With this era of fintech, we are expected to witness a surge in digital payments. Experts believe the growing volume in this space demands a scalable system for processing and managing financial data. This is where AI seems to fit into the game, due to its ability to manage and process large amounts of data on real real-time basis. Typically, in the payments landscape, the adoption of artificial intelligence (AI) and machine learning (ML) is believed to allow users access to personalised user experiences, fraud detection and real-time payment rails, among others. “Artificial intelligence and machine learning helps in detecting intelligent fraud by utilising its intelligence to extract insights by identifying irregularities, suspicious patterns, and historical data, thus enhancing its effectiveness in overall enterprise payments monitoring and management,” Subhasis Bandyopadhyay, VP, global BFSI domain head, Happiest Minds Technologies, told FE-TransformX.
Personalisation and real-time tracking with ML
Industry experts believe that machine learning can help users detect intelligent fraud by identifying irregularities, suspicious patterns and historical data, thus, enhancing its effectiveness in overall enterprise payments’ monitoring and management. It is believed that machine learning and artificial intelligence aid in securing digital payments by improving the authentication process with its verification of users, by using biometric recognition and behavioural analysis.
Machine learning is believed to have made payment settlements easier in routing money around different RTP (real-time payments) rails that make electronic transfers error-free and automate the authorisation and completion of those transactions in real-time mode. “Real-time monitoring helps to intervene promptly concerning atypical activities thereby mitigating risks timely. Moreover, with predictive analytics, one is able to predict threats as they unfold thus eliminating them before they take place as an act of staying away from sensitive financial information,” Lalit Mehta, co-founder, CEO, Decimal Technologies, explained.
User cases and market statistics
The digital payment ecosystem has grown exponentially in the last few years. In financial year 2023, about 114 billion digital transactions were recorded across the country, as per insights from Statista. User cases include TrueAccord’s HeartBeat, a machine learning tool, that helps lenders customise personal interactions in real-time, based on its ability to detect why a customer’s payments are late. Companies using machine learning have been able to reduce their bad debt provision by 35% to 40%, as per insights from McKinsey, a market reaserch platform.
Use cases include PayPal’s Braintree Auth payments tool, which uses PayPal’s consumer transaction data in conjunction with software developer Kount’s fraud detection capabilities to authorise high volumes of transactions and verifications in near real-time. “Machine learning and artificial intelligence also enable adaptive security systems that learn from evolving threats to stay one step ahead. Additionally, these technologies assist in risk assessment and allow financial institutions to set transaction limits and trigger alerts when suspicious activity is detected,” Sathvik Vishwanath, co-founder, CEO, Unocoin, said.
Contradictory views around use of machine learning in digital payments
Reportedly, there was a data breach of the personal details of about 1.5 million people at the Tata-owned Taj Hotels group, on November 23, 2023. The breach was reported by a threat actor named ‘Dnacookies’ which demanded $5,000 for the full dataset, as per media reports. This is expected to have stirred the adoption of machine learning into digital payments. “eWallets will see improved security, customer experiences, and integration with emerging technologies such as blockchain, Internet of Things (IoT) and biometrics, among others. As digital payments continue to grow, AI and ML will play a central role in shaping eWallets’ future, offering advanced features and driving industry growth,” Balaji Viswanathan, MD, CEO, Expleo India, a technology, engineering and consulting service provider, highlighted.
Artificial intelligence applied to payments did not have a high adoption rate among consumers in early 2023. A survey held in 14 different countries across North America, Europe, and Latin America observed that consumers were not comfortable yet with the idea. About 10% of the respondents mentioned that they might use artificial intelligence in two years’ time, when it becomes more established, as per insights from Statista.
Naysayers argue that without the incorporation of regtech, these automated functions might drive illegal activities such as data breaches and complicate data collection methods, among others. RegTech solutions can make data filings more efficient, transparent, and useful for everyone who generates, collects and consumes them. Use cases include policy initiatives such as the Financial Transparency Act in the United States (US) and Standard Business Reporting (SBR) in Australia.
The road ahead
In this fast-paced age of technological developments, machine learning can be more significant in protecting digital payments than just using algorithms. Further, it is believed, this can also reflect the collaboration between artificial intelligence and financial security. “Machine learning is both a vigilant overseer of digital payments and a protector against the constantly changing world of cyber threats. It gains knowledge, adjusts, and strengthens the barriers around financial transactions with a never-ending thirst for data. It recognises patterns, irregularities, and possible threats during payments, much like a silent guard,” Sanjay Kaushik, managing director, Netrika Consulting, concluded.
Follow us on Twitter, Facebook, LinkedIn
