Leadership Spotlight
The Decision Is Still Human with Nithin Reddy
The future of AI in financial services is ultimately simple: less time assembling the decision, more time making it well.
A Conversation with Nithin Anajeneya Reddy, Vice President, Data & AI Specialist, Financial Services
Financial institutions don’t lack AI initiatives, models, platforms, or data. The harder question is whether any of it is materially improving the decisions that matter.
For Nithin Anjaneya Reddy, that is where the conversation needs to move next. In lending, risk, and portfolio management, he sees AI’s real value not in removing human judgment, but in eliminating the fragmented, manual work surrounding it. The opportunity is to connect trusted data, evidence, AI, governance, and domain expertise so decisionmakers can understand what changed, why it matters, and where they need to act.
1. You’ve said AI is changing how decisions are made, not who makes them. What does that mean in practice?
When I look at where AI can create the most value in financial services, I don’t start with replacing human judgment. I start with the work that gets in the way of it.
In lending and risk, teams still spend significant time gathering financials, reviewing documents, reconciling information across systems, and understanding what has changed. AI can compress that work. It can bring relevant information forward and identify meaningful signals, allowing underwriters, risk officers, and portfolio managers to spend more time on the decisions that require their expertise.
The more interesting shift is from periodic, backward-looking reviews toward continuous decision support. Rather than waiting for the next credit review to discover that a borrower’s position has changed, teams can identify emerging signals earlier and understand what is driving them.
That is how I think about decision intelligence: AI strengthens the decision process. Accountability stays with the human.
2. Banks already have enormous amounts of data. Why doesn't more data automatically lead to better decisions?
Banks rarely have a shortage of data. The challenge is knowing which data matters, whether it can be trusted, and what it means in the context of a specific decision.
Take an early-warning alert. The alert itself has limited value unless a portfolio manager can quickly answer: What changed? Why did it trigger? What evidence supports it? Is it material? What should happen next?
That is why I don’t see data, governance, domain expertise, and AI as separate conversations. They need to come together inside the workflow.
A model might summarize a credit agreement. But a useful lending solution needs to understand which terms matter, how they affect a covenant or credit decision, and when an exception requires attention. Trust, explainability, lineage, and human oversight aren’t features you bolt on afterward. They are prerequisites for putting AI into real financial decision-making.
3. What are financial institutions getting wrong when they start an AI program?
Too often, the starting question is, “Where can we deploy AI?”
I would rather ask, “Where is the process breaking down?”
Why does a credit review take weeks? Where are analysts repeatedly gathering or re-keying information? Where do decisions get stuck? Which activities genuinely require expert judgment?
Once you start there, the conversation changes.
In underwriting, for example, AI can support document intake, financial spreading, policy checks, and preparation of a first-draft credit memo. But the underwriter still challenges assumptions, evaluates exceptions, and owns the recommendation.
The objective shouldn’t be to make an inefficient process run faster. It should be to redesign the workflow around a better decision.
4. Where does the bigger opportunity lie beyond automating individual steps?
There is clear value in financial spreading, document intelligence, underwriting support, covenant monitoring, portfolio surveillance, and early-warning signals. But I believe the real opportunity comes from connecting them.
Imagine a portfolio manager receiving an alert that something has changed with a borrower. Instead of searching across multiple systems and documents, they immediately see what changed, why it matters, the evidence behind it, and where they need to focus.
Then comes the part that matters most: judgment.
Is this noise? Is it a temporary variance? Or are we looking at the beginning of meaningful deterioration?
That’s when AI moves beyond productivity and starts becoming decision intelligence.
5. If AI pilots aren't the measure of progress, what separates the institutions that will actually create value?
AI earns the right to scale when people stop treating it as an experiment and start relying on it to make the workflow materially better, earning user trust, and then expanding. That makes scaling AI as much an operating-model challenge as a technology challenge.
Financial institutions already have models, platforms, and data programs. Simply adding another AI capability isn’t necessarily the answer.
The institutions that get this right will bring trusted data, domain expertise, strong governance, and human judgment together in the workflows where real decisions happen.
The goal isn’t more AI. It’s better decisions.
AI may radically change how financial work gets done. It can read the documents, surface the signals, connect information, and prepare the analysis. But Nithin’s argument draws a deliberate line between assembling a decision and making one.
That distinction matters. Because if every financial institution eventually has access to powerful AI, simply having technology stops being an advantage.
The breakthrough won’t be AI making the credit decision. It will be reaching the moment of judgment faster, with better evidence and fewer blind spots.
Latest Posts
The Pathway Gap: Why Proven Programs Still Lose the Budget Debate
Proving a program works is only the start. Find out why development evidence often doesn’t reach budget decisions, and how six questions and four lenses help close the pathway gap.
Navigating Geopolitical Uncertainty in Investment Banking
Geopolitical uncertainty now shapes which deals move forward, how assets are valued, and whether deals close. This paper shows how dealmakers can turn geopolitical disruption into decision-ready transaction insight.