Decoding Two Decades: Nilesh Vaidya Discusses the Transformative Journey of Retail Banking and the Rise of AI

Earlier this month, we conducted a poll on LinkedIn to gauge the preferred modes of interaction among retail banking customers. Two-thirds (66%) voted in favor of mobile applications, while over one-in-four (27%) preferred engaging with a human agent. Chatbots ranked third with only 7% of the vote. Unsurprisingly, only 1% choose to visit a bank branch!

Many global banks are trying to shape an effective channel strategy to customer touchpoint choices. Convenience has become an expectation as customers expect hassle-free, on-demand services that fit into their busy lives. This is why 24/7 access to mobile apps, chatbots, and other self-service options are the norm.

Yet, public acceptance for traditional chatbots that typically operate based on defined rules and templates, as we see in our poll, remains low. Capgemini’s World Retail Banking Report 2024, launched this month, reveals that 17% of retail banking customers surveyed do not trust chatbots and an even greater number (61%) contact agents due to their frustration with chatbots resolutions.

In 2024, the bank of the future will see chatbots as an untapped opportunity.

Customer preference for an elevated digital experience is industry-agnostic in current times. Retail banking customers demand personalized interactions or threaten to move to a competing brand. We spoke to Nilesh Vaidya , Global Industry Head of Retail Banking & Wealth Management at Capgemini, to understand how banks will benefit from the integration of AI and generative AI into their everyday operations.

1. Not a day goes by where we don’t hear the words “generative artificial intelligence”. What are you hearing from clients when assessing the current state of technology driven transformation at global retail banks?

We are in a digital-first world and banking is no different. Every area from lending and transactions to bill payments and wealth management is seamlessly being propelled by the power of a unified and transformative digital experience.

It has taken retail banks nearly 15 years of effort and investments to get where we are. Yet, the potential unlocked so far is only just the beginning. Emergence of new-age players such as Chime (US), Starling (UK), and NuBank (Brazil) have compelled the traditional banks and credit unions to shift their attention to innovation. And now, generative AI has opened an array of possibilities.

However, there is a substantial gap between aspirations and reality. While critical functions such as onboarding, lending, marketing, and contact center are now digital, 80% of bank employees rank automation of these key functions as ‘moderate’.

Typically, banks have directed majority technology investment towards front-end transformation, emphasizing immediate customer needs and enhancing user interactions. The disparity in investment is an outcome of the pressure facing banks to deliver a hyper-personalized customer experience with visible enhancements. It is just unfortunate that it arrives at the expense of foundational upgrades necessary to support these advancements.

In discussions with banking executives, I see a notable shift with greater emphasis on more mid- and back-office functions. This shift indicates a dual objective: cost management and establishing a strong foundation for future front-office improvements.

2. Where can generative AI make the most difference in retail banking?

Generative AI's impact on democratizing artificial intelligence has been profound. The 2024 World Retail Banking Report survey of bank CXOs found that 80% of bank executives believe generative AI represents a significant leap in advancing AI technology. As AI applications multiply and mature, it is important for banks to strike a balance between the enhancing customer experience and improving their own operational efficiency.

For instance, we at Capgemini have collaborated with a large global bank to improve their contact center chatbot's conversational AI system with generative AI to improve multi-modal customer journeys, and expect to achieve up to 2x improvement in the containment rate of chatbot from the current level.

I see three key areas as prime candidates generative AI for intelligent transformation in retail banking in the coming years: workforce productivity through copilots, documentation gathering, processing, and validating, and enhancing call center friendliness and efficiency.

Banks can optimize up to 70% of the time spent by banking employees on operational activities through workforce copilots. Similarly, onboarding team spends nearly 91% time on operational documentation, and compliance activities.

By utilizing cognitive capabilities, including AI and ML, intelligent document processing platforms automate tasks such as document categorization and data extraction, helping workforce optimize up to 66% of the time spend in onboarding - resulting into reduced cost, efforts, risk, and improved customer experience.

One of the most promising use cases we are seeing for generative AI in banking is the use of conversational agents or chatbots. In a typical scenario, a retail banking customer reaches out to the bank via a chatbot, the chatbot understands the intent, retrieves data from various sources, and responds using generative AI for a natural, personalized conversation.

3. What are the challenges to successful Generative AI deployments in retail banks?

Banks face hurdles in adopting generative AI due to legacy system complexities, integration challenges, and fragmented data across departments. Stringent regulatory frameworks and skill shortages add extra layers of complexity, while inherent risk aversion and cost concerns contribute to a cautious approach.

We assessed 250 banks based on business and technology criteria to gauge their readiness for intelligent transformation. Globally, 41% of banks performed at an average level on various parameters, with 29% scoring low. Surprisingly, only 4% of banks achieved high scores on both parameters, signaling limited readiness to fully embrace and scale intelligent transformation.

It is also important that when AI system makes a decision, it should be explainable so humans can quickly comprehend the how and why behind it. Banks, along with ensuring explainability in AI, can leverage these models for better outcomes by actively mitigating potential biases, inaccurate outputs, privacy concerns and the misuse of the technology by fostering responsible AI development practices.

To make this work, I suggest starting from the basics: creating a modern system to handle data efficiently, using smarter algorithms, and putting in simple rules to ensure AI is used responsibly.

In addition, banks need to identify and set benchmarks to measure the outcomes of their generative AI investments. The lack of these guardrails could lead to the concerning trend of "generative AI silent failure," where issues persist without immediate recognition.

This could begin with establishing an AI observatory and leveraging KPIs for continuous monitoring, reporting, and improving AI governance. These quantifiable benchmarks not only help identify potential risks and anomalies but also ensure alignment with standards and objectives, promoting proactive risk management and compliance.

In 2024, banks face a heightened need for efficiency. Balancing revenue generation with cost control is critical, prompting institutions to explore innovative value creation strategies even within capital constraints. This focus on efficiency transcends necessity, potentially becoming a key differentiator in a competitive landscape.