Indian financial services has moved past the AI pilot; the harder test is value
India’s financial services industry has spent the past three to four years investing in AI. It has also had enough time to establish that AI can improve specific tasks. The more difficult question now is whether those gains can be converted into changes in how banks and NBFCs actually operate.
Radhika Saigal, Financial Services Consulting Leader at EY India, sees the industry at an “interesting inflection point”. The returns from AI are real, she says, but they remain concentrated disproportionately in productivity and efficiency rather than in fundamental business-model transformation. That distinction is becoming increasingly important. Improving an existing task does not necessarily change the economics of the process around it.
“Financial institutions have become very good at proving that AI can work,” Saigal says. “The harder question is can you redesign an entire process around AI and actually change the economics of that process?”
The gap between demonstration and value
The difference can be illustrated through lending. An AI system that summarises a credit document may save an employee 20 minutes. That is a tangible efficiency gain, but it does not by itself alter how a lending operation works.
The more consequential proposition would be to use AI across document ingestion, analysis, underwriting, decision support and monitoring. In that scenario, AI is no longer being applied to one task within an existing process. The process itself is being reconsidered. This is what Saigal describes as the “pilot-to-production-to-value” gap.
The industry, in her assessment, has become increasingly capable of demonstrating that an AI use case works. The harder part is taking it into production and establishing that it has changed cost, speed, capacity, risk or another meaningful business outcome.
“The next phase of ROI will not come from having 500 AI pilots,” she says. “It will come from taking 5–10 high-value domains and rewiring them end-to-end around AI.”
The numbers here are not a measure of industry adoption but Saigal’s prescription for where institutions should concentrate their efforts. Her larger point is that AI investment needs to move away from the accumulation of use cases towards the redesign of selected business processes.
The question, therefore, shifts from “What AI use cases can we deploy?” to “What business outcomes can we fundamentally change with AI?”
The difficult part begins in the live banking environment
Indian banks and NBFCs have several characteristics that make them significant environments for AI deployment. They operate at enormous transaction volumes, have substantial data, digitally mature customers and sophisticated technology organisations. Those advantages also create complexity.
AI has to coexist with legacy technology, fragmented data, regulatory requirements, cybersecurity and model-risk controls. Building a model is therefore only one part of the deployment problem.
“The issue is not whether an Indian bank can build an AI model,” Saigal says. “The issue is whether that model can safely participate in a live banking workflow.” That means the model has to operate with the right data and permissions, while meeting requirements around explainability, auditability and human oversight. It also has to work reliably when exposed to the scale and operational complexity of millions of transactions or customer interactions.
The organisational structure of traditional technology delivery presents another constraint. Technology teams have typically built platforms while business teams have defined processes. AI cuts across that separation because the technology and the business problem often have to evolve together.
The challenge, consequently, is not simply to acquire or build better AI models. It is to connect those models to the systems, controls, processes and decisions through which financial institutions actually operate.
Why the engineer is moving closer to the business
That helps explain the growing relevance of the forward deployed engineer, or FDE. The role places an engineer close to the business problem rather than several organisational layers away from it. Instead of moving a requirement through a conventional sequence of business, product, architecture, engineering and operations, the FDE works directly with the business, understands the workflow, experiments with AI, builds the solution, takes it into production and iterates based on what happens in the real world. The importance of the role lies in the last mile of enterprise AI.
“The model itself may be relatively easy to access,” Saigal points out. “The difficult part is connecting it to enterprise data, legacy systems, APIs, controls, workflows and human decision-making.” That makes the FDE less a new version of an existing engineering role and more a response to the organisational distance that can separate technology development from business transformation.
For financial services, where AI has to function within tightly controlled operating environments, that distance can become particularly consequential.
GCCs face an ownership test
India’s financial services GCCs potentially have an advantage in this transition because they already bring together three capabilities required by the FDE model which are domain knowledge, engineering talent and proximity to global products and platforms. But having those capabilities is not the same as having ownership.
The traditional GCC model has been built largely around scale, efficiency and execution. Saigal sees an opportunity for Indian GCCs to take on a different role in which they become AI product and engineering engines for their global organisations. That would mean moving from “We execute what headquarters gives us” to “We own the problem, build the solution and scale it globally,” she says.
The distinction matters. If an India-based team builds an AI pilot that is subsequently sent to headquarters for deployment, the underlying operating model has changed little. The more significant shift would come if India-based teams were given responsibility for the AI product lifecycle from problem definition through deployment and scaling.
Financial services adds another dimension. An engineer who understands AI alongside banking’s risk, compliance, data privacy, controls and legacy platforms is operating at the intersection of technology and regulated business processes. But the opportunity depends on ownership being conferred, rather than simply adding more AI work to existing GCC responsibilities.
The CIO’s AI decision is really a process decision
For a mid-sized private bank, Saigal’s approach is to start with business outcomes rather than the number of AI experiments. She suggests focusing on high-value domains such as lending, customer servicing, operations, fraud, compliance or software engineering and defining specific outcomes around cost-to-serve, turnaround time, revenue, risk reduction, productivity or customer experience.
The teams pursuing those outcomes should also be organised differently, bringing engineers, product specialists, domain experts, data specialists and risk and control expertise together around the problem.
The larger change, however, is to redesign the workflow rather than simply attach AI to it. A co-pilot added to an inefficient process may make that process marginally faster. It does not necessarily remove the inefficiencies embedded in the process itself. “A co-pilot that sits on top of an inefficient process may make that process slightly faster,” Saigal says. “An AI-native process can fundamentally change how work gets done.”
Governance has to be treated in the same way. In financial services, responsible AI cannot be reduced to a review at the end of development. Security, privacy, model risk, explainability, audit trails and human oversight have to be incorporated into the architecture from the outset.
The race is from prototype to measurable value
The ingredients required for this transition are not necessarily in short supply in India. Saigal points to engineering talent, data and financial-services scale. What needs to change is how closely those capabilities are brought together around the business problem. “If I had to identify just one thing, it would be the speed at which we move from experimentation to ownership and deployment,” she adds.
That makes the next phase of AI adoption a test of organisational execution as much as technology capability. The institutions that matter most may not be those with the largest AI budgets, but those capable of moving an idea through the chain from business problem to prototype, production and measurable value.
“The next generation of AI leaders will not necessarily be the institutions with the largest AI budgets,” Saigal says. “They will be the institutions that can take an idea from a business problem to prototype to production to measurable value faster than everyone else.”
For Indian financial services, the competitive question is therefore becoming narrower and more demanding. It is no longer simply whether an institution can experiment with AI, or even whether it can put an AI application into production.
The harder test is whether it can redesign important parts of the business around the technology and make those changes work at scale.
“The competitive advantage will not be who experiments with AI first,” Saigal concludes. “It will be who industrialises it first.”
