AI Digital Bank: Reimaging the future

ARTIFICIAL Intelligence (AI) is poised to have an unprecedented impact on humankind in the history of technology in the past century.

AI refers to computer systems that mimic the general intelligence found in human beings. The application of AI in banking brings possibilities that were considered science fiction just a decade ago. Although the fundamentals of banking haven’t changed, it’s how it is done that has been disrupted by unrelenting digital and AI transformation. This article presents a reimagined, future-fit and AI-powered bank referred to as AI-Digital Bank.
The legacy banking business model is product based and vertically integrated and entails ownership of the entire value chain. Almost all products and services are owned and distributed directly by the bank.

Dennis Magaya, is a AI and Digital Transformation Strategist

A business model describes how the bank creates and delivers value to customers and shareholders. When the first wave of new players entered the market with new business models driven by mobile banking offerings and personalised user experience, incumbent banks’ revenue migration was deceptively small. At face value, the banking industry appeared to be in good health as demonstrated by several banks reporting revenue growth and profits. However, a deeper analysis shows that incumbent banks were experiencing a consistent revenue and value attrition. This is illustrated by the year-on-year shrinkage of banking’s overall contribution to GDP from 2009 to 2020 which dropped by 26 percent (UK), 15 percent (USA), 29 percent (Australia) and 22 percent (Europe).
To date, digital transformation is enabling even more players to enter the market with disruptive business models. Fintechs, techfins and digital only banks’ value propositions fragment the traditionally coupled lending, investments and transactions banking products into components from which they cherry-pick more profitable and less capital-intensive ones. Mobile money is an example of a fintech that focuses on money transfers and transactions, avoiding costly branches. The mobile money business for Africa was valued at US$912 billion in 2022 and this impacted the banks’ income. In some markets, as soon as a bank account receives a deposit, customers transfer the money into a mobile money wallet from which transactions are done, resulting in undesirable transient deposits and reduction of profitable transaction fees.
In Zimbabwe, international remittances were US$1,873 billion in 2023 which was mainly transferred by a fintech called Mukuru. Due to market informalisation, this cash is unlikely to enter the banking system.
Banks that use AI powered non-linear business models adapt quicker to customer centricity and value chain dislocations caused by market disruption. A non-linear business model transforms a bank into a value architect for its customers and ecosystem players. Value architects are firms that develop agile business models to play diverse roles in a value chain by relentlessly focusing on customer needs.

The AI algorithms that dominate the market are essentially prediction machines.

They unbundle products into micro-offerings and package them together with components from ecosystem partners, creating compelling propositions. To achieve this, a bank evaluates the areas where they are best in class, which capabilities and products to develop in-house and what to source from partners. It also entails the identification of utility areas to unlock value by using Banking as a Service (BaaS) or open banking model. To achieve this, an incumbent bank requires a roadmap to transition from a monolithic to multiple non-linear models allowing for stages of co-existence.

The AI-Digital Bank requires a new operating model that supports the non-linear business model described above. The governance, company structure and organisation structure are customer centric rather than being based on products. A board committee responsible for digital business models and operating model and AI is included.

A chief customer experience officer for the whole bank, and a chief data officer or chief AI officer are new additions to the executive team. AI is co-worker in all roles to achieve a human and technology combination that delivers better results. Productivity improves because AI doesn’t do organisational politics, it’s not forgetful and works 24/7.

The AI-Digital Bank strives to deliver a blend of omni-channel experience, intelligent customer experience and personalised process orchestration. Omni-channel functionality allows customers to start a service and move seamlessly across multiple channels namely, physical (branch), virtual (call centre) and digital (mobile app) within a single journey without restarting. Intelligent experience means technology that autonomously recommends actions, anticipates and automates key decisions or tasks. Personalised means products and services that are relevant and based on a 360-degree understanding of customer behaviour and context. Personalised process orchestration means when customers use channels, the processes to discover, access and consume services are tailored to suit individual needs.

AI enables a business logic layer where product and service recommendations on what to offer, when and which channel to use are made and communicated in real-time during the customer journey. This is important because from a profitability perspective, branches are about eight and 15 times costlier than call centres and digital channels respectively. Relevant non-banking products and services are integrated into the bank to fully address customer needs using AI to automate decisions and activities on behalf of the customer.

Bank products and services can also be extended beyond-the-bank journeys, such that intelligent experiences become the glue cementing the client’s relationship.

An enterprise-wide AI engine can be built using Symbolic AI (expert knowledge systems), Natural Language Processing (NLP), Machine Learning (ML) and computer vision depending on the bank’s strategy. The AI functionality can be built as a standalone system, incorporated as a module in an existing system or obtained via an Application Programming Interface (API).

Symbolic AI is used in structured and predictable environments where expert knowledge and rules are coded into software to automate decision making. ML is used together with data sets for detailed analysis, to make decisions and continuously learn from new data. Computer vision is used for identification and recognition of human or object images and videos. The AI technology choice to use depends on its complexity and availability of data.

The AI-Digital Bank’s fuel is big data capabilities covering visualisation, analytics and data lake so that the bank can handle both structure and unstructured data. Social media and omni-channels are sources of valuable unstructured data eg videos, audios, pictures and text. AI models require clean and huge data sets so that the hallmarks of the future bank such as usage-based prices, individualisation, intelligent products and digital operations are delivered. Imagine a bank that predicts your routine daily actions and adjusts the mobile App accordingly so that your service journeys are shorter.

Deploying AI capabilities across the firm requires a scalable, resilient and adaptable set of core-technology components. With the rapid increase in customer engagement across both the bank and non-bank platforms, the future bank should deploy hyper-scalable infrastructure to process high-volume transactions in milliseconds. Cloud computing, microservices architecture and Application Programming Interfaces (APIs) enable banks to quickly deploy new services, integrate with third-party platforms, and respond to market changes.

AI powered cybersecurity to protect against vulnerabilities within applications, operating systems, hardware and networks is required.
The AI-Digital Bank of the future’s strategic targets and performance management are impacted by the new non-linear business model and digital operating models. Therefore, the traditional key performance indicators such as income, loan book, bank accounts and financial ratios should be enhanced to include future-proofing aspects. By Dennis Magaya

Dr Magaya is a AI and Digital Transformation Strategist. He is founder and CEO of Rubiem Solutions which offers business strategy consultancy services and has operations in the SADC region. He can be contacted on Dennis@rubiem.com, +263717770666