Analysis of over 400 AI-enabled malware samples shows that most remain confined to research and sandbox environments, with only a small fraction observed on protected endpoints across three countries. These samples span five malware families including FunkSec ransomware and Oyster backdoor. Existing behavioral detection, cloud sandboxing, and endpoint analytics successfully detect and block all observed samples. The AI component primarily accelerates malware development rather than enabling evasion of defenses. Distribution patterns are opportunistic rather than targeted.
A multinational law enforcement operation called Operation Endgame has successfully disrupted SocGholish, a malware framework operated by threat actor TA569 since 2017. The operation took down 106 servers and domains and remediated nearly 15,000 compromised WordPress websites. SocGholish uses fake browser update prompts on compromised websites to trick victims into downloading malicious JScript payloads, providing initial access to corporate networks for ransomware deployment and data breaches. Analysis revealed that 55% of Infoblox cloud customers were exposed to SocGholish in 2026, demonstrating widespread impact across multiple industries including government, education, and healthcare. The framework employs domain shadowing techniques and operates through a four-stage attack chain involving traffic acquisition, filtering, fake update lures, and on-device implant execution. SocGholish infrastructure has facilitated access for various ransomware families and has been extensively used by the notorious Evi...
Threat actors are exploiting AI agent skill formats as a novel attack vector, using convincingly packaged OpenClaw skills to distribute malicious payloads. The latest campaign employs pure social engineering, with skills containing no malicious code themselves but instead tricking users into downloading Windows binaries. The attack leverages a fake GitHub infrastructure hosting GachiLoader, which delivers Rhadamanthys infostealer through fileless injection. The operation uses two delivery mechanisms: Node.js Single Executable Applications and an Electron dropper, both converging on the same payload. GachiLoader employs sophisticated evasion techniques including anti-VM checks, sandbox detection, and privilege escalation, while using a Polygon blockchain smart contract as its C2 resolver for enhanced persistence and obfuscation.
Gen Threat Labs has identified Remus, a new 64-bit infostealer attributed to the Lumma Stealer family, emerging after Lumma's takedown and the doxxing of its alleged core members. First campaigns date back to February 2026, with the malware switching from Steam/Telegram dead drop resolvers to EtherHiding and employing new anti-analysis checks. Remus shares multiple characteristics with Lumma including identical string obfuscation techniques, AntiVM checks, direct syscall/sysenter handling, indirect control flow obfuscation, and a unique Application-Bound Encryption bypass. The analysis details test builds labeled Tenzor from September 2025, representing a transitional step between Lumma and Remus. While maintaining Lumma's stealing arsenal for browser passwords, cookies, and cryptocurrency, Remus introduces blockchain-based C2 resolution via EtherHiding, additional anti-sandbox checks targeting analysis tool DLLs, and enhanced device fingerprinting capabilities.
Iranian intelligence services are increasingly engaging with the cyber crime ecosystem, leveraging criminal tools, services, and operational models to support state objectives. This trend is particularly evident among actors linked to the Ministry of Intelligence and Security (MOIS), such as Void Manticore and MuddyWater. These actors are not merely imitating criminal behavior but actively associating with the cyber criminal ecosystem, using its infrastructure, malware, and affiliate-style relationships. This approach enhances their operational capabilities, complicates attribution, and contributes to confusion around Iranian threat activity. Examples include the use of ransomware branding, commercial infostealers, and overlaps with criminal malware clusters. This shift from imitation to active engagement with cyber crime offers both improved deniability and expanded technical capabilities for Iranian actors.
A new malware distribution campaign utilizing compromised YouTube accounts to spread infostealers has been identified. The campaign employs GachiLoader, a heavily obfuscated Node.js loader, to deploy the Rhadamanthys infostealer. GachiLoader implements anti-analysis techniques and uses a novel PE injection method called Vectored Overloading. To aid analysis, researchers developed an open-source Node.js tracer tool. The campaign has affected over 100 videos with 220,000 views across 39 compromised accounts since December 2024. The malware evades detection, elevates privileges, and disables Windows Defender before retrieving its payload.
Matanbuchus 3.0 is a sophisticated C++ malware downloader offered as Malware-as-a-Service since 2020, designed to deliver additional malicious payloads including ransomware and remote access trojans like Rhadamanthys and NetSupport RAT. It employs advanced obfuscation techniques such as junk code insertion, encrypted strings, and API hashing to evade detection. The malware features anti-analysis mechanisms including an expiration date and persistence via scheduled tasks. It communicates with its command and control servers using encrypted Protocol Buffers over HTTP(S), supporting a wide range of commands for payload execution, data collection, and system manipulation. While no known exploits are currently reported in the wild, its modular design and use in ransomware campaigns make it a medium-severity threat. European organizations are at risk due to the malware’s capability to facilitate ransomware attacks and backdoor access, potentially leading to data breaches and operational disruption. Mitigation requires targeted detection of its persistence mechanisms, network traffic analysis for encrypted C2 communications, and blocking associated domains and URLs. Countries with high technology adoption and ransomware targeting history, such as Germany, France, the UK, Italy, and the Netherlands, are most likely to be affected.
A multi-stage malware execution chain originating from a ClickFix lure has been discovered, leading to the delivery of infostealing malware like LummaC2 and Rhadamanthys. The campaign utilizes steganography to hide malicious code within PNG images. Two distinct ClickFix lures were observed: a standard 'Human Verification' and a convincing fake Windows Update screen. The execution chain involves mshta.exe, PowerShell, and .NET assemblies, ultimately extracting and injecting shellcode into target processes. The steganographic technique encodes malicious data directly into image pixel data, using specific color channels for payload reconstruction and decryption in memory. This sophisticated approach helps evade signature-based detection and complicates analysis.
The October 2025 Infostealer Trend Report highlights the ongoing and evolving threat posed by Infostealer malware families such as Rhadamanthys, ACRStealer, and LummaC2. Attackers have shifted distribution tactics, increasingly leveraging legitimate websites to evade detection and search engine restrictions, a technique known as SEO poisoning. A notable development is the mass distribution of a new Loader malware employing DLL sideloading to execute malicious payloads stealthily. The report also details changes in LummaC2 distribution patterns and the use of sophisticated disguise and phishing techniques to target companies. These Infostealers aim to exfiltrate sensitive information, posing risks to confidentiality and potentially enabling further attacks. The threat does not require known exploits or zero-day vulnerabilities but relies on social engineering and advanced evasion methods. European organizations are at risk due to the widespread use of affected malware and the targeting of companies via phishing. Mitigation requires proactive detection of DLL sideloading, monitoring for SEO poisoning campaigns, and enhanced phishing defenses.
Check Point Research identified a sophisticated malware distribution campaign on YouTube called the YouTube Ghost Network. This network uses over 3,000 malicious videos to spread infostealer malware, primarily Lumma and Rhadamanthys, targeting users seeking game cheats and pirated software. The operation involves compromised YouTube accounts assigned roles such as video uploaders, community posters, and interaction simulators. Active since 2021, the campaign saw a significant rise in activity in 2025. It employs evasion techniques including password-protected archives and frequent updates to malware payloads and command-and-control infrastructure. The campaign highlights evolving malware distribution tactics on popular platforms.
TA585 is a sophisticated cybercriminal threat actor that operates its entire attack chain, from infrastructure to email delivery and malware installation. The actor demonstrates innovation in the evolving cybercrime landscape, using unique web injection campaigns and complex filtering techniques. TA585 frequently delivers MonsterV2, a versatile malware with remote access trojan, loader, and stealer capabilities. MonsterV2 is used by multiple threat actors and avoids infecting computers in Commonwealth of Independent States countries. The malware is actively maintained and updated, with pricing ranging from $800 to $2,000 per month. TA585's campaigns often involve compromised websites, fake CAPTCHAs, and GitHub-themed attacks to deliver various payloads.
Rhadamanthys, a complex multi-modular stealer, has released version 0.9.2 with significant updates. The malware now uses PNG files to deliver payloads, implements new evasion techniques, and introduces changes to its custom executable formats. Key modifications include a new message box mimicking Lumma stealer, updates to string encryption, and enhanced configurability. The malware continues to evolve, focusing on refinements and customization options while maintaining its core design. These changes aim to disrupt analysis tools and detection methods. The authors are professionalizing their operation, treating Rhadamanthys as a long-term business venture with tiered pricing and expanded product offerings.
This analysis examines Infostealer trends in August 2025, focusing on distribution volume, methods, and disguises. AhnLab's automated systems collect and analyze malware, providing real-time IOC services. Infostealers, often disguised as cracks, are distributed through SEO poisoning. Notable variants include LummaC2, ACRStealer, and Rhadamanthys. Distribution methods evolved from personal blogs to legitimate websites, bypassing search engine restrictions. Malware is primarily distributed as EXE files (89.7%) or through DLL-SideLoading (10.3%). Two significant trends emerged: mass distribution via Slack Marketplace and ACRStealer's domain masquerading technique, which now targets security company domains to evade detection.
In 2025, a significant surge in phishing attacks targeting major U.S.energy companies was observed. The campaign primarily focused on Chevron, ConocoPhillips, PBF Energy, and Phillips 66, utilizing sophisticated impersonation techniques. Attackers employed HTTrack-based cloning to replicate legitimate websites, creating over 1,465 phishing domains. The infrastructure was distributed across multiple hosting providers and countries to evade takedowns. Notably, Chevron faced the highest volume of impersonation attempts with 158 fake domains. The phishing sites combined credential harvesting with investment scam frameworks, enhancing their profitability. Many malicious domains showed low detection rates across security vendors, exposing gaps in current defense systems. The analysis highlights the need for improved threat intelligence integration and faster mitigation strategies in the energy sector.
A massive distribution of SmartLoader malware has been discovered through GitHub repositories masquerading as legitimate projects. These repositories focus on topics like game cheats, software cracks, and automation tools to attract users. The malware is distributed via compressed files containing a legitimate Lua loader executable, a malicious batch file, and an obfuscated Lua script. Once executed, SmartLoader establishes persistence, sends system information to a C2 server, and downloads additional payloads. The malware has been observed downloading InfoStealer malware such as Rhadamanthys, Redline, and Lumma Stealer. Users are advised to download software only from official sources and to carefully verify the credibility of GitHub repositories before use.
A Malware-as-a-Service operation utilizing Amadey for payload delivery has been identified, with connections to a SmokeLoader phishing campaign targeting Ukrainian entities. The operation exploits fake GitHub accounts to host payloads and tools, bypassing web filtering. Emmenhtal, a multistage downloader, is used to download Amadey and other malware. The activity involves various malware families and GitHub repositories for staging custom payloads. Similarities in tactics and indicators between the SmokeLoader campaign and Amadey MaaS activity have been observed. The operation demonstrates adaptability in delivering diverse tooling, including legitimate software like PuTTY. The threat actors employ sophisticated obfuscation techniques and leverage public platforms for malware distribution.
This analysis provides insights into Infostealer malware trends observed in June 2025. The data, collected through various automated systems, reveals changes in distribution methods and malware types. While LummaC2 has been dominant, June saw increased activity from Rhadamanthys, ACRStealer, Vidar, and StealC. A new variant of ACRStealer emerged, using advanced techniques like HTTP host domain spoofing and anti-analysis methods. Distribution via crack disguises decreased, with 94.4% in EXE format and 5.6% using DLL-SideLoading. A unique malware type was observed, creating an uncontrollable window prompting browser updates. Some samples now hide compression passwords in image files, indicating evolving evasion tactics.
This analysis examines the distribution trends of Infostealer malware in May 2025. It highlights the use of SEO poisoning to distribute malware disguised as cracks and keygens. LummaC2, Vidar, StealC, Rhadamanthys, and Amadey were the main Infostealers observed. Distribution methods included posts on legitimate websites, forums, and Q&A pages. Malware was primarily distributed in EXE format (95.4%), with a decrease in DLL-SideLoading (4.6%). Notable trends include the emergence of BAT script malware, use of the Wormhole file-sharing service for distribution, and the use of Unicode characters in compression passwords to bypass security measures. The report provides insights into distribution volumes, methods, and disguises based on data collected and analyzed by advanced security systems.
Proofpoint has identified Amatera Stealer, a rebranded version of ACR Stealer with enhanced capabilities and evasion techniques. Distributed via ClearFake website injects, it utilizes sophisticated attack chains and web injects. Amatera Stealer employs NTSockets for stealthy C2 communication, WoW64 Syscalls to bypass user-mode hooking, and supports HTTPS requests. It focuses on stealing information from browsers, crypto wallets, and various software. The malware can also execute secondary payloads. Amatera Stealer is actively developed and sold as a malware-as-a-service, with subscription plans ranging from $199 to $1,499.
The CopyRh(ight)adamantys campaign is a large-scale phishing operation distributing the latest Rhadamanthys stealer malware (version 0.7). It uses copyright infringement-themed bait emails impersonating companies mainly in the Entertainment/Media and Technology/Software sectors. The campaign employs automation and possibly AI tools to tailor and distribute lures globally. Rhadamanthys stealer targets sensitive information by leveraging various techniques including OCR-based text recognition and credential theft. Although the malware claims AI-powered capabilities, it primarily uses traditional machine learning OCR methods. The campaign is financially motivated, not linked to nation-state actors, and poses a medium severity threat. European organizations in targeted sectors face risks of data theft and credential compromise. Mitigation requires targeted phishing awareness, enhanced email filtering, endpoint detection, and credential hygiene. Countries with strong media and tech industries and high phishing exposure are most at risk.
