Reserve Bank of India (RBI) Governor Sanjay Malhotra has urged Indian banks to embrace artificial intelligence (AI) as a major tool to transform lending, financial inclusion, customer service, operational efficiency and fraud detection. He also warned that rapid AI adoption must be backed by strong governance, explainability, data privacy, cybersecurity and human oversight.
While speaking at the FIBAC 2026 conference in Mumbai, RBI Governor Malhotra said AI could define the 2020s for Indian banking in the same way liberalisation shaped the 1990s and digitalisation transformed the 2010s.
“Artificial Intelligence is not a single technology to be procured, nor a project to be completed. It is a new way of doing business, of running a bank,” Malhotra said.
He said banks to decide whether they will deliberately shape their AI journey or allow the technology to shape them by default.
AI can expand access to credit
Malhotra said AI could fundamentally change the economics of credit delivery, particularly for borrowers who have limited formal financial histories.
STORIES YOU MAY LIKE
Traditional lending models can struggle to assess new-to-credit borrowers, gig workers and small businesses without formal books. AI models can instead use alternative information such as cash flows, GST filings, utility payments and digital footprints.
This could significantly expand the pool of borrowers considered “bankable” by lenders while lowering the marginal cost of underwriting loans.
AI-powered credit risk models could also help banks identify emerging stress earlier. Liquidity forecasting and scenario analysis could allow lenders and regulators to detect problems before they appear in traditional financial statements.
AI could boost financial inclusion
The RBI Governor said AI could become a major accelerator of financial inclusion in India. Voice-based interfaces in Indian languages could make banking easier for customers who face language barriers. Predictive models could also identify borrowers at risk of default early enough for banks to provide support instead of simply beginning recovery action.
India’s existing digital public infrastructure gives the country an advantage in deploying AI, Malhotra said.
He pointed to Aadhaar, UPI, DigiLocker, ONDC, Account Aggregator and the Unified Lending Interface (ULI) as platforms that can support further innovation.
“AI, layered on top of this stack, has the potential to do for financial judgment what UPI did for financial transactions: make it instant, granular, and available to the last mile,” he said.
AI can improve customer service and reduce costs
AI can also help banks improve customer service and productivity, according to Malhotra.
AI-assisted relationship managers could identify suitable products and risk alerts, allowing employees to serve more customers. AI could also support grievance redressal and provide personalised financial guidance.
Banks could use AI to automate document processing, reconciliation and internal audit sampling. The technology could also automate transaction reporting and regulatory returns.
This could lower compliance and operational costs while allowing skilled employees to focus on work that requires human judgment.
AI can help banks fight fraud
Malhotra also highlighted AI’s role in combating digital fraud. “Fraud today moves at the speed of an API call,” he said, adding that traditional rules-based fraud systems can struggle to keep pace with criminals who constantly change their methods.
Machine-learning models can continuously learn from transaction patterns and identify anomalies in real time.
This could allow banks to use AI to detect fraudulent transactions before losses materialise.
RBI flags seven key risks from AI
While the RBI supports greater AI adoption, Malhotra warned that banks must manage the risks associated with the technology.
The first is the “black box” problem. Advanced AI and generative AI models may not always explain how they reached a particular decision.
If an AI system rejects a loan application, for instance, both the customer and regulator should be able to understand the reason behind the decision.
The second risk is bias and exclusion. AI systems trained on historical lending data could reproduce existing biases against certain geographies, occupations or communities.
Malhotra said fairness must be built into AI systems from the beginning rather than treated as a compliance requirement.
The RBI also warned about concentration and herding risks. If several banks rely on the same AI models or technology vendors, a common error or vulnerability could become a system-wide problem.
Similar AI-based trading models could also cause banks to behave in the same way during periods of market stress, potentially increasing volatility.
Banks must manage vendor and data risks
Third-party dependence is another concern. Smaller banks may rely on technology vendors rather than develop AI models themselves.
Malhotra said outsourcing arrangements must include AI-specific accountability, audit rights, the ability to seek explanations and a credible exit plan if a vendor or model needs to be replaced.
Data privacy and security will also become increasingly important as AI systems require large amounts of information.
“The Digital Personal Data Protection Act is the floor, not the ceiling, of what customers should expect from their bank,” Malhotra said.
The RBI also flagged cyber risks, including data poisoning, model manipulation and attacks designed to fool AI-based fraud detection systems.
‘The model decided’ cannot be an excuse
The RBI Governor’s strongest warning concerned the possible erosion of human judgment and accountability. He said banks cannot shift responsibility for decisions to algorithms, regardless of how sophisticated the technology becomes.
“The model decided’ can never be an acceptable answer to a customer, an auditor, or the Reserve Bank,” Malhotra said.
Banks must retain the ability to explain AI decisions, intervene when necessary and override a model where required.
RBI asks banks to strengthen AI governance
The RBI wants banks to treat AI governance as an immediate priority rather than a future compliance exercise.
Banks should maintain a complete inventory of every AI system they use, including AI embedded in products supplied by vendors. They should also establish board-approved AI governance policies with clear accountability for outcomes.
The central bank wants lenders to build the ability to explain AI-driven decisions that materially affect customers, particularly in lending and fraud.
Banks should also red-team and stress-test AI systems before deployment and periodically afterwards.
Human oversight must remain in place wherever an AI error could cause material harm to customers or financial stability.
RBI to follow a principles-based approach
Malhotra said the RBI will take a principles-based and proportionate approach to regulating AI rather than impose rigid rules.
The risks faced by a large bank using proprietary AI models may differ significantly from those faced by a smaller lender using an off-the-shelf product from a technology vendor.
The RBI also plans to work with banks as AI technology evolves instead of regulating it from a distance.
The central bank will continue to use its regulatory sandbox to provide a safe environment for testing innovative use cases.
It also plans to facilitate common utilities such as MuleHunter and the proposed Digital Payments Intelligence Platform to strengthen fraud detection.
RBI remains focused on financial stability and credit growth
Malhotra also outlined progress on the RBI’s broader regulatory priorities announced at FIBAC 2025.
The central bank has taken measures covering credit risk capital, the expected credit loss framework, project finance, related-party transactions, dividend policy and net open positions.
The RBI remains on track to implement applicable Basel III guidelines from April 1, 2027, on a calibrated glide path.
On ease of doing business, the central bank has reduced regulatory burdens on boards, consolidated regulatory instructions, streamlined forms of business and rationalised current-account and working-capital norms.
It has also continued to strengthen digital public infrastructure such as Account Aggregator and ULI to lower the cost of credit origination and underwriting, particularly for MSMEs and underserved borrowers.
Banks that understand AI will have an edge
Malhotra said India has significant opportunities to use AI to improve credit access, customer service, financial inclusion and fraud detection.
But the RBI does not want banks to pursue AI adoption simply for the sake of adopting the technology.
“The banks that will win in the AI era will not necessarily be the ones that adopt the most AI, or the fastest,” Malhotra said.
Instead, he said banks that understand what they are deploying, maintain clear accountability and protect customer trust will be better placed to succeed in the AI era.
aniltikotekar4sme.com 