Skip to content
New DelhiFree OSINT tools and field notes for investigators in India · Newsletter · Contact
IGRS logoIGRSIntel Grid Research Station
India Cyber Brief, 22 November 2026: Delhi Police arrest 22 over fake banking apps

India Cyber Brief, 22 November 2026: Delhi Police arrest 22 over fake banking apps

Nine stories from the last 36 hours: a 22-arrest APK fraud bust by Delhi Police, a CERT-In warning on payout fraud at payment providers, email spoofing and mule-account arrests, and new WhatsApp, Signal and Android malware developments.

Delhi Police arrest 22 over fake banking apps linked to 22,139 victims

Delhi Police's West District ran a five-day operation between 11 and 15 November across Jharkhand, Uttar Pradesh, Bihar and West Bengal, with seven teams of 64 personnel raiding sites in Jamtara, Deoghar, Giridih and nearby areas. Twenty-two people were arrested, including people said to have built the malicious apps. Police link the groups to 22,139 victims, more than Rs 20 crore in proceeds and 2,080 complaints on the national cybercrime portal. DCP (West) Hareshwar Swami said the accused posed as bank staff and sent senior citizens APK files dressed up as KYC, life-certificate or credit-limit updates, which then read SMS and OTPs. Police say they examined 32 APK files and 56 Telegram profiles, and that APK-fraud complaints in Delhi fell 48% in the week after the raids.

Why it matters: Life-certificate season makes pensioners an easy target, and any "bank" link that installs an app should be treated as fraud.

Source: ThePrint (PTI), Millennium Post

CERT-In reports unauthorised transactions at firms handling government-linked payments

CERT-In said in an advisory posted on Friday, 20 November, that it had received reports of unauthorised transactions at service providers that support government-linked payments, Reuters reported. According to the advisory, payment fraud is no longer limited to banks: attackers are going after vendors and partner payout platforms. The method described is to change business records or approval relationships first and then let normal payment processes move the money, so each transfer looks legitimate. The advisory did not say how much was paid out or who was responsible.

Why it matters: This follows CERT-In's early-November API warnings, and finance teams should now check who can edit beneficiary and approver records, not only who can release payments.

Source: Reuters via Yahoo News

Ahmedabad cyber cell arrests four in Mumbai over a Rs 6.55 lakh email spoofing fraud

Ahmedabad City Cyber Crime Branch arrested four men from Nalasopara and Virar in Palghar for diverting a business payment of Rs 6,55,111. Police say the gang got into an email thread between Global Pantry Grocery Company and Maya Agrovet Limited, changed the owner's email ID and sent a forged bank letter naming a different account holder. One accused allegedly supplied seven bank accounts. The FIR, registered on 14 November, cites S.318(4) and S.319(2) BNS among other sections, along with S.66C and S.66D of the IT Act.

Why it matters: Business email compromise rarely needs malware; a phone call to a known number before changing bank details stops most of it.

Source: The Hans India (IANS)

Deoghar police arrest seven posing as payment-app support staff

A special team led by DSP Kailash Prasad Mahto raided the Sarath police station area of Deoghar district on 20 November and arrested seven men, seizing 11 phones and 11 SIM cards. Police say the group called victims pretending to be from Flipkart, Amazon Pay, Google Pay, PhonePe and Airtel Payments Bank, offering cashback or help with blocked cards. They also spread malicious APK files disguised as PM-Kisan, electricity-bill and RTO challan documents. Some numbers and IMEIs are linked to complaints in other states. The amount stolen has not been disclosed, and the companies named were not involved.

Why it matters: IMEI and number links to other states' complaints are exactly what investigators elsewhere can now query to close their own cases.

Source: The420.in

Shahdara cyber police arrest a 24-year-old mule-account supplier

Cyber Police Station Shahdara arrested a 24-year-old from Udham Singh Nagar, Uttarakhand, on 18 November as an alleged supplier of mule accounts. The case began with a Rs 2.22 lakh investment fraud reported by a local resident, registered as an e-FIR on 5 July. About Rs 80,000 of the money was traced to an account in a firm's name. Police recovered six phones, 15 debit cards, nine cheque books, four passbooks and nine SIM cards, and say two of the accounts appear in several complaints.

Why it matters: A case that took four months to reach the account supplier shows why victims should report within the first hour, while money can still be frozen.

Source: The420.in

WhatsApp patches three flaws, one suspected of use in a scam campaign

WhatsApp has fixed three vulnerabilities. CVE-2026-23871 (CVSS 3.5) is weak validation of external ad-reply URLs in WhatsApp and WhatsApp Business for Android before version 2.26.17.4, which could reveal a device's IP address; WhatsApp says it found the bug while disrupting a scam campaign and suspects it was used with spam. CVE-2026-102769 (CVSS 6.8) is an SQL injection in Android versions before 2.26.35.74 that can delete local files when a crafted "order details" message arrives from an unofficial client. A third flaw affects Meta AI's confidential-computing checks. WhatsApp reports no exploitation of the latter two.

Why it matters: India has more WhatsApp users than any other country, so updating the app is the simplest protection here.

Source: The Hacker News

AI agents found a zero-day in medical-imaging software and gained root

Amazon Threat Intelligence says suspected China-linked actors, assessed with moderate confidence, used AI agents run through the open-source Cairn framework to find a path-traversal flaw in MONAI Label, an open-source server used to label medical images. The bug, CVE-2026-133968 (CVSS 9.8), affects version 0.8.5 and earlier in the default no-login setup, and gave the attackers root on a university server. A fix is merged in the source code, but no patched release had been published. Advice is to block untrusted network access, turn on authentication and run the service under a limited account.

Why it matters: Indian hospitals and medical colleges running AI research tools often leave them open on the network, and this shows attackers can now find such bugs automatically.

Source: The Hacker News, AWS Security Blog

Bazuka: new Android banking malware hides theft behind a fake home screen

ThreatFabric has described Bazuka, an Android malware framework found in late October with a dropper, a core and 23 modules covering overlays, keylogging, SMS and two-factor code interception, remote control and crypto-address swapping. Its new trick is to display a screenshot of the phone's home or lock screen as an overlay, so the victim sees an idle phone while the attacker works underneath. The operators' panel uses a language model to score and pick victims. Droppers pose as a "TikTok premium" app and abuse accessibility services and wireless debugging.

Why it matters: A phone that looks idle but feels slow or warm after installing an app from outside the Play Store needs checking before any banking is done on it.

Source: The Hacker News

Signal blocks screenshots by default to keep AI agents out of chats

Signal now blocks screenshots by default on Android and iOS so that operating-system AI agents cannot read conversations from the screen. The setting can be changed under Settings, Privacy, App Security. Signal says the fix is incomplete because it cannot stop file-system access, and macOS is not covered. The change follows a Reuters report that Meta's Muse agent read Signal messages on macOS through screenshots and on Android through notifications.

Why it matters: Investigators and sources who rely on encrypted chat should check which AI assistants on their devices have screen or notification access.

Source: The Hacker News

Analysis: Three of today's Indian cases rely on fake APK files, and two rely on changed bank or approver details. Both are stopped by the same habit: confirm any payment change or app request on a number you already know.

Where to report

Report cyber fraud at once on the 1930 helpline or at cybercrime.gov.in. Report suspicious calls and messages through Sanchar Saathi Chakshu, and organisations should report incidents to CERT-In within six hours. Cheating through these schemes falls under S.318 BNS, personation under S.319 BNS, and personation using a computer under S.66D IT Act.

Sources

Frequently Asked Questions

How did the APK fraud in the Delhi Police case work?
Police say callers posed as bank staff and sent senior citizens app files presented as KYC, life-certificate or credit-limit updates. Once installed, the apps read SMS messages and OTPs, which let the gang take money from bank accounts.
What did CERT-In warn payment providers about on 20 November 2026?
According to Reuters, CERT-In reported unauthorised transactions at firms supporting government-linked payments, where attackers changed business records or approval relationships and then used normal payment processes to move money.
What should I do if I installed a suspicious APK sent by a caller?
Switch on flight mode, call your bank to block cards and UPI, report on 1930 or cybercrime.gov.in, and uninstall the app or reset the phone before using banking apps again.
Which law applies to email spoofing payment fraud in India?
Police commonly cite S.318 BNS for cheating and S.319 BNS for cheating by personation, with S.66C (identity theft) and S.66D (personation using a computer) of the IT Act.