As India’s banking system grapples with a surge in digital fraud and mule accounts, senior regulators and government officials are increasingly framing artificial intelligence not as a cure-all, but as a tool whose effectiveness will hinge on data quality, system-wide integration, and institutional preparedness.
Artificial Intelligence as an Early-Warning Tool
At recent industry discussions, senior policymakers have pointed to artificial intelligence as a means to move financial oversight from reaction to anticipation. By analysing global and domestic macroeconomic conditions, AI models, they argue, could help predict stress events before they cascade into broader financial disruption. Such predictive capacity, officials said, could play a role in preventing incidents that might otherwise prove catastrophic for the financial system.
The emphasis, however, has been on potential rather than promise. Speakers stressed that AI’s value lies in its ability to synthesise vast and disparate data sets—something traditional systems, built to respond after an event has already occurred, are not designed to do.
Certified Cyber Crime Investigator Course Launched by Centre for Police Technology
Data Quality and the Limits of Automation
That caution was underscored by T Rabi Sankar, Deputy Governor of the Reserve Bank of India, who highlighted that the outcomes of any AI-driven fraud detection system would only be as reliable as the data it consumes. Banks, he said, often rely on narrow data pools—such as payments data or institution-specific records—limiting the ability of automated systems to detect complex fraud patterns.
According to him, systems that draw on system-wide data rather than siloed information would, by definition, be better equipped to identify fraud before it occurs or to manage it more effectively after the fact. The distinction, he suggested, was not between human judgment and machines, but between fragmented and integrated information.
Platforms, Pilots, and Institutional Coordination
To that end, regulators have urged banks to actively participate in upcoming pilots involving shared intelligence platforms, including MuleHunter and digital payments intelligence systems. The message has been one of collaboration: banks, regulators, and technology teams working in concert to ensure that such platforms are not only tested, but deployed at scale and with speed.
Officials have encouraged continuous engagement with the Reserve Bank’s Innovation Hub and its FinTech Department, framing these initiatives as collective infrastructure rather than optional tools. The success of these efforts, they indicated, would depend less on technology alone and more on institutional willingness to share data and align processes.
Rising Fraud, Governance Gaps, and the Role of AI
Despite significant investments in technology, concerns about the scale of digital fraud remain acute. M Nagaraju, Secretary in the Department of Financial Services at the Ministry of Finance, described the situation as “not very good,” citing the continued rise of digital frauds and mule accounts within the financial system.
Traditional mechanisms, he said, have been put in place, but they have not stemmed the problem. What is required, in his view, is deeper deliberation on feasible solutions—an area where artificial intelligence, including generative AI, could play a vital role. Speaking at an event organised by the Indian Banks Association in Mumbai, he pointed to recent official data showing that in 2024–25, private sector banks accounted for 59.3 percent of the total number of fraud cases reported, while public sector banks accounted for 70.7 percent of the total amount involved.
