HomeAnalysisNBFC Cyber Fraud Exposes Delhi-NCR’s Weakest Digital Link

NBFC Cyber Fraud Exposes Delhi-NCR’s Weakest Digital Link

The reported cyber fraud against non-banking financial companies (NBFCs) in Delhi-NCR is not simply a story of stolen money. It exposes how the digitisation of lending has shifted the attack surface from individual borrowers to the software, cloud infrastructure and disbursement controls used by financial institutions. According to a report citing Delhi Police sources, hackers infiltrated the systems of 12 finance companies over three months and diverted more than Rs 65 crore. The police have arrested 20 people in connection with the cases and frozen Rs 18 crore before it could be withdrawn as cash or converted into cryptocurrency.

The reported pattern matters because the criminals did not primarily rely on the familiar methods used against individual citizens, such as fake calls or digital-arrest scams. Instead, they targeted the systems that process approved loans. The cases described by the police indicate that the attackers sought to enter the software of smaller finance companies, identify loans that had been approved but not yet disbursed, and alter the destination account before the money was released.

That sequence turns a routine back-office function into a point of urban financial vulnerability. A loan approval is usually treated as an internal transaction between a lender and a verified customer. In the cases described, the approval remained in the system, but the final transfer was redirected to an account controlled by the fraud network. The reported losses therefore arose not from a customer being persuaded to disclose an OTP, but from the manipulation of the institutional system that was supposed to execute the payment.

The Delhi-NCR cases were described as part of a wider cybercrime network. The report says that more than 150 complaints involving NBFCs were registered across the country during the same period, with losses exceeding Rs 300 crore. These figures place the local cases within a broader pattern, although the supplied report does not provide a national breakdown by company size, state, lending product or recovery status.

The alleged attack route also shows why conventional perimeter security may not be sufficient for cloud-based financial operations. According to the police account cited in the report, several smaller finance companies use cloud platforms to run their websites and applications. Their firewalls reportedly whitelist the IP addresses of those cloud servers because the platforms are treated as trusted infrastructure. The attackers allegedly created accounts on the same cloud services and used trusted IP addresses to send fraudulent loan-disbursement commands.

In this model, the weakness is not necessarily that the firewall has been removed. The reported problem is that a legitimate-looking technical route may have been granted more trust than the transaction itself deserved. If an instruction appears to originate from a permitted cloud environment, the system may accept it even when the account details, timing or volume of disbursement differ from normal activity. The supplied report does not establish whether the affected companies lacked specific controls, but it does identify the gap between network authentication and transaction-level verification.

The investigation also points to the importance of the period between loan approval and payment. The malware allegedly remained hidden in company systems for seven to eight days while collecting information about approved loans that had not yet reached customers’ bank accounts. That waiting period would allow the attackers to understand the lender’s workflow and identify transactions close to disbursement.

The reported use of malware that observes a process before acting is significant for the way smaller lenders monitor their systems. A visible disruption may trigger an immediate response. A concealed programme that studies approved loans and intervenes only at the payment stage can remain undetected if monitoring focuses mainly on service outages, failed log-ins or unusual network activity. The report does not say how the malware entered each system or whether the companies used identical software, so the precise technical route remains case-specific.

One case reported from October involved an alleged loss of Rs 3 crore from a single company. Police said the investigation found the use of a Chinese artificial-intelligence platform called Cyber Strike AI, which they described as capable of sending hundreds of command scripts rapidly and simultaneously. The report does not establish that artificial intelligence independently carried out the entire operation. Rather, the police account links the platform to the speed and scale at which commands were allegedly deployed.

That distinction is important. The reported cases suggest that AI-enabled or automated tools may amplify an existing intrusion, but the underlying vulnerability still lies in the lender’s application and payment controls. The relevant institutional question is therefore not only whether a company uses an AI security product. It is whether its loan-disbursement system can distinguish an authorised request from a technically valid but fraudulent command.

The cases also expose a monitoring problem inside the lending cycle. Delhi Police said many smaller finance companies reconcile their accounts and loans weekly or fortnightly. If that description is accurate, a fraudulent disbursement may not be identified for several days. By then, the money may have moved through multiple accounts, making recovery more difficult. The reported freezing of Rs 18 crore shows the value of rapid intervention, but it also indicates that a substantial portion of the alleged Rs 65 crore loss had already moved beyond the first layer of detection.

This is where the urban dimension becomes visible. Delhi-NCR’s financial system is not confined to bank branches or physical offices. It includes distributed software teams, cloud-hosted applications, payment rails, customer databases and accounts spread across the region and beyond. A small finance company serving borrowers in the metropolitan area can depend on infrastructure operated elsewhere, while the proceeds of a fraud can move through accounts and cryptocurrency channels outside the city. The administrative challenge is consequently shared by police investigators, lenders, technology providers and financial institutions.

The report describes the Delhi Police cyber unit, IFSO, as the investigating agency in the 12 NCR cases. Its action included arrests and the freezing of funds before cash withdrawal or cryptocurrency conversion. The alleged involvement of 20 people in creating and supplying mule accounts highlights another layer of the system: the technical intrusion and the movement of stolen money may be handled by different participants. That separation can make investigations more complex because the person who alters a loan command may not control the account that receives the funds.

The mule-account network also demonstrates why cyber fraud is not only a software problem. The digital attack becomes financially effective when there is a mechanism to receive, layer and move the proceeds. The supplied report does not identify the banks, payment intermediaries or account holders involved, nor does it describe the legal action against each alleged participant. It does, however, indicate that account infrastructure was central to the operation and that investigators were tracking the funds alongside the digital intrusion.

The available evidence does not provide a full policy framework for NBFC cybersecurity. It does not specify whether the affected companies had reported prior warnings, whether they underwent security audits, or what standards applied to their cloud configurations and loan-management software. It also does not establish whether the cases resulted from a common vendor, a shared vulnerability or separate weaknesses exploited by one network. Those gaps prevent a broader claim that all smaller NBFCs face the same exposure.

What the cases do establish is a policy and governance problem around responsibility. A finance company may control the loan application and disbursement process, while a cloud platform hosts the relevant systems and a banking channel transfers the money. When a trusted cloud address is allegedly used to issue a fraudulent command, accountability cannot be understood only as a question of employee vigilance. It involves how authentication, application permissions, payment approval and reconciliation are divided between institutions.

The reported timeline further complicates oversight. A weekly or fortnightly reconciliation cycle may be compatible with ordinary accounting practices, but it creates a window in which an unauthorised disbursement can be concealed. The source does not say whether regulators or lenders are considering changes to reconciliation intervals, transaction alerts or independent approval requirements. Those measures therefore cannot be presented as established responses. The immediate documented response in the supplied material is police investigation, arrests and fund freezing.

The numbers give the clearest indication of scale. In Delhi-NCR, 12 companies reportedly lost more than Rs 65 crore in three months, an average of more than Rs 5 crore per reported company if the losses were evenly distributed. That average is only a calculation and does not describe the actual distribution. Nationally, more than 150 complaints reportedly involved losses of over Rs 300 crore during the same period. The October case involving Rs 3 crore shows that a single breach can produce a substantial loss before a periodic reconciliation identifies it.

These figures should still be read as reported complaint and investigation figures, not as a complete measure of the problem. The supplied article does not clarify how many incidents went unreported, how much money was ultimately recovered, or whether all complaints involved the same technical method. The distinction matters because police investigations may reveal a pattern across cases while individual companies may have experienced different forms of intrusion.

The larger urban question is how quickly digital lending infrastructure is expanding compared with the internal controls that protect it. The report’s account suggests that the attackers understood the operational sequence of loan approval, system access and payment. They allegedly exploited the trust built into cloud infrastructure and the delay built into financial reconciliation. That combination links technology design to institutional routine: a weakness becomes valuable when it can be used repeatedly and remain unnoticed long enough for funds to be moved.

For Delhi-NCR, the immediate significance is not limited to the companies named in the report. The cases show that urban financial resilience now depends on the security of systems that residents may never see. Borrowers may interact with an app or website, but the safety of the transaction depends on backend permissions, cloud access, disbursement commands and the speed with which irregularities are detected. The evidence supplied so far confirms the scale of the alleged losses and the police response. It leaves open the questions of how the systems were initially breached, whether common vulnerabilities connected the cases, and what controls affected lenders will change as the investigation proceeds.


RELATED ARTICLES

Most Popular

Latest News