Human intelligence and trusted data will define the next frontier of AI in banking

Speaking at a recent Standard Bank AI Conference, senior leaders said the banking industry is moving beyond the first phase of AI adoption, where many organisations have focused on pilots and productivity experiments. The next phase will be about embedding AI into the way banks work, make decisions, serve clients and build new sources of value.

The shift is significant because access to Generative AI models and cloud technology is becoming increasingly available across the financial services industry. Standard Bank believes the real differentiator will sit in the quality and coverage of a bank’s data, the depth of its client relationships, the strength of its governance, its aptitude to learn and adapt with speed and the ability of its people to use AI responsibly and practically.

Andrew van der Hoven, Standard Bank Chief Digital & Product Officer says: “From a product and digital perspective, we must always start with solving real customer and business problems. Generative AI enables teams to build better solutions, faster, and at a scale that was not possible five or six years ago. The opportunity is not only to optimise existing processes, but to reimagine the bank from ground up. We are not thinking like optimisers but thinking and acting like builders. If we can build a system that learns faster than anybody else and compounds, that becomes a durable competitive advantage.”

Standard Bank’s view is that AI will be most powerful when it enhances human capability rather than replacing it. Machines can process information, detect patterns, automate repetitive tasks and support faster decisions. People bring judgement, empathy, ethics, contextual understanding and trusted relationships. The value comes from combining both.

The impact of AI is already being realised across Standard Bank. Standard Bank launched tools at the conference designed to help its employees exploit generative AI to better meet the needs of its customers. A highlight was the launch of FRAME, an in-house customer product development harness, which accelerates how Standard Bank designs, tests and delivers customer solutions. In early use cases, this capability has reduced the time between idea to live solution by up to 70%.

Furthermore, Standard Bank showcased its AI agent framework. A deployed example is an agentic Call Centre voice recognition solution that enables customers to speak freely and be automatically connected with a banker with the right skills for the specific challenge they are facing. This is expected to reduce customer frustration and transferred calls by 25%. Furthermore, a knowledge agent is supporting voice bankers to get the most recent information for customers and reducing the time customers have to spend on a call. These are real solutions for customers.

But Standard Bank cautions that AI success should not be measured by the number of use cases alone. Alongside these, Standard Bank launched its GenAI Foundry, a group-wide capability that enables technology and business teams across the bank to build, deploy and scale generative AI solutions safely and repeatedly, putting the technology directly into the hands of the teams closest to the customer. Rather than centralising development in a single team, the Foundry operates as a federated capability: a shared foundation of models, tooling and governance that any team can draw on to build customer-facing solutions without having to solve the underlying safety, compliance and infrastructure challenges themselves.

The bigger test is whether AI changes the way an organisation works and whether it delivers measurable value for clients, employees and shareholders. Ensuring trust and appropriate guardrails are key in probabilistic systems.

Khomotso Molabe, Standard Bank Deputy CIO says: “AI is not a technology adoption problem, it is a business transformation opportunity. The mistake would be to get married to the technology and lose sight of the value we are trying to create. Technology will keep changing. Our job is to use it to transform the business, while staying humble enough to know that some choices may need to change. At the same time, we must not make decisions that could degrade trust. The best technology strategy is one that allows us to move forward, learn quickly and change course if we are wrong.”

This balance between speed and responsibility is central to Standard Bank’s AI approach. The group has stated its ambition to become Africa’s pioneering AI financial services organisation, with AI expected to fundamentally improve how clients experience the bank; improve how work gets done and create significant value for both the bank and its clients.

Molabe says, “We have invested in building a modern data platform and busy migrating critical data to accelerate its AI ambitions. This is supported by strong governance and maturing data management disciplines. This enables better access for teams who are solving real client problems.”

Standard Bank believes this progress must be supported by a strong data foundation. As AI models become more widely available, the competitive advantage will increasingly move from the technology layer to the data layer. For banks, this means having a fuller, more responsible understanding of clients, including core banking data, behavioural insights and journey data.

“Data is the fuel for all types of AI,” said Molabe. “We continue to build stronger data foundations, stronger governance and better access for teams who need to solve real client problems.”

In everything, trust should remain central to the AI opportunity. AI can help improve financial education, fraud detection, onboarding, credit decisioning, personalised advice and support for entrepreneurs. But it must be deployed with appropriate governance, human oversight, regulatory discipline and respect for client privacy.