Applying AI to Risk Management in Banking and Finance. What’s the latest?

How Is AI Transforming Risk Management in Banking and Finance?

Over the last decade, artificial intelligence (AI) has become a symbol of improving efficiency and productivity while reducing cost in the financial sector. The relentless competition from new Fintech companies has forced traditional banks to upgrade their legacy systems and architecture. Furthermore, the growing digital ecosystem has created a new brand of tech-savvy customers who want their banks to do more.

AI is a major game-changer in risk management. Inherently, financial institutions are prone to risk due to the type of information they handle on a day-to-day basis. AI is the perfect way to streamline the management of those risks in a growing, competitive industry. This article reviews some of the risks that banking and finance is exposed to, and how AI can mitigate them.

What Do You Mean by AI?

When referring to AI, we’re not talking about the Terminator-type world Hollywood brought to life in 1984. AI is simply machine learning: using existing data to spot patterns and behaviours likely to indicate future actions. This is how Amazon recommends new products for you or how Netflix suggests the shows for your next “binge”.

AI in Credit Risk

Credit is one of the largest risks to a financial institution. This is money loaned to individuals or businesses who might default on their obligation to repay. There could be circumstances outside of the bank’s control leading to defaults (e.g., a client loses their steady income stream), but there are avenues through which AI can negate these incidences.

In March 2019, an Experian report showed a default rate of 3.68% on credit cards. In this basic instance, machine learning algorithms could have been used to better assess customers’ credit histories. The insights provided would have revealed additional vulnerabilities or caught patterns easily missed by human eyes. Zest Finance advertised they can reduce default rates by approximately 20% using AI-based models.

Market Risk

When it comes to reducing the risk in market trading, AI is perfectly equipped. Machine learning algorithms can analyse vast volumes of data points in seconds–which would otherwise take humans at least several days to review.

The resulting insights give traders optimal price points (relative to the risk of losses), and let them forecast at much greater accuracy. Trading firms mitigate more risk while achieving higher returns. One such example is Trading Technologies, which combines machine learning with powerful processors so to identify trading patterns across multiple markets.

Fraud Risk

Malicious acts such as identity theft and money laundering are very common with financial institutions. In 2019 alone, there were over 650,000 reported cases of identity theft.

As AI models are able to analyse data in massive volumes, they can quickly spot patterns from several different channels in tandem and send alerts about potential fraudulent activity for a virtually limitless number of banking clients at once. There are many companies specialising in this area because of the potential cost savings. For example, AK Sigorta specialises in reducing the number of fraudulent insurance claims using AI. Teradata says they can drastically reduce fraud cases, by up to 50% for Danske Bank as one example.

Larger banks are now using multiple layers of protection within their applications to protect customers. These include fingerprint and biometric identification practices, part of the new norm of mobile phones.

Underwriting Risk

Insurance companies are constantly trying to spot fraudulent claims.

In underwriting, customers are assigned premiums taking into account various risk factors associated with their respective likelihood to claim. These generally include: age, domicile, previous claims history and possibly their credit score. Further, there may be some insurance-specific questions related to the type of cover they are taking out (motor, home, travel, life, and so on).

If the data points are not sufficient to give a clear picture as to whether a customer is likely to claim, it could potentially be very costly to the insurance company. AI is used to collate data points from multiple sources and give a more rounded risk profile for each customer. For example, in home insurance, a company might use data from Zoopla, mortgage surveys or Land Registry logs to better understand the state of the property they are covering. Specialist information companies like Whenfresh are now working with insurers on their data needs.

Challenges for AI in Risk Management

While AI can bring many benefits to risk management, there are challenges to overcome. Many banks are still using legacy systems with inconsistent and incomplete data sets which are hard to extract. Even when the sets are pulled together from disparate sources, that does not mean they are shared willingly. Perhaps, they are shared only for regulation and compliance.

The first hurdle, however: a 2019 survey revealed that over 20% of people within risk management are not even aware of how machine learning can impact their role.

If financial services is to reach full maturity and utilise AI applications for risk management, there is a long road ahead.


AI can automate many aspects of risk management within the banking and financial services sector. As it stands, the technology is doing a great job of augmenting human processes to help businesses better understand potential risks. In the future, algorithms will continue to learn, begin to work completely autonomously, and allow humans to focus on more complex tasks. It might require a particularly brave trader to make the leap of allowing AI to automate all of their investment decisions, but it will be done.

This next decade will reveal just how dramatically we will come to rely upon machine learning.

Disclaimer:  The author of this text, Jean Chalopin, is a global business leader with a background encompassing banking, biotech and entertainment.  Mr. Chalopin is Chairman of Deltec International Group,

The co-author of this text, Robin Trehan, has a bachelor’s degree in economics, a master’s in international business and finance, and an MBA in electronic business.  Mr. Trehan is a Senior VP at Deltec International Group,

The views, thoughts, and opinions expressed in this text are solely the views of the authors, and do not necessarily reflect those of Deltec International Group, its subsidiaries, and/or its employees.