New Delhi: Behind the glowing screens of a thousand “digital arrests” lies a hidden plumbing system to channelise the stolen money. For years, cybercriminals have laundered their spoils through a labyrinth of “mule accounts” — temporary financial hideouts that vanish as quickly as they appear. Enter MuleHunter.AI, the Reserve Bank Innovation Hub’s latest predator to clean the digital jungle.Recently highlighted by Union home minister Amit Shah as a key shield against the rising tide of cybercrimes, MuleHunter.AI doesn’t just watch the money, it also learns the “heartbeat” of a scam — by the time a fraudster attempts to move the illicit gains, it has closed the gates, turning the “golden hour” into a dead end for the digital underworld.The tool, currently being utilised by around two dozen banks to identify and purge fraudulent accounts, is proving to be a game changer. “Unlike traditional banking audits that often flag suspicious activity days or weeks after a crime has occurred, this tool is built for real-time detection. This allows banks to freeze suspicious cash transfers as they are happening,” said a cyber cell officer.The tool represents a shift from rigid, rule-based filters to sophisticated machine learning. While old systems may only flag a transaction if it exceeds a certain limit, MuleHunter.AI analyses 19 subtle behavioural signatures identified through collaboration across the banking sector.One primary focus is ‘velocity anomalies’, where funds are transferred almost immediately after a deposit across a web of unrelated accounts. It also looks for ‘behavioural mismatches’, such as when a dormant student or pensioner account suddenly exhibits high-frequency, high-value activity that does not align with the owner’s profile.“The tool’s ability to perform digital fingerprinting makes it a game-changer. It can detect when a single IP address or mobile device is managing a cluster of seemingly unrelated accounts across different regions,” the officer explained. “It even monitors for robotic or unnatural navigation patterns within banking apps, which often suggests that a fraudster is controlling the account remotely through a trojan or screen-sharing software.” Cybercrime experts noted that the rise of digital arrest scams — where victims are coerced into “virtual custody” by fraudsters posing as law enforcement officials — relies entirely on the availability of mule accounts. “The success of such scams depends on the speed of the payout. By the time a victim realises he has been duped, the money has usually been layered through five or six different bank accounts. MuleHunter.AI breaks this chain by identifying the mule account before it can even be operationalised,” said an investigator.By providing it as a shared public infrastructure, RBI ensures that even smaller banks without enough resources can defend their customers. As the tool continues to detect nearly 20,000 mule accounts every month, officials hope it will finally turn the tide against the “cyber slavery” and extortion rackets.
