RBI’s fraud detection tool MuleHunter AI expands reach with over 15 more banks nearing rollout

The banks that have implemented the MuleHunter AI include Canara Bank, Punjab National Bank, Bank of India, Bank of Baroda, and AU Small Finance Bank.

At least 15 banks are in the process of implementing the MuleHunter AI – a tool to detect mule bank accounts - in next two months, in order to improve fraud detection, Reserve Bank of India's chief general manger (CGM) Suvendu Pati said on August 1.

The banks that have implemented the MuleHunter AI include Canara Bank, Punjab National Bank, Bank of India, Bank of Baroda, and AU Small Finance Bank.

These five banks have already implemented the AI tool, Pati added, while Federal Bank is in the advance stages of implementation and will soon go live on its platform.

"Federal Bank, it will go live in some time. But all other banks are in the preparation stage," Pati said in a speech at Future Proof Forensics 2025 event of the Institute of Chartered Accountants of India in Mumbai.

RBI’s CGM added that with the AL/ML tool, Canara Bank has achieved an accuracy level of 95 percent in detecting mule accounts.

Last year, Reserve Bank of India Innovation Hub (RBIH), a subsidiary of RBI, has developed an artificial intelligence and machine learning (AI/ML) model called MuleHunter AI, to help banks and financial institutions detect mule accounts amid rising concern around fraud.

A mule account is an account created by one person but operated by another person, and these are often used for money laundering and tax evasion.

Mule accounts are in violation of several rules and can be prosecuted under the Prevention of Money Laundering Act (PMLA). Such arrangements are illegal under tax laws, and even Sebi and Reserve Bank of India (RBI) rules prohibit the use of such accounts.

In April, RBI Deputy Governor Swaminathan J had said the central bank will ensure fraud risk management and proactive risk management system, which are operational at the bank level, are improvised so that frauds can be prevented.