Fraud can outrun conventional bank systems in the AI era, warns SBI’s Setty
State Bank of India chairperson C.S. Setty on Tuesday warned that the wider use of artificial intelligence (AI) is creating new vulnerabilities for banks, and that oversight, model risk, and human accountability will have to be rethought as systems become more autonomous.
“Fraud can evolve at a speed that conventional systems may struggle to match,” Setty said at FIBAC 2026, warning that Indian banks will have to strengthen their defences at the same time as they widen their use of AI.
He said increased use of the technology is introducing new vulnerabilities across the sector, and that more sophisticated technology can create even more sophisticated cyber threats.
He also called for a rethink of how banks supervise the technology internally. “Increasingly, autonomous AI systems require us to think differently about oversight, model risk, transparency, and responsibility,” Setty said, adding that the approach to governance will have to evolve as these systems take on more of the work.
On where the line should fall, he tied the level of scrutiny to what is at stake in a given decision. “The more consequential the decision, the greater must be our emphasis on explainability and human accountability,” he said.
Setty argued that the sector cannot treat any of this as a purely technological question, since banking is ultimately built on trust. He described trust as a theme that industry leaders have been returning to and said that the core of banking is trust.
Future of Indian banking
The workforce implications were framed in similar terms. The banking workforce of the future will work very differently from the workforce of today, Setty said, and equipping people for an AI-led environment is now part of the industry’s task.
The risks were set out alongside an expansion agenda rather than against it. Much of the early momentum in Indian banking has understandably been in retail, he said, where large volumes of data and transactions create immediate opportunities. “The next wave of AI in banking must take us deeper into the economy, to rural India, small businesses, and customers whose financial histories may not fit conventional models,” Setty said.
Agriculture in particular presents a significant opportunity, he said, with AI able to support better farm-level decisions while data-driven risk assessment, digital records and satellite imagery help banks improve both credit access and portfolio management. “The challenge is to take these capabilities beyond pilots and make them affordable, practical, and accessible at the last mile,” he said.
The measure of success, in his view, should be economic rather than technological. “The real test of AI will not be how sophisticated our technology becomes, but what it enables us to accomplish,” Setty said, pointing to timely credit for a small business, better risk management for a farmer, and a safer and more resilient financial system. Most importantly, he said, the industry has to achieve all of this while preserving the confidence that people place in their banks.
