Category: Banking

AI in Banking: Beyond the Bots

Unlocking value from AI will distinguish winners from losers.

By Mark Leher, VP of Data and Analytics

It wasn’t so long ago that artificial intelligence (AI) in banking seemed futuristic. Now, AI in the banking industry is officially mainstream as use cases mature from nice-to-have, cutting-edge options into day-to-day capabilities that offer concrete benefits and extraordinary savings. It’s about efficiency, speed, data and insights that just weren’t possible before.

The trend toward AI is continuing to gather steam. According to a recent study by Business Insider:

  • 80% of financial institutions are “highly aware” of the benefits of AI and machine learning.
  • 75% of banks with more than $100 billion in assets are currently implementing AI.
  • 46% of FIs with less than $100 billion in assets are currently implementing AI.
  • By 2023, FIs are projected to save $447 billion by using AI, the majority of that being derived from customer-facing apps like chatbots, and back-office uses like anti-fraud and risk applications.

And this, from a March 2022 report by Temenos and the Economist:

  • 81% of IT executives in banking agree that “unlocking value from AI will distinguish winners from losers.”
  • 85% have a clear strategy for adopting AI in the development of new products and services
  • 78% say that using AI will help them achieve their business goals and priorities, with 46% of those saying “to a great extent.”

Those numbers are very clear. It’s not if AI is now essential banking, it’s how is it being used and to what end.

Growing use cases of AI in banking

What are some of the top uses of AI in banking today?

Chatbots. These almost seem old hat by now, don’t they? But they are evolving from simple apps that allow banking customers to ask questions 24 hours a day. Take Bank of America’s Siri-like bot, Erica. The virtual assistant can respond to more than a million unique financial questions and has nearly 20 million users, who had 105 million interactions with her last year, up from 27.8 million the year before. A well-trained AI driven chatbot can enhance the customer experience.

Cybersecurity. Today, digital banking is ubiquitous. When is the last time you paid for, well, anything with cash or a check? Increasingly, we’re even moving away from using a physical credit card and simply tapping our phones to buy most everything. With all of these digital transactions, the need for cybersecurity and fraud detection is ramped up exponentially. But this isn’t just for customers who suddenly find two first-class tickets to Vegas on their credit card bill. It’s for banks themselves, too. According to a 2020 report by Richey May, the financial industry was the most targeted industry for cyberattacks, making up 29% of the total amount of cyberattacks in the U.S. That’s up from just 8% the year before. An alarming jump, causing the industry to ramp up AI protections.

Consumer Lending. Credit history and credit scores are the gold standard when bankers are deciding whether to give a customer a loan — or are they? Credit reporting is not an exact science and many bankers wish there was a better way that wasn’t filled with so many errors and outdated information. And basic credit reports simply have no data on current transaction history. With AI, bankers can get a broader picture of a customer’s spending habits, behaviors and more, getting a holistic view of the person’s financial wellness.

Predictive Modeling. With AI, FIs can analyze account holders’ everyday purchase transactions and their utilization of banking products. What they find is a set of insights that paint a picture of an individual’s financial life. By analyzing data about everyday purchases, FIs can get a clear view of an account holder’s financial priorities. Those everyday purchases are extremely predictive of an account holder’s future spending behavior, allowing an FI to offer relevant products at just the right time. Predictive audiences can also be created for a large number of business cases, attrition or mortgage sales for instance. Taking into consideration account holders who have attrited, or those who have closed on loans, in order to predict those likely to do the same in the future.

By leveraging all of this data with AI, financial institutions are able to not only be more efficient and productive, but also deepen the account holder experience in an age when few people walk into their local branch anymore.

About Author:

Mark Leher is the Vice President of Data and Analytics at Segmint, where he is responsible for business models related to Segmint’s technology, taxonomy design, client management for the taxonomy business and strategy for new data solutions. 

Share