SBI to deploy ‘agentic AI’ for customer service

Mumbai: State Bank of India aims to integrate agentic AI for customer interactions as part of its long-term technology strategy. An agentic AI system can process customer requests, suggest solutions, and complete transactions independently, functioning like a relationship manager.
Speaking at a Microsoft event, Nitin Chugh, deputy managing director, SBI, said the country’s largest bank is using a GenAI-powered approach to transform its banking services. He added that SBI is first deploying agentic AI for bank staff, with the ultimate goal of extending it to customers.
SBI is transitioning to a hybrid cloud infrastructure that integrates its private Meghdoot cloud with the public Azure cloud. This move ensures data security while leveraging the scalability and flexibility of public cloud services.
The bank has already developed AI use cases. One such initiative is the Deceased Account Settlement Queries Chatbot, which assists branch and business unit staff in handling inquiries related to deceased account settlements. It offers scenario-based query resolutions aligned with internal SOPs and circulars, reducing friction in customer interactions, ensuring timely and accurate information, and improving employee productivity.
Another AI-driven solution is the generative AI-powered askSBI chatbot, designed to assist employees in handling complex business scenarios. It serves as a centralized point of contact for internal staff, improving operational efficiency and reducing reliance on manually curated documents. By streamlining operations, the chatbot enhances service quality for SBI’s workforce.
According to an IDC survey commissioned by Microsoft, the BFSI sector sees the highest return on investment from generative AI, with an average return of nearly 5x for every $1 invested, compared to an average ROI of 3.4x across industries.