Incident Response

Chinese Espionage Crews Circle SentinelOne in Year-Long Reconnaissance Campaign

Anti-malware vendor said it spent the past twelve months deflecting a stream of network reconnaissance probes from China-nexus threat actors

China

Anti-malware vendor SentinelOne said its security teams spent the past twelve months deflecting a steady stream of cyberespionage reconnaissance probes from China-nexus threat actors now seen targeting cybersecurity vendors.

SentinelOne said the attackers never gained a foothold inside its network but there were supply chain scares when a third-party contractor that handles laptop logistics for employees was briefly compromised.

In a technical deep-dive, the company’s SentinelLabs unit documented how the same infrastructure hammered more than 70 organisations between July 2024 and March 2025, including a South-Asian government IT agency and a major European media group, before turning reconnaissance scans on SentinelOne’s own internet-facing servers. 

The company said campaigns relied on well-known Chinese espionage staples. ShadowPad, a modular backdoor previously tied to APT41, showed up following dozens of breached gateways from Check Point, Fortinet and SonicWall gear, often delivered through recently disclosed exploits. 

ShadowPad use was also connected to an APT cluster the company calls PurpleHaze. That group was seen using a Go-based implant that tunnels traffic over SSH-in-WebSockets, and hid command-and-control servers behind what SentinelOne calls an “operational relay box” network, a rotating fleet of VPS nodes registered in bulk and managed from China. 

The SentinelOne researchers found overlaps with infrastructure and tactics long associated with APT15 and UNC5174, including Ivanti zero-days that were still under embargo when the hackers began chaining them.

Advertisement. Scroll to continue reading.

“We assess with high confidence that the threat actor’s activities were limited to mapping and evaluating the availability of select Internet-facing servers, likely in preparation for potential future actions,” the company said.

SentinelOne warns that the ongoing activity underscores a blind spot in the industry’s threat model: cybersecurity vendors themselves are increasingly high-value targets because compromising them can yield visibility into thousands of downstream customers. 

“Cybersecurity companies are high-value targets for threat actors due to their protective roles, deep visibility into client environments, and ability to disrupt adversary operations,” the researchers noted.

The company’s research team argues that disclosing its own near-misses, complete with file hashes, domains and IP addresses, is meant to remove the stigma of reporting attacks and make it harder for nation-state actors to reuse the same playbook. 

The latest disclosure follows a separate wave of North Korean fake-worker schemes and opportunistic ransomware scans targeting the prominent EDR vendor. SentinelOne said its HR teams fielded roughly 1,000 job applications from 360 fake personas tied to North-Korean revenue-generation schemes.

The company said none of the applicants were hired but its researchers milked the data for intelligence on resume-forging and deep-fake interview techniques. 

SentinelOne said profit-driven ransomware gangs have also tried to obtain console or agent access by buying stolen credentials or bribing insiders.  

Related: SentinelOne Targeted by North Korean IT Workers, Ransomware Groups 

Related: Justice Department Disrupts North Korean ‘Laptop Farm’ Operation

Related: Mandiant Offers Clues to Spotting and Stopping North Korean Fake IT Workers

Related: Chinese APT Tools Found in Ransomware Schemes, Blurring Attribution Lines 

Related Content

Artificial Intelligence

Xint says a flaw in OpenAI’s compaction process let researchers move reasoning traces between models, potentially making it cheaper to distill a rival AI.

Artificial Intelligence

The state-sponsored hackers targeted hundreds of organizations and gained privileged access to some environments.

Artificial Intelligence

The DHS said agencies should remove DeepSeek products as well as third-party products and services developed using DeepSeek from their IT systems.

Artificial Intelligence

Tasked with writing an application for a US defense contractor, the model made the code 19% more likely to be hackable.

Artificial Intelligence

At least three threat actors have been abusing Cairn for persistent access and financial theft.

Vulnerabilities

Armory Castle exploited CVE-2026-102764 for initial access, deployed custom tools, and deleted logs and artifacts.

Cybercrime

Backed by a federal court order, Microsoft and Health-ISAC seized domains powering YAM Store, CoGUI, and Iron Man.

Government

Flax Typhoon and other APTs used MicroScan and FishHub to scan and hack US and foreign critical infrastructure.

Copyright © 2026 SecurityWeek ®, a Wired Business Media Publication. All Rights Reserved.

Exit mobile version