The Reserve Bank of India (RBI) has been ramping up its efforts to curb financial fraud in the country. A significant move in this direction is the introduction of the MuleHunter platform, an artificial intelligence (AI) and machine learning (ML) based fraud detection system developed by the RBI Innovation Hub.
Over the next two months, at least 15 more banks are expected to adopt MuleHunter to strengthen their ability to detect mule accounts, a major challenge in the fight against cyber and financial crimes.
What Are Mule Accounts?
Mule accounts are bank accounts created and operated by fraudsters to receive, transfer, or launder illicit funds obtained from unsuspecting victims.
These accounts are usually opened using fraudulently obtained credentials and serve as a conduit for moving stolen funds quickly through multiple layers of accounts. This layering makes it nearly impossible for banks to claw back the money to its rightful owner.
Typically, funds move through these mule accounts within minutes or hours, leaving a very narrow window for banks to detect and stop the transactions.
Current Implementation Status
So far, five major banks, Canara Bank, Punjab National Bank (PNB), Bank of India, Bank of Baroda, and AU Small Finance Bank, have already implemented MuleHunter.
In addition, Federal Bank is set to adopt the platform in the coming days.
According to Suvendu Pati, Chief General Manager at RBI, MuleHunter is currently operational in six banks and discussions are underway to onboard another 15 to 20 banks in the next 1.5 to 2 months.
How MuleHunter Works
MuleHunter uses AI-driven algorithms and machine learning models to identify unusual transaction patterns that may indicate mule account activity.
Unlike other fraud detection platforms, MuleHunter reportedly delivers 90% positive alerts, compared to around 80% false positives seen in many existing systems.
This high accuracy means banks can act more quickly and confidently on suspicious transactions, reducing wasted resources on false alarms.
Efficiency Gains and Learning Ability
One of MuleHunter’s strengths is that it continuously learns from transaction data. As more banks join the system, it gains access to a wider range of transaction patterns, making detection even more accurate.
Currently, the platform has already identified 90 distinct fraud patterns used by cybercriminals operating mule accounts.
Interesting Fraud Pattern Insights
The platform has uncovered a unique trend: the highest number of mule account transactions occur between 11 PM and 1 AM.
This timing is strategic for fraudsters because many bank call centres are closed during this period for maintenance or operational reasons, meaning victims cannot report or stop suspicious activity immediately.
Bank Requirements for Implementation
Although MuleHunter is being offered free of cost to banks at present, implementation requires:
- Infrastructure investment by the banks
- Skilled data scientists to monitor and analyse flagged transactions
- Integration with the bank’s existing fraud detection and transaction monitoring systems
The platform is designed for both savings accounts and current accounts, making it versatile for different banking needs.
Short-Term vs Long-Term Challenges
Pati acknowledged that while MuleHunter is still in its learning phase, fraudsters may continue to find loopholes in the short term.
However, as the system evolves and banks share more transaction data, law enforcement agencies and bank fraud detection teams will become far more effective in preventing mule account activity over the medium to long term.
Future of AI in Fraud Detection
The RBI has also set up a committee on Artificial Intelligence to explore broader applications in the banking sector. The committee’s report is expected to be released soon.
With quantum computing on the horizon, Pati warned that the banking industry must remain alert to emerging threats and prepare its defences in advance.