Possibly using AI, hackers diverted approved loans at 12 non-banking financial companies (NBFCs) in Delhi-NCR over three months, taking over Rs 65 crore, The Indian Express reported. The report cited senior Delhi Police officers, who said these cases are part of over 150 first information reports (FIRs) filed by NBFCs across India.
Instead of the popular fraud mechanism “digital arrests”, AI fraudsters reportedly use to hack into financial systems, like NBFCs, hackers had found loans that were approved but not yet disbursed and then they diverted the amounts to mule accounts, the Delhi Police told Indian Express.
They also said that hackers are typically using malwares to breach the front end of an NBFC’s systems since many of the small firms that hackers target have flaws in their software code. The malware could stay in their systems for over 7-8 days, as per the report.
“We have detected traces of AI being used in these frauds. These incidents are largely attributable to weaknesses in the public-facing cyber systems of NBFCs.” – Gaurav Tyagi, DCP at Intelligence Fusion and Strategic Operations
Earlier in November, Reserve Bank of India met Banks regarding the rise of AI-assisted cyber-attacks on India’s financial sector and gave the banks a “dressing down”. The RBI asked the banks to secure their systems in a confidential advisory. Head and Tales’s Arti Singh who reported on the confidential advisory, the RBI has issued another confidential advisory concerning the API hijacking on November 16.
The opaque system of disclosures:
Similarly, in another instance, in October, Economic Times reported, AI-powered cyberattacks have reportedly siphoned off about Rs. 300 crore from smaller non-banking financial companies (NBFCs) across India over two months, resulting in more than 150 FIRs being filed. MediaNama argued at that time, that relevant regulatory authorities should’ve disclosed these cyberattacks instead of the public knowing these from the media.
However, this has been a wider pattern. For instance, the Central Bank of India’s Rs. 13.97 crore loss became public through a Hindustan Times report on its police complaint before it was disclosed to the National Stock Exchange regarding the vulnerability in their financial networks without disclosing the alleged lost money.
Similarly, SBI told stock exchanges about a data exposure two days after a dark web claim surfaced. CoinDCX’s chief executive called breach reports a false alarm, then admitted a Rs. 152 crore loss three days later.
Additional CP at Delhi police’s IFSO, Sanjay Bhatia, asked the financial entities to follow the CERT-In issued advisories regarding the rise in AI-driven cyberattacks on financial systems.
CERT-In’s recent AI-powered cyberattacks:
October 9: CERT-In and CSIRT-Fin, its financial-sector arm, issued a “Critical” advisory to NBFCs, digital lenders, fintechs, payment service providers and wallet operators. The advisory described how attackers steal payment gateway and partner-bank application programming interface (API) credentials and then directly access disbursal APIs. This allows them to bypass KYC checks and per-transaction limits.
However, the advisory does not name any affected entity or identify the specific threat actors or their modus operandi. It also does not disclose how many incidents CERT-In has recorded or the amount of money stolen, according to its records.
October 15: A second advisory warned about AI agents accessing Indian systems without authorisation. It stated that attackers with financial motives had conducted cyber campaigns against Indian entities using publicly available, AI-native offensive security frameworks.
However, the advisory did not specifically identify the tools used or name any affected entities.
October 27: CERT-In conducted a full-day drill involving more than 300 officials from over 55 financial institutions, predominantly regional rural banks and cooperative banks. Its public post does not mention NBFCs or any financial losses.
The modus operandi:
According to the Indian Express, an NBFC usually hires a fintech company to host its website on a data centre run by a cloud platform such as Amazon Web Services, a senior police officer told the paper. The firewall then adds the data centre’s outbound IP addresses to a whitelist. A disbursal request is approved only if it comes from one of those addresses.
Anyone can open an account on these cloud services, the officer said. Fraudsters can therefore send commands from addresses the NBFC’s firewall already trusts. The command asks the NBFC’s application programming interface (API), the software link through which its systems instruct the bank, to release a loan. It carries a mule account in place of the borrower’s account, and the bank approves it instantly. In the police account, the firewall’s trust in those addresses lets the command pass as routine.
A senior officer of Delhi Police told Indian Express that they have arrested 20 people in the 12 NCR cases. All are linked to creating or procuring mule accounts, which are bank accounts that receive and pass on stolen money. Rs. 18 crore of the Rs. 65 crore was seized before it left those accounts as cash or crypto.
Indian Express named only two companies: Hero FinCorp that reportedly lost Rs. 12 crore on August 17 and Unifinz Pvt Ltd which claimed to have lost Rs. 12.49 crore in 550 transactions over two days in its complaint dated October 21. The other entities remain unnamed. These transactions range from Rs. 66 lakh in 66 transactions to Rs. 10.59 crore in 1,131 transactions, that happened between September 18 and November 3.
Also read
- Rs. 300 Cr lost in AI cyberattacks on NBFCs: Why did the public learn about it from the media, not regulators?
- Hackers bypassing KYC via API hijacks, CERT-In warns fintechs
- AI agents accessed Indian systems without authorisation: CERT-In
The post AI-assisted hackers breach NBFCs in Delhi-NCR; divert Rs 65 cr in approved loans appeared first on MEDIANAMA.
