patterns, conduct due diligence, reduce the amount of false-positive fraud alerts and increase the effectiveness of fraud detection, KYC and AML programmes.

Download our free ebook, Preparing for the Next Fintech Evolution With AI

Discover how AI will drive the next wave of disruptions in the financial sector and how you can use that to your benefit.

AI in AML and KYC

The financial sector is required to operate within Anti-Money Laundering (AML) regulations that are based on the Financial Action Task Force (FATF) and the International Standards on Combating Money Laundering and the Financing of Terrorism & Proliferation (FATF40), as outlined by the 3rd and 4th EU Anti-Money Laundering Directive (4AMLD).

The standards set in the 4AMLD require financial institutions to identify and verify business clients and beneficial owners, apply customer due diligence and enhance due diligence when higher-risk situations have been assessed.

And the FATF guidance expects finservs to take a risk-based approach to countering money laundering and terrorist financing.

AI, ML, data analytics and other related technologies can quickly analyse large volumes and differentiate between various types of data, automate workflows to enhance AML and KYC.

Financial institutions can use AI to enhance due diligence by:


  • Link analysis: identify customer links with questionable jurisdictions or companies
  • Unstructured data analysis: analyse data from the web, news, social media
  • Pattern recognition: identify patterns in customer behaviour
  • Workflow automation: generate documents, reports, audit trails and notifications

How Cognitive Search Helped a Multinational Finserv Save Time Conducting Due Diligence

International due diligence involves exhaustive research, like tracking more than 40,000 global sources in media, corporate records, financial transactions and legal cases.

Results from the search text strings are painstakingly reviewed for each entity, following which reports are created manually. This is a laborious process that takes financial institutions weeks.

With numerous partners, vendors and counterparty relationships worldwide, this multinational company wanted to refine its software to identify compliance risks, analyse risks in real-time, reduce manual effort and lower costs.

Conventional searches were simply not smart enough, and employees were wasting vast amounts of time reviewing irrelevant results.

The company worked with Cognizant and Amazon Web Services (AWS) to develop an API to connect the client’s software to a machine learning model built on IBM’s Watson Explorer 11 and Watson Content Analytics.

In less than five weeks, they developed cognitive APIs powered by deep learning algorithms. The contextual search model allowed the company’s third-party application to detect and deliver improved reports on news relating to risk and fraud, while the machine learning model understood linguistic nuances, meaning and relationships specific to the financial industry.

Further, the algorithm extracted relevant information from enormous and diverse data sets and compiled research reports with contextually relevant results. And it also automated the search process, integrated the research workflow and cut the time spent by analysts in manually reviewing irrelevant material by more than half.

With AI, the company was able to complete 14% of its due diligence reports in one hour and was freed up to generate up to 30% more due diligence reports a year. They could also analyse compliance and financial risks in real-time.

Sign up for the Oxford AI in Fintech and Open Banking Programme to learn how you can leverage data and AI to streamline work, gain insights and become more competitive.

Oxford AI in Fintech and Open Banking Programme is delivered as part of a collaboration with Saïd Business School, University of Oxford and Esme Learning. All personal data collected on this page is primarily subject to the Esme Learning Privacy Policy.


© 2021 Esme Learning Solutions. All Right Reserved.