At least 15 more banks are expected to adopt the RBI Innovation Hub’s AI-powered MuleHunter platform within the next two months to strengthen efforts in identifying fraudulent mule accounts, a senior RBI official said. Currently, five banks—Canara Bank, Punjab National Bank, Bank of India, Bank of Baroda, and AU Small Finance Bank—have implemented it, with Federal Bank set to follow shortly.
Mule accounts, typically created using stolen identities, are used to transfer illicit funds rapidly through multiple layers, making fund recovery difficult. MuleHunter, which uses machine learning, is being offered to banks at no cost and complements existing anti-fraud systems. It boasts 90% accuracy in identifying fraudulent accounts, compared to 80% false positives from other systems, making it easier for staff to act swiftly.
So far, 90 fraud patterns have been detected, with most activity occurring between 11 PM and 1 AM. While fraudsters may exploit the system during its early learning phase, its effectiveness is expected to improve as more banks join and the system evolves.
