Machine learning a constituent of AI (Artificial Intelligence) is a great leap forward and an indispensable technology for much of today’s generation. The term was coined in 1959, by Arthur Lee Samuel refers to a field of study that “gives computers the ability to learn without being explicitly programmed.” Machine learning has become a buzzword in much of the computer-driven industries today- Retail, HR, Operations, Logistics, BFSI there would be hardly any industry which is left untouched by the wonders of machine learning and artificial intelligence.
Artificial intelligence (AI) and machine learning are rapidly being adopted for a range of deliverables in the BFSI or Banking, Financial Services & Insurance industry. The adoption of these deliverables or applications has been driven by both the demand and supply factors. The supply side is governed by factors such as technological advances and the availability of financial sector data, while the demand factors include profitability needs (including reduction in income leakages, cost reduction, and revenue gains); inter competition (providing customer-centric services for customer retention), and the demands of financial regulatory stakeholders (for data reporting, adhering to AML guidelines).
AI and machine learning are being adopted across the financial system, in the BFSI industry at large. The data slices include Banking, Financial services and Insurance segments.
AI & Machine Learning in Banking Domain
Banking domain has seen a technological disruption of sorts, from client-facing chatbots to complex credit scoring algorithms, machine learning is everywhere. Large-scale customer data are inputted into algorithms to assess the credit quality and customer repayment ability.
Chatbots to enquire Customer’s banking needs:
Many institutions have adopted AI interfaces-chatbots for client interactions. India’s City Union Bank (CUB) has become the first bank to introduce a chatbot to answer customer queries. Called Lakshmi the humanoid banker is built on the IBM Watson artificial intelligence (AI) engine and is based in CUB’s Kumbakonam branch in Chennai.
Credit Scoring Models
Credit scoring models use machine learning to speed up accurate lending decisions, while potentially limiting incremental risk. Credit scoring models are used by banks both for corporate and retail segments. Customer’s historical financial transaction and payment history data from financial institutions serve as the foundation of most credit scoring models which use tools such as regression, decision trees, and statistical analysis to generate a credit score using limited amounts of structured data. Currently, banks are increasingly turning to additional, unstructured and semi-structured data sources for accuracies like analysing a customer’s social media activity, mobile phone use and text message activity.
Regulatory Compliance and Supervision
AI and machine learning techniques are being used by regulatory stakeholders for regulatory compliance, and supervision. The applications include regulatory reporting and data quality; systemic risk analysis, monetary policy, surveillance and fraud detection.
AI & Machine Learning in Financial Services Domain
Capital Markets: Trading and portfolio management
AI and machine learning are used to devise trading and investment strategies. In the capital markets, Machine Learning powers trading execution (primarily sell-side) and portfolio management (buy-side). Trading firms generate huge chunks of data which is fed to Machine learning models to improve the client experience. Past trading behaviour is analysed extensively to anticipate a client’s next trading order. Additionally, AI and machine learning pro-actively manage risk exposures by flagging intervention alerts when client’s trading account positions may have increased risk exposures.
Customised Financial Services products
AI and machine learning have bought a customisation revolution for the customer. Personalised investment products are being offered across equity, debt, and mutual funds segments according to customer’s age composition and risk appetite. While a young professional may be offered an ELSS (Equity Linked Saving Scheme) Product that couples up as an investment and tax saving source, a retired person may be offered debt oriented schemes that promise moderate returns with investment security.
AI & Machine Learning in Insurance Domain
Cost and Profit analysis
The insurance industry uses machine learning to study complex data structures to lower costs and improve profitability metrics. Insurance-related technology, ‘InsurTech,’ often relies on analysis of complex big data. Machine Learning and AI help in formulating an insurance product and analyse its potential gains. Adoption of AI and machine learning involve improvements in the underwriting process and segment claimants to high/ moderate and low-risk categories.
AI may help reduce claims processing times and operational costs by successfully predicting the probability of occurrence of any miss happening. Insurance companies are exploring predictive analytics, through AI and machine learning algorithms. Remote sensors (connected through the ‘internet of things (IOT)’) can detect, and in some cases prevent, insurable incidents before they occur, like car accidents or industrial damages.
Data V/S Data Privacy
The widespread adoption of AI and machine learning has inspired both hope and concern about the rapidly growing complexities and capabilities. The issues revolving around data privacy to access the data being processed by AI and machine learning tools is a worldwide debate. While AI, big data and machine learning are harnessed to generate profits, they operate on customer ‘life cycle data’ which is in many cases too personal. In the global outlook, data is divided by international borders and attracts legal ownership rights and data privacy protections.
The gains to BFSI and its customer from AI & Machine learning has been a win-win situation for both the sides. While BFSI industry seeks to multiply its profits and reduce costs, the customer looks forward to a delightful financial experience. With this premises, big data, AI, Machine learning are the big bets for the future.