AI in Financial Services: Revolutionising Banking in India

AI in Financial Services: Revolutionising Banking in India

India leads the world in digital banking users, with a diverse and dynamic customer base driving the need for inclusive financial growth. Today’s customers expect easy access to personalised banking services, pushing banks to enhance productivity and meet rising compliance demands.

The India Stack—a set of open APIs for government, businesses, and individuals—has revolutionised financial services by expanding access in a once cash-dominated economy. Aadhaar, India’s unique ID system, has reduced the cost of customer verification, contributing to a rise in bank account coverage from 53% in 2015 to 78% in 2021.

This digital transformation has paved the way for technology, particularly artificial intelligence (AI), to reshape banking. AI enables better delivery of banking products and services, while generative AI’s ability to analyse large datasets and generate insights is poised to be a game-changer in the industry. Banking remains a leader in AI adoption across various areas, including credit evaluation and more.

1. Underwriting

Retail credit has been India’s major source of loan deployment, and it is growing rapidly. The increased number and pace of loan applications have mandated shorter risk assessment and evaluation cycles while assuring greater risk management and compliance.

Underwriting is a complex process that necessitates adhering to bank policies, regulatory compliance, risk assessment, and customer due diligence. AI algorithms are being used to evaluate creditworthiness by analysing alternative data sources such as frequent utility payments and consumption habits. Generative AI also enables the assessment of unstructured datasets, allowing underwriters to make more informed and accurate credit decisions.

2. Wealth Management

Over the previous decade, India’s investment accounts have steadily increased. Individual investors’ average Asset Under Management (AUM) increased by 21% in June 2023. However, India’s AUM as a proportion of GDP is 16.9%, compared to China’s 20% and South Africa’s 62%.

Banks are particularly positioned to capitalise on retail development, as India has the second-greatest number of high net-worth individuals among the BRICS countries.

All of these circumstances have provided a unique opportunity for Indian banks to step up and meet the increasing demands of both new and existing customers.

Banks can now deliver individualised investing and algorithmic portfolio management through the increasing use of artificial intelligence. For wealthy consumers, banks can utilise generative AI to analyse company data, summarise financial reports, analyse a customer’s portfolio, and offer investment suggestions based on the individual’s risk tolerance, financial goals, and so on. Banks will be able to use AI to generate portfolios for ordinary clients seeking safer investments. AI has enabled hyper-personalization at the individual level, on a large scale, and with high consistency.

3. Risk and Compliance Management

Digital banking has democratised user access to financial products, increasing the risks that banks confront. The volume and complexity of risks are making risk management more difficult. Furthermore, when product services and distribution channels expand, their scale and scope grow.

Traditional AI, mixed with generative AI, has the ability to significantly improve the process of finding, assessing, and managing risks faced by banks. One of the challenges is the danger that a growing number of false positives – transactions that are incorrectly marked as suspicious – would overwhelm areas such as transaction monitoring and fraud. For example, every transaction that is identified as fraudulent or a potential money-laundering violation must be triaged, confirmed, and flagged; this process can overwhelm computers when done in large numbers. AI will assist risk management teams in identifying a variety of threats on a large scale.

Banks continue to invest in technology to automate compliance processes, which are complex and time-intensive. The size of compliance efforts continues to rise in terms of volume and velocity as the number of transactions and channels through which they occur increase. Generative AI will provide banks with crucial tools for managing this size.

Its algorithms can process enormous volumes of data, analyse regulatory changes, and develop risk assessment drafts from that knowledge. Furthermore, AI can generate responses to questions from internal stakeholders and regulatory bodies.

4. Customer Service

Banks have increased their digital banking offerings during the last decade. However, prior to the introduction of generative AI, the majority of these services required human intervention to advise and coach users. Chatbots, for example, lacked full conversational capability and functioning. Changes to the underlying knowledge base necessitated retraining of existing AI models. Generative AI has now overcome these concerns.

With improved training and inference skills, generative AI will be able to apply knowledge and patterns in new encounters, which will be extremely useful in customer service. These capabilities are not restricted to customer service; they open up a plethora of possibilities in the field of conversational banking.

What is ahead for India’s banking sector?

Almost all AI systems today fall under the category of “System 1 AI,” which replies to enquiries synchronously with a response. For example, when a consumer asks a chatbot for information about a credit card, the chatbot analyses the query and returns the best answer from its knowledge library.

AI systems have generally lacked the ability to construct a problem-solving strategy, summon reasoning, and allocate time and effort to tackle a problem in a systematic manner (System 2 AI). You can’t tell an AI to “Create an annual report” and expect it to figure out all the data sources, get the data, build the P/L, generate the text, and put together the final report.

This has altered with the introduction of generative AI. Some AI models provide views into how System 2 AI can function. Advances in AI (particularly generative AI) enable the integration of System 2 mechanisms with well-developed System 1 AI underpinnings. Several enterprises, including Axis Bank, are conducting experiments on building AI platforms that can harness the rational decision-making ability of System 2 AI in combination with System 1 to solve problems beyond the realm of System 1 AI.

We must not lose sight of the ethical component while making these advancements. Banking is a heavily regulated sector; therefore, transparency, data security, accountability, and unbiased decision-making are essential. Banks should use AI to address these critical areas comprehensively.

The explainability of AI models is critical since it is necessary to understand the fundamental reasons underlying the decisions made by these models. When utilising AI models, we must comply with data privacy and security standards. There is a possibility that they will display biases found in training data or disclose sensitive information. Bias in AI algorithms can lead to incorrect decisions or, worse, may end in upholding racial, gender and other discriminations. Accountability for wrong decisions made by banks due to algorithms is another complex issue since these algorithms are primarily opaque.

link