With banks having already crossed several stages of gradual digitalization, they are now poised to undergo another significant paradigm shift in the form of autonomous banking. This is more than simply digitizing tasks; this involves taking decision-making to the next level, where intelligent and reliable systems can take autonomous decisions in compliance with certain guardrails.
Autonomous banking is about introducing intelligence, accuracy, and speed in core processes for better outcomes. This can be made possible with the maturation of certain components, including reliable AI models, clean data environments, and modular architectures with an API-first approach. These will help banks move from workflow-based reactive approaches to proactive systems based on optimization of business processes for better results.
Musharrat Shahin, Chief Sub Editor, FE B2B, interviews Monu Kurien Mathew, SVP & Head – Business Solutions (BFSI) at HCLTech, on autonomous banking.
As banks move beyond digital transformation, what is the real business shift to autonomous banking, and why is it becoming critical now?
For years, banks have been digitizing processes. Autonomous banking is different — it is about redesigning the operating model itself. The shift is from people using technology to execute work to technology executing work under human oversight.
The reason this is happening now is that the underlying building blocks have finally matured.
AI models are more reliable, data platforms are cleaner, and modular architectures allow intelligence to be composed across systems.
Banks have now moved from proofs of concept to production-grade systems. So banks can now shift from reactive workflows to proactive systems that anticipate what's needed and act ahead of time. That's the real inflection point.
In simple terms, digital transformation made banking faster; autonomous banking makes it smarter. And in a market where speed, precision, and personalization all matter at once, that shift is becoming critical.
With CXOs focused on returns, what are the most measurable outcomes banks are seeing from autonomy, cost savings, faster decisions, or better risk control?
The short answer is - all three. But the most visible impact is speed. Processes like onboarding, underwriting, and credit decisions are moving from days to near real-time, and that directly improves customer conversion and experience.
The second big impact is cost. When routine work is handled intelligently, banks can process higher volumes without scaling manual effort at the same rate. That improves productivity and lowers cost-to-serve.
The third, and in many ways the most strategic, is risk. AI-led autonomy allows banks to make faster decisions without losing control, because risk is assessed continuously, not just at fixed checkpoints. So, this isn't just about efficiency. It's about faster decisions, lower operating friction, and better-quality outcomes at the same time.
There is a lot of hype around AI. What are some common myths about autonomous banking that you think need to be corrected?
The rapid progress in AI has created genuine momentum and opened up powerful possibilities for banks. At the same time, it has also led to some understandable misconceptions that are worth addressing.
The first misconception that we push back constantly is that autonomous banking is just automation with a new label. It is not. Automation improves tasks; autonomy rethinks how decisions and workflows happen end-to-end.
Another misconception is that banks need a completely new technology stack before they can start. In reality, many banks already have building blocks. The real challenge is orchestration, not invention.
And perhaps the biggest misconception is that autonomy reduces the role of people. In fact, it does the opposite. It takes humans out of repetitive execution and moves them into higher-value roles, oversight, judgment, exceptions, and strategy. So, autonomy is not about replacing human judgment; it is about using it where it matters most.
Given that legacy systems still dominate, how can large banks realistically move toward autonomy without a complete overhaul?
Large banks do not need a big-bang replacement to become more autonomous. In fact, that is usually the least practical route. What they need is smarter architecture around the core.
AI creates an intelligence layer that can sit above legacy systems, orchestrating decisions and workflows while the core continues to act as the system of record. That means banks can modernize progressively using APIs, data virtualization, and automation, without ripping out decades of infrastructure overnight.
What we are really seeing is gradual core hollowing. Intelligence moves outward first, dependency on the rigid core reduces over time, and optionality increases. So autonomy is not about replacing legacy in one shot; it is about surrounding it with intelligence and evolving deliberately.
In a highly regulated environment, how do you ensure autonomous systems remain transparent, auditable, and compliant at scale?
In banking, if autonomy is not auditable, it is not usable. Regulation is not a constraint around autonomous banking — it is part of the design requirement.
The most effective model is what I describe as “Human plus Machine.”
Machines handle the bulk of routine decisions — the ones that are within well-defined parameters. Humans are responsible for exceptions, edge cases, and anything that requires judgment outside established guardrails. But critically, humans also set and supervise those guardrails. The AI does not get to define its own boundaries.
In many cases, autonomy can actually improve compliance discipline because it reduces inconsistency and creates a stronger audit trail. The goal is not just to make decisions faster, but to make them defensible. In a regulated industry, efficiency matters, but trust matters more.
With use cases like autonomous mortgages reducing timelines, what have you learned about balancing speed with risk and accountability?
Mortgages are a great example to dig into, because it's a process that has historically been slow for understandable reasons. You are dealing with large amounts, long tenures, varied income structures and the documentation is incredibly varied.
But the real lesson is that speed only works when risk moves with it. In autonomous models, risk assessments do not sit at the end of the process; they run continuously as data moves through the workflow. That makes the journey both faster and more controlled.
It also changes the role of teams. Humans are no longer reviewing every standard case; they can focus on exceptions, anomalies, and judgment-heavy situations. The best autonomous models do not trade speed for control — they improve both together.
As AI takes over more decisions, where should banks still rely on human judgment in the Human + Machine model?
Autonomy works best when responsibilities are clearly defined. Machines should handle what is predictable, repeatable, and rules-based. That is where they bring the most value — speed, consistency, and scale.
Humans should step in where context matters: borderline cases, policy exceptions, regulatory interpretation, ethical trade-offs, or moments of market uncertainty. Those decisions require nuance, accountability, and experience.
That is why Human + Machine should be seen as a design principle, not a compromise. Machines bring precision and consistency; humans bring judgment and responsibility. And in banking, judgment is still the difference between a fast decision and a responsible one.
With global competition rising, will autonomous operations become a true differentiator or a basic expectation for all banks?
In the near term, autonomy will be a clear competitive differentiator. Early adopters and movers will gain advantages in cost efficiency, speed in decision-making, customer experience, and risk precision, and those benefits will compound over time.
Over the longer term, it will become a baseline expectation.
However, the real deal-breaker will be who uses AI tools to rethink the operating model. Banks that automate isolated processes will improve. Banks that redesign journeys end-to-end will lead. That is where the competitive gap will open up.
