• Consumer Financial Management: Financial services can be categorised into four consumer finance use cases: pay, invest, borrow and insure. Fintechs must be credited for the quickest implementation of Gen AI in building automated budgeting and financial management tools, that previously required manual intervention from either customers or service providers. Analysing customer portfolio data, providing personalised financial investment and advice is another application.
• Customer Support: Customer support and management are among the most extensive areas of workflow automation in banking. Chatbots and virtual assistants play pivotal roles in addressing customer inquiries, offering immediate responses, assisting customers with tasks such as setting up new
accounts, managing routine disputes and ensuring timely escalation to a Level 2 human customer experience team when necessary.
• E-commerce Enablement: E-commerce enablement is another widely known application of AI in finance. By personalising the online shopping experience through predictive analytics, automated decision-making mechanisms and enhanced fraud detection capabilities, AI-driven finance platforms are ensuring a seamless user experience.
• Fraud Detection and Risk: As cybercrime advances in targeting banks and fintechs, fraud prevention businesses have developed advanced capabilities using AI to power biometric data capture, behavioural analytics, mobile ID and document verification. Minerva and Sardine are two emerging companies in this space.
• Regulatory Compliance: AI is revolutionising regulatory compliance and governance within financial institutions. The ability to simplify the complexities of navigating through evolving financial regulatory requirements and ensuring compliance is solidifying AI’s position as an indispensable asset for financial institutions.
• Credit Risk Assessment: Lenders can leverage AI to analyse datasets to enhance the accuracy of credit risk assessment and create credit underwriting models. AI also empowers fintechs and other financial institutions to provide real-time payment approvals by analysing the data to verify the authenticity of a transaction.
• Investment: In addition to building sophisticated applications for consumer investment and wealth management, financial institutions are using AI to develop proprietary trading and market analysis algorithms for quick investment decisions based on real-time market data.
• Employee Financial Services & Benefits: The payment infrastructure, which comprises payrolls, employee benefits and management, is one of the most promising ones in finance AI. They help reduce the time and headcount required to manage HR functions while providing innovative features such as real-time earned wage access and catching fraud.
• Open Banking: AI can enhance the utility of open banking by offering personalised financial services based on individual consumer data. This creates a continuous cycle of data feeding into AI algorithms, thereby generating insights that make open banking more effective and consumer-centric.
4 Untapped Opportunities in the Global Financial Sector
• Retail Customer Interaction: Although Gen AI and natural language chatbots have become more widespread for Level 1 customer support, AI is yet to prove itself fluent enough in human interaction to replace tellers or simple decision menus for cash deposits and withdrawals. Similarly, Gen AI and natural language processing have not been able to replace the more intricate support processes necessitating human intervention and escalation to managers (Level 2 and 3 support). This is a significant opportunity for banks.
• Underbanked Financial Services: Extending more products to unbanked or underbanked customers remains an untapped area. Such customers fall into the category of late targets for banks as they yield less per account compared to high net-worth customers and carry higher risks. Leveraging AI to assess credit-worthiness and identity verification could result in banks expanding credit to them.
• Advancements in Open Banking: The merging of AI and open banking is bound to produce innovative solutions for consumers and financial institutions. Upcoming developments in technologies, like blockchain and quantum computing, will further enhance data security and computational capabilities, making real-time, high-volume data analysis more efficient. Integrating sophisticated voice-activated AI assistants with open banking platforms could revolutionise consumer interactions.
• Risks Associated with Embedding AI in Finance: Despite the advantages, Gen AI also brings a unique set of risks. These include impaired fairness, bias in results (‘hallucination’), intellectual property risks, privacy concerns and security threats. Furthermore, although open banking promises transparency and control, the merger with AI raises significant regulatory and ethical concerns with data privacy at the forefront.
• Regulators’ Response to AI in Finance: To combat financial fraud, governments have drafted mandatory regulations for financial institutions. However, different countries have different regulatory approaches. These fall into three broad categories; provision of guidance for the use of AI in financial services; monitoring and enforcement of actions to restrict disapproved uses; and rules on permissible uses of AI. The variations in approach, focus areas, implementation standards and timing result in differing outcomes for AI adoption regionally. Whereas the future demands financial institutions take a regulatory-first approach to implementing Gen AI, regulators too need to be adaptive and set forth forward-looking regulations.
The financial institutions that will benefit the most from Gen AI are those that have large amounts of customer data (allowing them to train more nuanced models) and forward-thinking product teams. They are also the ones most likely to initiate partnerships with fintechs; aimed at leveraging innovative solutions and building sophisticated products internally by harnessing proprietary machine learning models. Financial sector leaders need to go beyond an incremental or copy-paste approach and fully embrace the opportunities.
Nabeel Qadeer is Deputy Group CEO, BenchMatrix.
nabeel.akmal@gmail.com
• Retail Customer Interaction: Although Gen AI and natural language chatbots have become more widespread for Level 1 customer support, AI is yet to prove itself fluent enough in human interaction to replace tellers or simple decision menus for cash deposits and withdrawals. Similarly, Gen AI and natural language processing have not been able to replace the more intricate support processes necessitating human intervention and escalation to managers (Level 2 and 3 support). This is a significant opportunity for banks.
• Underbanked Financial Services: Extending more products to unbanked or underbanked customers remains an untapped area. Such customers fall into the category of late targets for banks as they yield less per account compared to high net-worth customers and carry higher risks. Leveraging AI to assess credit-worthiness and identity verification could result in banks expanding credit to them.
• Advancements in Open Banking: The merging of AI and open banking is bound to produce innovative solutions for consumers and financial institutions. Upcoming developments in technologies, like blockchain and quantum computing, will further enhance data security and computational capabilities, making real-time, high-volume data analysis more efficient. Integrating sophisticated voice-activated AI assistants with open banking platforms could revolutionise consumer interactions.
• Risks Associated with Embedding AI in Finance: Despite the advantages, Gen AI also brings a unique set of risks. These include impaired fairness, bias in results (‘hallucination’), intellectual property risks, privacy concerns and security threats. Furthermore, although open banking promises transparency and control, the merger with AI raises significant regulatory and ethical concerns with data privacy at the forefront.
• Regulators’ Response to AI in Finance: To combat financial fraud, governments have drafted mandatory regulations for financial institutions. However, different countries have different regulatory approaches. These fall into three broad categories; provision of guidance for the use of AI in financial services; monitoring and enforcement of actions to restrict disapproved uses; and rules on permissible uses of AI. The variations in approach, focus areas, implementation standards and timing result in differing outcomes for AI adoption regionally. Whereas the future demands financial institutions take a regulatory-first approach to implementing Gen AI, regulators too need to be adaptive and set forth forward-looking regulations.
The financial institutions that will benefit the most from Gen AI are those that have large amounts of customer data (allowing them to train more nuanced models) and forward-thinking product teams. They are also the ones most likely to initiate partnerships with fintechs; aimed at leveraging innovative solutions and building sophisticated products internally by harnessing proprietary machine learning models. Financial sector leaders need to go beyond an incremental or copy-paste approach and fully embrace the opportunities.
Nabeel Qadeer is Deputy Group CEO, BenchMatrix.
nabeel.akmal@gmail.com
