This report details two interconnected malware campaigns targeting Chinese-speaking users in 2025, using large-scale brand impersonation to deliver Gh0st RAT variants. The first campaign, active from February to March, mimicked three brands across over 2,000 domains. The second campaign, starting in May, impersonated over 40 applications with more sophisticated infection chains. Both campaigns used cloud infrastructure for payload delivery and DLL side-loading for evasion. The adversary demonstrated an evolving operational playbook, advancing from simple droppers to complex multi-stage infections. The campaigns' infrastructure remained active for months, indicating a persistent and well-resourced threat actor focused on Chinese-speaking targets globally.
The Chinese-affiliated threat actor UNC6384 is exploiting the ZDI-CAN-25373 Windows vulnerability to deploy PlugX malware targeting European diplomatic entities, specifically in Hungary and Belgium. The attack vector involves spearphishing emails with malicious LNK files themed around diplomatic conferences. The campaign uses DLL side-loading of legitimate Canon printer utilities to evade detection and maintain persistence. UNC6384’s operations have expanded from Southeast Asia to Europe, focusing on espionage related to foreign policy, defense, and economic matters. This campaign demonstrates advanced social engineering and rapid exploitation of new vulnerabilities. The malware provides persistent remote access for intelligence gathering. No known public exploits exist yet, but the threat is active and targeted. The medium severity rating reflects the targeted nature and complexity of the attack. European diplomatic organizations should prioritize mitigation to protect sensitive information and maintain operational security.
A sophisticated cyber campaign targeting Chinese individuals in the FinTech, cryptocurrency exchange, and trading platform sectors has been uncovered. The operation uses spear-phishing emails with malicious .LNK files embedded in fake resumes. When executed, these files initiate a multi-stage infection process, ultimately deploying ValleyRAT malware. The malware establishes persistence through scheduled tasks, performs system reconnaissance, and exfiltrates sensitive data. The campaign's infrastructure is primarily hosted in Hong Kong, with multiple domains using the .work TLD to impersonate job portals. The attackers employ various techniques to evade detection, including anti-VM checks and attempts to disable antivirus software.
LAB52 has identified a new backdoor called NotDoor, attributed to APT28, a Russian intelligence-linked threat group. NotDoor is a VBA macro for Outlook that monitors incoming emails for specific trigger words, enabling data exfiltration, file uploads, and command execution on victim computers. The backdoor is deployed via Microsoft OneDrive.exe using DLL side-loading, and it establishes persistence by modifying registry keys. NotDoor employs obfuscation techniques and a custom string encoding method. It can execute commands, exfiltrate files, and upload files to the victim's machine. The malware demonstrates APT28's continuous evolution in bypassing defense mechanisms, posing a significant threat to NATO member countries across various sectors.
Chinese threat actors have launched a new phishing campaign using DeedRAT, a modular backdoor. The campaign exploits a DLL side-loading vulnerability in VIPRE Antivirus Premium's MambaSafeModeUI.exe. DeedRAT now includes a new NetAgent module, expanding its capabilities. The malware uses TCP for C2 communication and employs various persistence techniques. Notable features include a custom encryption algorithm using a linear congruential generator, API protection, and junk functions to confuse analysts. The backdoor's continued development and increased obfuscation suggest the threat actors are actively enhancing their tools and techniques.
NailaoLocker, a new ransomware variant targeting Windows systems, uses AES-256-CBC encryption and uniquely incorporates SM2 cryptography with hard-coded keys. It employs DLL side-loading for execution and uses I/O Completion Ports for multi-threaded file processing. The ransomware includes both encryption and decryption modes, with a built-in SM2 key pair. However, testing revealed the embedded private key fails to decrypt files properly, suggesting it may be a trap or an incomplete build. NailaoLocker's use of Chinese SM2 standards for key protection marks a departure from typical ransomware practices. While the decryption logic functions correctly with valid key material, the variant's true intent remains unclear.
