By Aditi Kashyap
Las Vegas, Dec 25 (PTI) Indian financial institutions are rapidly moving beyond the proof-of-concept (POC) stage with Generative AI and now deploying the technology at scale to drive revenue rather than just cost efficiencies, a top executive at Amazon Web Services (AWS) has said.
Kiran Jagannath, Head of FSI and Conglomerates at AWS India and South Asia, stated that while the initial conversation around Artificial Intelligence (AI) focused on productivity, the narrative is shifting toward "agentic AI", autonomous agents that can execute complex tasks, which he predicts will be the dominant technology narrative in Indian banking over the next three years.
"It's a great time to be in financial services in India. If you look at what's happening over the last five to ten years, the entire digitisation movement in India is bringing a whole lot of opportunities across the board.
"The pandemic changed how banks, insurers, and others interact with and engage their customers. Pre-pandemic, we were still largely a branch-based engagement model. But post-pandemic, during lockdowns, when branches were closed, digital channels became the primary channels of engagement," Jagannath said in an interview to PTI.
He emphasised that AI-driven efficiency is now directly translating to top-line growth. Citing an example, he noted that fintech player Cashfree used Generative AI to accelerate merchant onboarding, reducing a process that took many hours to less than 10 minutes, thereby preventing business loss.
"Last year, we were talking a lot about productivity and efficiency that generative AI is driving. A lot of those use cases are in production already this year. This year, what we're talking about is something even further, which is how agents can start driving more efficiency in business processes.
"And even though the word uses efficiency, these business process efficiencies actually drive revenue and top-line growth," he noted.
Jagannath highlighted a significant structural shift in the sector: the modernisation of legacy core banking systems. For years, banks built digital layers on top of older systems, but that is changing.
He observed that banks are adopting different strategies, with some "hollowing out the core" to modernise it, while others are taking a "dual-engine approach", where new customers are onboarded onto modern, cloud-native cores while the old system is progressively upgraded.
Jagannath also pointed to the BSE as an example of leveraging cloud technology for critical risk mitigation. BSE, working with an AWS partner firm, built a system capable of coming online within 15 minutes if some stockbroker platforms go down, ensuring investors can still square off positions in the futures and options (F&O) segment.
Looking ahead, Jagannath stated that the role of the Chief Information Officer (CIO) has evolved, with technology leaders now having a "seat at the table" on boards.
"In three years, I think the one that we are going to be talking more about is how agents have been innovating various experiences, how they have been transforming various business and technology processes," he added.
While acknowledging the potential of quantum computing, he asserted that agentic AI remains the most immediate and transformative prospect for the sector.
"With technology making it very simple for even business users to build something...I think collaboration, where business and technology stakeholders come together and solve business problems, is going to be the need of the hour. That might require a cultural shift within the organisations, and that will have to be obviously top-down," Jagannath said.