Artificial intelligence (AI) and Machine Learning (ML) are playing a significant role in the Asia Pacific (APAC) due to their use in credit and risk functions for enhanced credit assessment, credit scoring, and fraud detection.
AI’s priority has increased so much in every sector, that in the future, even in banks and financial institutions, AI will be a much-needed technology rather than staying an option. According to data specialists and leading finance executives who stated in a webinar that with AI in these two domains, a lot can be achieved such as matching customer expectations, exploring new business opportunities, and addressing fast developing fraudulent activities.
Fintech Fireside Asia’s latest panel discussion witnessed the participation of top-level executives representing Union Bank of the Philippines, credit bureau TransUnion, and lending startup Funding Societies & data solutions provider Mobilewalla went on to discuss the AI adoption’s state across APAC’s financial ecosystem, exploring the way predictive modeling and Machine learning are currently being utilized in the lending process.
Enhancing Financial Inclusion
Anindya Datta, Founder, CEO, and Chairman of Mobilewalla considers AI as an opportunity for delivering cutting-edge business models that can go ahead of the conventional models and reach the people without bank accounts, a possibility that is specifically relevant in Southeast Asia.
It is the region where more than 70% of the adult population is currently underfunded.
According to Anindya, the major decision-making part of lending lies in assessing the person’s potential to pay back and whether they will do so on time.
The reason for it being intriguing in the emerging markets, especially in APAC is that the credit footprint is not huge, and many don’t have credit scores.
When it comes to technology adoption, ML is likely one of the most widely used framework technologies in fintech that is used to determine creditworthiness.
By depending on non-conventional information like customer mobility and average household phone cost, AI can minimize the information asymmetry for people lacking credit history, expanding credit availability to those whose creditworthiness cannot be measured with the help of traditional metrics, according to Anindya.
According to Dr. David R. Hardoon, Chief Data and AI Officer at Union Bank of the Philippines, all that matters is making lending more compatible, backing at-scale personalization, and understanding that banks deal with different underlying comrades and individuals with varying degrees of circumstances and needs.
AI is about the principle of hyper-personalization, and has its focuses on understanding the specific customer from a behavioral perspective, not simply from a repayment standpoint, but also from a capacity and necessity standpoint, David explained. Understanding the usage and the moment of lives linked with that and shifting from a lending perspective to a credit management perspective.
Exploring the possibilities
According to Ishan Agrawal, Group CTO, Funding Societies, Modalku, the scope of AI and ML is beyond credit scoring, and it can also be utilized in the lending lifecycle. Acquisition, electronic know-your-customer (eKYC), detection of fraud, underwriting, and collection are some of the processes in which AI and ML can be of use in the lending lifecycle.
Ishan went on to add that AI has applications all the way through the lending lifecycle. If someone is in the fintech space or carrying out any type of lending, be it buy now pay later (BNPL), credit card, SME (small and medium-sized enterprise) lending, AI will have a very important role to play, not only in credit scoring but across the board.
AI is evolving at a rapid pace, and it will be exciting to see what AI will be capable of in the next five years.
In customer acquisition, advanced analytics enable banks to provide highly personalized offers and superior experiences, resulting in higher conversion rates. This is accomplished by gaining a better understanding of each new customer’s path to the bank and obtaining an accurate picture of a customer’s context and direction of movement.
AI and ML can be used in credit decisions to analyze extensive and diverse datasets. As a result, banks can qualify fresh customers for credit services, decide their loan limits and pricing quickly and also minimize the fraud risks by anomaly detection.
Detecting fraud manually is a very tedious process, according to Ishan. For instance, a person’s behavioral analytics needs to be checked, information needs to be gathered on the way they fill an application, their typing speed and pattern, and the time duration taken by them to finish the application. It is impossible to detect fraud in all these issues without the help of AI.
Credit bureaus welcome AI and open banking
Apart from fintech companies and binding banks, credit bureaus have begun to energetically leverage alternative data and utilize ML to provide exceptional insights and expand their coverage.
TransUnion Philippines launched CreditVision Link last year, a solution that can credit score Filipino adults with the help of conventional credit data as well as alternative data comprising telco data such as reloads, payments, mobile data usage, and device data.
In Hong Kong, we have 85 percent coverage, however in the Philippines, we only catch about 25 percent of the population, said Jerry Ying, TransUnion’s Chief Product Officer, Asia Pacific. With CreditVision Link, we can now look at around 70-80 million people instead of covering around 25 million of the Filipino population. That’s a significant improvement.
AI plays a significant role in this scenario – to compensate for customer coverage where some of the newer customers in the credit segment are unable to obtain a lending.
TransUnion is also actively investigating open APIs as a key technology. The credit bureau in the United Kingdom has launched an open banking service that enables consumers to share their data with third parties when applying for credit. This aims to enhance the consumer experience by making lending decisions faster and providing more precision in risk assessment.
We’re looking into introducing that in Hong Kong, Jerry said. We’re looking at the way to access data, use AI to decipher inquiry transactions and interpret matters such as an individual’s cash flow and income level, and pull that data straight from third-party sources.
