The BOSSAI campaign involves a single operator using a mesh of 12 AI agents to conduct persistent attacks across 11 countries. The attacker leveraged both abliterated and commercial AI models in specialized roles such as infrastructure building, vulnerability scanning, and exploitation. Primary targets included emergency services, universities, financial institutions, and government systems in Taiwan, Thailand, and Mexico. Confirmed breaches include exfiltration of medical records from Taipei emergency services, student data from Thai universities, and over 200,000 Mexican residential phone records. Despite attempts to jailbreak commercial AI models, the attacker relied on less effective abliterated models. The investigation uncovered the full operational infrastructure, including command histories and victim data, highlighting both the capabilities and limitations of AI-augmented autonomous exploitation.
An extortion campaign targeted Japanese real estate websites in October 2026, compromising one company and pivoting to nearly 20 related sites via shared hosting credentials. The attacker, known as kkkkevin777, stole over 16,000 customer records including plaintext passwords and sensitive personal data. Ransom pages with deadlines were deployed, threatening to expose stolen data. Ten days after the extortion wave, PHP2Shell exploitation tooling was found on an exposed staging server, indicating preparation for further attacks. Additional data from a Taiwanese company was also exfiltrated. The server contained weaponized PHP2Shell scripts, web shells, and tools for lateral movement, though no confirmed live exploitation via PHP2Shell was documented.
A Korean-speaking threat actor linked with moderate confidence to North Korea's Lazarus Group is conducting a global campaign exploiting two critical PHP vulnerabilities (CVE-2026-109618 and CVE-2026-109619) that allow unauthenticated remote code execution. The campaign targets government agencies, financial institutions, manufacturing, logistics, aviation, and trading platforms across multiple countries including the US, Japan, and EU. Exploitation attempts have been observed since 17 October 2026, involving web shell deployment and session poisoning. The vulnerabilities affect all internet-facing PHP installations below versions 8.5.12, 8.4.27, 8.3.36, or 8.2.35, with Apache mod_php being at highest risk.
In October 2026, a Chinese-speaking threat actor conducted a global AI-assisted hacking campaign using the ARTEX penetration testing framework combined with multiple AI coding assistants. The attacks targeted organizations across finance, retail, education, cryptocurrency, and telecommunications sectors in over 20 countries worldwide. The campaign utilized custom scripts to maintain context across AI sessions and divided tasks among specialized AI agents. The attackers sought bulk personal data, transaction records, and mobile phone numbers, with indications of intent to sell stolen information on underground markets. The workflow included peer review among AI models to improve effectiveness. The campaign is financially motivated and opportunistic in nature.
In October 2026, multiple organizations in Japan suffered unauthorized access incidents leading to personal information leakage. Attackers used artificial intelligence to scan for known vulnerabilities and exploit APIs, including server-side request forgery. They analyzed Android application packages and employed tools to reverse engineer Flutter apps and target various services such as web applications, MySQL databases, Jenkins servers, and email gateways. The attackers attempted to evade endpoint detection and response systems and deployed web shells to gain internal network access. The campaign also affected web services beyond Japan, indicating a wider scope of malicious activity.
The Hotejka affiliate program, a rebrand of ROI777, operates the Help Traffic Distribution System (TDS) to deliver tech support scams. It uses browser fingerprinting and geolocation to redirect victims through compromised websites and malvertising to fake Microsoft security alerts that lock browsers and display fraudulent phone numbers. Victims who call are guided to install remote desktop tools like UltraViewer, granting attackers ongoing system access. Since April 2026, Hotejka has abused Cloudflare Pages to create over 20,000 disposable domains, generating about 300 new domains daily. Affiliates such as Balada Injector, Detour Dog, and DollyWay drive traffic to this TDS. Infoblox developed predictive detection algorithms that identify malicious domains about seven days before deployment, significantly reducing exposure.
A coordinated disruption targeted Chinese-speaking cybercriminals operating three major phishing kits—CoGUI, CoAceV, and CoFish_Ws—that distributed hundreds of millions of credential phishing messages primarily against Japanese organizations. These phishing campaigns impersonated 66 well-known consumer and finance brands to steal usernames, passwords, and payment data. The threat actors used sophisticated evasion techniques, including approximately 17,000 unique rotating domains. The disruption involved civil lawsuits and domain seizures against ecosystem enablers YAM Store and Iron Man, conducted collaboratively by Microsoft, Health-ISAC, and security partners. This action is expected to significantly disrupt multiple threat groups involved in financial fraud within the Chinese phishing ecosystem.
Microsoft uncovered a Chinese-language cybercriminal ecosystem involving phishing kit developers, an illicit marketplace called YAM Store, and downstream threat actors. This ecosystem enables large-scale phishing campaigns targeting organizations worldwide, including in Japan, the US, Australia, and India. The phishing kits impersonate legitimate brands and facilitate credential theft. YAM Store sells fraudulently acquired Microsoft and Google cloud environments to support these operations. Microsoft and partners disrupted the primary developers and infrastructure behind this ecosystem.
A sophisticated phishing campaign targets YouTube content creators by impersonating well-known brands such as Hollyland, Nike, and Spotify. Attackers send personalized sponsorship emails referencing specific videos from the creators' channels, directing them to fake platforms that mimic legitimate campaign dashboards. The scam culminates in a Google sign-in prompt that either steals credentials directly or requests excessive permissions. Compromised accounts are then locked by changing recovery options and used to propagate further scams to followers. This modular campaign operates globally, with confirmed targeting in Peru, Japan, and English-speaking regions, using multiple domains with consistent design and functionality.
In July 2026, China-aligned threat actor TA419 conducted credential phishing campaigns impersonating prominent economists and AI policymakers, targeting AI experts at US think tanks, universities, and legal organizations. The group sent benign conversation starter emails themed around AI policy topics, such as invitations to join an AI Policy Advisory Committee. Upon receiving responses, TA419 deployed multi-stage URL redirection chains leading to Adversary-in-the-Middle credential phishing pages using a customized Frameless Browser-in-the-Browser tool. Active since April 2025, TA419 consistently targets individuals at US and Japan-based think tanks, defense contractors, universities, and law firms. This activity likely supports Chinese intelligence objectives to monitor US AI policy and regulatory developments amid strategic competition involving AI technologies, model distillation concerns, and export controls between the US and China.
Between July and August 2026, a phishing campaign targeted organizations primarily in Japan and other East and Southeast Asian countries. Attackers sent phishing emails in Japanese and Korean, impersonating product damage complaints to trick victims into visiting fake document-sharing sites. These sites delivered ZIP files containing malware executables with double extensions and DLLs using advanced loader techniques. The final malware payloads belonged to the PureRAT and PureLogs families. The campaign employed frequent changes in loader structures and multiple evasion methods to avoid detection. Infrastructure analysis revealed consistent patterns and identifiers across campaigns.
A suspected Chinese-speaking threat actor linked to Red Heron conducted extensive exploitation campaigns from May to September 2026. The attacker exploited multiple vulnerabilities in WordPress (CVE-2026-63030, CVE-2026-60137) to breach a western government organization, stealing over 18,000 sensitive records including credentials and personally identifiable information. Additionally, the actor exploited CVE-2026-7273 in ZyXEL GS1900 switches, compromising 996 devices across 48 countries and exfiltrating configurations and credentials. The adversary demonstrated sophisticated capabilities such as custom tool development possibly using large language models, privilege escalation, and systematic data exfiltration via PowerShell-based techniques.
Noodle RAT is a modular remote access trojan with distinct Windows and Linux versions, used since at least mid-2016 by Chinese-speaking threat actors. It has been involved in espionage and cybercrime campaigns targeting multiple Asia-Pacific countries. The malware features shared command-and-control infrastructure and modular capabilities, with the Windows variant operating as an in-memory backdoor and the Linux variant providing reverse shell and proxy functionalities. Multiple threat groups have deployed Noodle RAT, which appears to be actively maintained and possibly commercially available.
BambooToken is a multiplatform malware family active since at least February 2023 that uses the MQTT protocol for covert command and control. It targets Windows and Linux systems primarily across Asia and South America, infecting backend servers in various sectors including mobile apps, legal, financial, software, hotels, and GitLab instances. The malware employs sideloading via Tendyron's OnKey authentication software, common in Chinese banking and government networks. It has capabilities for host enumeration, antivirus detection, and potential keylogging and clipboard theft. The attacker uses MQTT's publish-subscribe model and Cloudflare proxies to conceal infrastructure and operations. The campaign demonstrates sophisticated operational security and widespread infections.
MacSync Stealer is a macOS information stealer and remote-access stager delivered through ClickFix social engineering and malvertising campaigns. Operating under a malware-as-a-service model, it employs sophisticated evasion techniques including process daemonization, single-byte XOR obfuscation, and in-memory AppleScript execution to bypass Apple Gatekeeper, XProtect, and EDR solutions. The malware exfiltrates credentials, browser data, cryptocurrency wallets, and SSH keys through fault-tolerant 10MB chunked uploads to command-and-control infrastructure. MacSync targets professionals in software engineering, cryptocurrency, fintech, and corporate environments across North America, Europe, and Asia-Pacific regions, with campaigns impersonating legitimate services like Google Meet, Claude AI, Docker, and TradingView to trick victims into executing malicious Terminal commands.
The Iranian-linked cyber-espionage group Tortoiseshell has expanded its malware toolkit with a new reverse SSH tunneling utility disguised as wtsapi32.dll and a C++ backdoor similar to TWOSTROKE malware. The SSH tunnel uses the Windows OpenSSH client to connect to command-and-control servers. The backdoor supports executing files, shell commands, in-memory DLL execution, and file manipulation. Infrastructure analysis shows domains linked to multiple countries including the UAE, Saudi Arabia, UK, Belgium, Canada, Australia, Japan, and the United States, indicating broader targeting beyond the Middle East and US defense and military sectors. The group has been active since 2018 and maintains persistent infrastructure despite domain suspensions. No known exploits in the wild or patches are reported.
Kimsuky conducted spear phishing campaigns targeting South Korean and Japanese entities in early 2026. The attacks used LNK malware distributed via OneDrive share links to establish scheduled tasks that periodically retrieved PowerShell scripts from command-and-control servers. These scripts profiled infected systems, exfiltrated email data from Thunderbird and Outlook, and logged keystrokes. The threat actor leveraged legitimate remote control tools such as Chrome Remote Desktop and AnyDesk to evade antivirus detection and maintain persistent access. Additionally, a malicious Chrome extension with AI-generated code was used to steal Gmail data. The operation employed rotating infrastructure and compromised legitimate Korean servers as command-and-control nodes to hinder tracking efforts.
Three financially motivated threat actors acquire expired malicious domains through dropcatch to inherit traffic from previously compromised websites. Stuffy Squirrel specializes in hiding activity within legitimate scripts and has operated since 2020, selling traffic to affiliate advertising networks. Shady Squirrel uses custom JavaScript and Keitaro injections with multi-step cloaking, partnering with initial access brokers to deliver tech support scams and SocGholish malware, notably facilitating SocGholish's return within weeks of Operation Endgame disruption. Swiping Squirrel, the most prolific actor, operates in greyhat territory by selling fraudulent traffic to zero-click advertising platforms like ZeroPark, often resulting in malvertising and malware distribution. These actors control thousands of domains collectively, exploiting lingering infections from previous compromises without conducting new attacks themselves.
BlueShell is an open-source remote access trojan developed in Go language, primarily used by Chinese-based threat actors. A variant of BlueShell has been identified in post-intrusion activities by APT groups including BlackTech, targeting organizations in Japan, South Korea, and Thailand. This variant differs from the original through a dedicated dropper mechanism, proxy server-based C2 communication, and anti-forensic capabilities. The dropper deploys the variant to /tmp/kthread, disguises it as a Linux kernel worker process, and removes filesystem traces. Recent variants observed since 2024 include XOR-encoded configuration data and proxy functionality, indicating continuous development. The malware performs hostname verification, validates C2 certificates, and implements commands for file transfer, remote shell, and SOCKS5 proxy capabilities.
The OpenAI models targeted services beyond Hugging Face as they attempted to solve the tasks they were given. The post JFrog Zero-Days Exploited in OpenAI-Hugging Face Hack appeared first on SecurityWeek .
