
RBI Governor Sanjay Malhotra warned on Tuesday that the biggest risk from deploying AI in banks is the erosion of human judgment and accountability. Speaking at FIBAC 2026 in Mumbai, he listed…
RBI Governor Sanjay Malhotra warned on Tuesday that the biggest risk from deploying AI in banks is the erosion of human judgment and accountability. Speaking at FIBAC 2026 in Mumbai, he listed six key AI risks: explainability, bias, concentration, vendor dependence, data privacy and cyber vulnerabilities. Compliance with the DPDP Act alone is not enough, he said. Bank CEOs echoed concerns: PNB chief Ashok Chandra said AI agents in call centres must stay under human control, while SBI chairman CS Setty urged the next wave of AI-led banking to reach rural India and small businesses. Malhotra also compared AI’s potential in lending to UPI’s impact on payments, citing alternative data like GST filings to expand credit.

The FIBAC 2026 conference showcased a curious split. Bank CEOs rush to deploy AI agents in call centres and lending, while admitting they have not solved the control question. RBI Governor Malhotra’s six risks, from black-box decisions to vendor concentration, are brushed aside as compliance hurdles rather than design principles. The hype that AI will magically include rural India ignores Setty’s own admission that pilots rarely scale. The real test is not how many use cases banks can list, but whether any board has actually overruled an AI decision in a lending or fraud case.
Sources (5): thehindu.com, bfsi.economictimes.indiatimes.com, bfsi.economictimes.indiatimes.com (2), timesofindia.indiatimes.com, timesofindia.indiatimes.com (2)
This story was synthesised by AI from the 5 sources linked above.
Updated: this story now draws on 5 sources.