Mumbai, Sept 09: Beams Fintech Fund, in partnership with Alvarez & Marsal (A&M), today launched Beyond the AI Pilot: Scaling Value in BFSI, a report examining the evolution of AI adoption across India’s banking, financial services and insurance sector, and the challenge of translating operational impact into sustained economic value.
The report finds that financial institutions are deploying AI across efficiency, cost reduction, sales, risk and fraud management, and customer experience, with applications spanning customer acquisition, onboarding, underwriting, servicing, collections, compliance and product creation. While operational impact is increasingly visible through productivity, throughput, turnaround time, and service automation, attributable and repeatable P&L impact remains harder to establish.
Evidence from the report highlights the growing impact of AI. Tata Capital has seen a ~30% improvement in underwriting productivity, while Kissht has improved first-time-right rates by 30%. Niyo increased AI-handled customer support from 10% to 90% while keeping support headcount flat despite approximately 4x customer growth. In claims, InsuranceDekho / Artivatic reduced adjudication time from ~6 hours to seconds while retaining human review for complex cases.
Sushil Zaregaonkar, Managing Director, Business Transformation Services, Alvarez & Marsal, said: “The question for financial institutions is no longer whether AI can improve an individual task. It is whether they can redesign the workflow, operating model and governance around that capability to capture the benefit. Institutions that treat AI as a layer added to existing processes may see productivity gains; those that redesign how work gets done have a greater opportunity to translate those gains into structural advantage.”
The report identifies fragmented data, workflow dependencies, integration, governance, security, talent, and accountability as key barriers to scaling AI beyond successful pilots. It also highlights workflow decomposability as critical to determining where AI can be effectively deployed.
For investors, this shift raises a different question: where will durable value accrue?
Sagar Agarvwal, Founder and Managing Partner, Beams Fintech Fund, said: “The AI opportunity in financial services is moving into a more consequential phase. The market is beginning to separate demonstrations of technical capability from evidence of durable business value. For investors, the question is increasingly whether a company is embedded deeply enough in a financial workflow to own an outcome, whether its data or distribution creates defensibility, and whether its economic strength rather than weaken as deployment scales.”
As AI moves from individual use cases towards broader operating-model change, the report places it within the longer evolution of technology in financial services.
Bhavik Hathi, Managing Director and Co-head of Transaction Advisory Group, Alvarez & Marsal, said: “Financial services has continuously evolved through successive technology shifts, from pen-and-paper processes to software-led systems, and then to internet, API and mobile-led models. AI now marks the fourth generation of this evolution, with a fundamental difference: its impact can extend beyond digitizing or automating individual tasks to changing how processes are structured, decisions are made, and human capabilities are deployed. As institutions move from pilots to scaled adoption, the opportunity will be to fundamentally rethink how work is performed, rather than simply adding AI to existing processes.”
The report also finds that build versus buy is becoming less binary, with institutions determining which AI capabilities to own based on strategic differentiation, proprietary data, workflow depth and scale economics, while sourcing more specialist capabilities externally. It maps more than 100 AI vendors active in or adjacent to Indian BFSI, with defensibility increasingly shifting towards differentiated workflows, proprietary data, distribution and demonstrable outcomes.
Taken together, the findings point to a clear shift: AI has moved beyond the question of whether it can work. The next phase will be defined by whether financial institutions can embed it into how work gets done and convert operational gains into economic value that can be measured, repeated and captured at scale.
