AI In Banking and Financial Services
Rohit Talwar
Futurist Keynote Speaker - Leadership, Learning, Innovation, and Transformation for the AI Era
Article 1 of 4 – Who’s Smarter?
Rohit Talwar – CEO, Fast Future
February 22nd, 2024
This is the first in a four part series exploring how Artificial Intelligence (AI) could impact financial services as its ever smarter capabilities enter the field over the course of the next few years.
In this first article I start by examining the current state of play and consider how AI could outsmart the bankers. Subsequent articles will go on to look at how AI could change the rules of the game in banking, how established players can win in this environment, and how new entrants might fight back and disrupt. The third article will look at the opportunities that could emerge for incumbents and new entrants, the smart money opportunity, and the role of AI in enabling enhanced financial inclusion. The fourth and final article looks at the implications for regulators, central banks, and managing the ethical and societal challenges.
We are beginning to accept that the rise of AI is inevitable and inescapable. Everywhere we look, we see stories of how it is – or could soon be - transforming lives, activities, work, jobs, businesses, sectors, and nations. The areas receiving the most attention right now are banking and financial services. This interest is driven by a combination of the size of the institutions, their wealth, the volume of transactions, the number of users, and their centrality to life on the planet.
Hence the excitement in the sector over the potential to grow revenues, cut costs, improve security, enhance customer service, speed up new developments, innovate around products, develop new pricing models, tackle fraud, and improve regulatory compliance. Looking a little further ahead we see even more potential in all of these fields and in the creation of entirely new asset classes and sectors, in the emergence of smart savings and intelligent money, and in the potential for radically different approaches to personalisation, privacy, security, and governance.
In the next few years we can expect AI to evolve rapidly to go beyond its currently narrow but deep task specific capabilities. Current examples of such narrow applications include generative applications like ChatGPT and Bard, a range of robotic process automation (RPA) tools, and highly specialised solutions in areas such as compliance, fraud, and know your customer (KYC) verification. These are excellent in their assigned tasks, but they wouldn’t be very good at each other’s or in other domains such as drug development, autonomously driving a vehicle, or making decisions about a bank’s strategy. However, in the coming years we could see that change with the emergence of artificial general intelligence (AGI) that is as capable as humans across all our cognitive capabilities.
The Current State of Play
In financial services, we are already seeing widespread and accelerating adoption of AI for consumer facing activities. These include enhancing service and engagement through personalized banking 24/7 virtual assistants that provide financial advice and support, help manage accounts, track spending, and make payments (e.g. Bank of America, HSBC). Other common applications include streamlining the mortgage lending process (Wells Fargo).
Others such as Vanguard are using AI wealth management and robo-advisors to support human financial advisors in giving tailored investment advice based on individual financial goals and risk tolerance. In investment banking, the use of AI-enabled big data analytics is being used in algorithmic trading to make investment decisions and manage risk (BlackRock). Others are using AI to assess market sentiment through analysis of real-time news and social media content (Bloomberg).
Recommended by LinkedIn
In order to enhance personalization and engagement, improve retention, and optimize marketing efforts to maximise customer lifetime value, machine learning models are frequently used to predict investment behaviour and product take up (Capital One).
In the ‘back office’, analysis of large historic transactional datasets is helping to improve efficiency, accuracy, and consistency in applications such as behaviour prediction and fraud detection and prevention (Barclays, American Express). Other common uses include interpretation of loan agreements and contract creation (JP Morgan), risk assessment, and credit management (Goldman Sachs). For small business loans, some are going further, by analysing non-traditional data sources to assess creditworthiness. The technology is also in common use for regulatory compliance processes such as anti-money laundering (Standard Chartered), KYC, and AI-enabled biometric recognition and digital identity verification (BBVA).
A number of players are also adopting RPA technology to automate repeatable routine applications such as treasury and cash management (Citibank). In insurance, AI and chatbots are accelerating and streamlining underwriting and claims processing (Lemonade).In the crypto assets space, AI is being used in conjunction with blockchain technology to enhance security and fraud detection.
Can AI Outsmart the Bankers?
As confidence with AI increases, growing use can be expected of AI in areas where bankers simply don’t have the mental bandwidth or resources available. For example, by combining historic data with complex, non-linear risk factors, and publicly available market sentiment information such as social media posts, AI will be able to undertake valuable analytical and predictive tasks at scale and uncover insights that might not be readily apparent to humans.
Potential applications include identifying subtle hidden customer behavioral patterns - such as predicting financial distress or identifying unmet customer needs. This could lead to better risk management, tightening of controls, and more accurate pricing and development of micro-personalized and flexible products and services. Such analysis could also highlight fraud detection nuances – identifying anomalies and patterns indicative of complex and sophisticated fraud or money laundering schemes untraceable by humans.
Internally, AI will help highlight optimization opportunities – capturing, aggregating, and analysing exponentially increasing volumes of operational and staff behaviour data across the organisation. This will help surface inefficiencies and optimization opportunities in processes like loan approval, customer service, and compliance.
At the strategy level, deep analysis of highly interconnected financial markets and geopolitical developments could reveal complex correlations and network effects that could impact systemic risk and financial stability. Combining such data with insights on technology advances, and on hyperlocal developments, could help spot emerging risks such as cybersecurity threats. This analysis could also surface opportunities such as emerging financial tools and assets before they become mainstream knowledge. Economic and financial forecasts could become increasingly accurate using ever more advanced and complex and adaptive predictive AI algorithms. These would draw on the above insights, coupled with increasingly comprehensive global economic data to identify leading indicators and patterns that are not widely recognized.
Rohit Talwar is the CEO of Fast Future – a foresight advisory firm focused on helping clients harness the trends, shifts, and ideas shaping the future. He has a particular focus on the implications and applications of rapidly evolving technologies such as AI.
+44 (0)7973 405145