Artificial intelligence (AI) has become one of the hottest topics in fintech, resulting in many organisations looking to get involved in the space. However, research from Capgemini, the tech consulting and information firm, found only six per cent of retail banks have established an enterprise-wide roadmap for the use of generative AI (gen AI) capabilities at scale.
A lack of funding has meant expansion is becoming difficult, resulting in many organisations looking inward at how they can improve their services amid an uncertain economy. The research published by Capgemini found that many chief experience officers (CXOs) plan to increase investment in digital technology by up to 10 per cent to help achieve inward growth. However, despite the desire, many banks cannot scale new technologies, like generative AI to ensure they are being used most efficiently.
For this report, Capgemini evaluated 250 retail banks across diverse business and technology parameters to understand their infrastructure data maturity and commitment to artificial intelligence.
It found most banks are ill-prepared to thrive in an intelligent banking future. Globally, only four per cent of retail banks achieved a high score on business commitment and technology capabilities. Meanwhile, 41 per cent scored average, indicating a widespread lack of readiness to embrace and effectively implement intelligent transformation.
Regional disparities further underscore this issue. In North America, 27 per cent of banks displayed low readiness, followed by Europe with 31 per cent, and Asia-Pacific (APAC) exhibiting a significant lag, with 48 per cent of banks scoring low.
Banks must act quickly to avoid ‘generative AI silent failure’
Focusing on intelligent solutions, that are embedded with AI-driven capabilities, will allow banks to navigate ongoing structural challenges, ultimately ensuring sustainable growth. However, success must be measurable: among those surveyed, just six per cent of banks have established key performance indicators (KPIs) to measure AI impact and continuous monitoring. More than 60 per cent of banks are still identifying and developing KPIs, while 26 per cent of banks that have already set some KPIs are not measuring them.
According to the report, banks risk succumbing to ‘generative AI silent failure’ due to the delayed realisation of suboptimal results and outcomes from their experiments with the technology. For instance, just two per cent of executives indicate they are regularly tracking the business impact KPIs of their generative AI performance.
In addition, 39 per cent of executives express dissatisfaction with the outcomes of their AI use cases further reinforcing this disconnect. To combat this, the study suggests banks set up an AI observatory to track, monitor, and report AI and generative AI real impact when implemented at scale.
Gen AI must be explained
“One year after generative AI cemented itself as a core boardroom conversation, we’re seeing how banks risk becoming technological laggards if they aren’t rapidly adopting solutions and preparing to take advantage of its capabilities,” said Nilesh Vaidya, global industry head of retail banking and wealth management at Capgemini.
“Generative AI can have a lighthouse effect when used responsibly and wisely across operations. There is also a need for increased efforts to make gen AI explainable and appropriately transparent.
“The time to act is now to establish practices that build much-needed trust and customer intimacy. Success will come down to developing a roadmap that balances hype with a pragmatic, traceable and measurable approach.”
Bank employees welcome generative AI copilots
Generative AI holds massive potential to elevate efficiency and customer experience across the retail banking value chain. Over two-in-three (70 per cent) bank employees are focused on operational activities. This rises to 91 per cent for those employees on customer onboarding teams, leaving little time for customer interactions.
Over 80 per cent of bank employees give a ‘moderate’ rating to automation across their functions (onboarding, lending, marketing, contact center), identifying a significant gap between the bank’s aspirations and reality.
Bank employees reported being most enthusiastic about generative AI copilots’ potential to automate fraud detection, data visualisation and analytics automation, as well as drafting and sending personalised content to customers. The report determines that banks could optimise up to 66 per cent of the time spent on operations, documentation, compliance, and other onboarding-related activities through AI-powered intelligent transformation and generative AI copilots.
Conversational AI could alleviate customer call abandonment
The pandemic shifted customer service offers across to digital channels as self-service tools like chatbots became the norm. Despite this change, customers express dissatisfaction. Nearly two-in-three (61 per cent) bank customers contacted agents because they were unhappy with chatbot resolutions, while 17 per cent simply distrusted chatbots and preferred human agents.
Traditional rule-based chatbots lack the flexibility and adaptability of advanced AI-driven systems due to their inability to handle complex or unanticipated queries. More than 60 per cent of customers rated their experience with chatbots as only average. These conditions mean that call abandonment is on the rise, reaching 12 per cent for Tier I banks and nearly 18 per cent for Tier II banks globally.
According to the report, banks should create intelligent contact centres that leverage chatbots with conversational AI capabilities and intelligent copilots to assist agents in their day-to-day tasks.
