VOLTA is a sophisticated Windows infostealer malware that uses AI to automate the analysis and prioritization of stolen victim data for cybercriminals. It collects credentials, sessions, and cryptocurrency wallets, then uses large language models to rank assets by monetization value, enriching data with stolen session access and email content. This automation reduces the manual effort attackers need to monetize stolen data and enables even low-level actors to bypass security controls and prioritize high-value targets. Organizations are advised to implement application control policies to detect suspicious Chrome headless processes, secure credential storage, revoke compromised tokens, change passwords, and treat exposed cryptocurrency wallets as fully compromised.
A vulnerability in ChatGPT's Lockdown Mode allows attackers to bypass its protections and perform silent exfiltration of chat data, phishing, and malware delivery. The flaw exploits the HTML page preview feature in ChatGPT code blocks, enabling attackers to open pop-up windows outside the sandbox, exfiltrate entire chat histories without user interaction, and display phishing content. This bypass undermines Lockdown Mode's restrictions on outbound network requests and content loading. The attack surface includes data theft, social engineering via interface manipulation, and malware distribution within the trusted ChatGPT interface.
Several White House staffers are using devices with residential proxy services that allow others to rent their internet addresses, including a Russian proxy service linked to past election cyberattacks. Security firms detected persistent proxy activity originating from the White House network (AS2650) involving multiple proxy providers, enabling public access to these IP addresses. Residential proxy services can anonymize traffic but also pose security risks, including facilitating malicious activities and exposing local networks to malware. These proxies often come from employee devices and have been found widely in government and corporate networks. Vulnerabilities in such services have been exploited to gain control over devices within local networks. This situation highlights significant security concerns for trusted government networks.
A botnet named XCPE, composed of over 1.3 million Android TV set-top boxes infected with malware, was dismantled by Google and industry partners. The malware covertly rented out users' Internet connections as residential proxies for criminal activities such as advertising fraud and account takeover. The botnet also infected other Android-based IoT devices including healthcare devices and karaoke machines. The malware was intentionally pre-installed by the manufacturer Videostrong Technology Co., which Google has legally acted against. The takedown is expected to disrupt access to over 10 million additional compromised devices. Several residential proxy providers were implicated in the operation of this botnet. Google has since addressed the security weaknesses exploited by Videostrong.
VOLTA-AUTODOXER is a self-deleting Windows infostealer that uses AI to prioritize stolen victim data by monetization value. After harvesting credentials, sessions, wallets, and other sensitive data, it builds a detailed evidence bundle about the victim, enriched via stolen session access. This bundle is sent to an LLM via OpenRouter, which acts as an intelligence analyst to rank the victim's assets and assess their security posture. The malware uses a multi-model fallback approach to ensure the AI completes the assessment despite model refusals. VOLTA operates as a builder model, allowing customers to compile customized binaries with their own API keys. There is no attacker-owned infrastructure, and exfiltration occurs in multiple stages. The AI integration represents a novel approach in malware by automating victim profiling and prioritization.
Microsoft Threat Intelligence analyzed two human-operated intrusion cases where attackers used Microsoft Teams to impersonate IT support and socially engineer users into granting remote access via Quick Assist. The attackers then staged portable runtimes, performed host and network discovery, and abused Active Directory Certificate Services (AD CS) to obtain authentication certificates for privileged accounts. Two distinct AD CS abuse techniques were observed: ESC1 template abuse allowing certificate issuance for privileged accounts, and ESC8 NTLM relay to AD CS Web Enrollment to obtain domain-controller certificates. These certificate-based authentications enabled persistent privileged access independent of passwords. The incidents highlight risks from social engineering combined with insecure or overly permissive AD CS configurations rather than software vulnerabilities. Microsoft Defender detections and containment effectiveness varied between cases, influencing the extent of attacker progression.
Cybercriminals exploit the popularity of ebooks by distributing malware disguised as ebook files, using search result manipulation, phishing scams, and fake pirate library sites. These attacks can lead to malware infections such as cryptocurrency miners and infostealers, data leaks including personal and financial information, and fraudulent charges via fake bookstore sites. E-readers themselves are less commonly targeted, but vulnerabilities exist in advanced models. Users' reading histories and personal data can also be exposed through data leaks or account compromises. Safe practices include using official bookstores, keeping devices and apps updated, and avoiding unofficial firmware.
This report summarizes multiple cybersecurity incidents and vulnerabilities including a critical 16-year-old buffer overflow in UEFI firmware affecting most servers worldwide, a disputed AI-led intrusion at Anthropic, data breaches at the University of Copenhagen and ENGIE, and a Cloudflare AI data leak. The UEFI flaw allows local network attackers to execute code at high privilege during network boot. Anthropic disputes the scope of an AI-driven intrusion by a startup. The University of Copenhagen breach exposed personal and national ID data of over 500,000 individuals. Cloudflare fixed a cross-tenant data leak in its AI inference platform caused by a cache size mismatch and a bug in the inference engine. Other incidents include suspected Iranian cyberattacks on military freezers and attempts to breach the US Federal Election Commission systems.
North Korean threat actors are increasingly integrating AI technologies into their cyber operations to enhance speed, efficiency, and complexity. These actors use AI to accelerate social engineering, targeting, malware development, and operational scaling, but their activities remain human-directed rather than fully autonomous. Key groups such as Jade Sleet, Coral Sleet, and Jasper Sleet have operationalized AI across their workflows, while others are at various stages of adoption. Their operations primarily target cryptocurrency, financial organizations, IT companies, government entities, and education sectors to generate revenue, collect intelligence, and pursue political objectives. Microsoft provides mitigation guidance and Defender detections to help defend against these evolving threats.
A critical zero-day vulnerability (CVE-2026-20127) in Cisco SD-WAN controllers has been exploited by a sophisticated threat actor linked to the Volt Typhoon 'Silent Train' network. The vulnerability allows remote attackers to bypass authentication and gain administrative access to SD-WAN controllers without credentials. Attackers deployed KeyDrift malware to maintain persistent access and steal encryption keys, targeting organizations in sectors such as defense, aerospace, telecommunications, and critical infrastructure. The campaigns have been ongoing since at least late 2025, affecting entities primarily in the United States, with a focus on regions including Hawaii, California, and Texas.
NIK C2 is a newly observed AI-enabled command-and-control (C2) framework targeting Windows systems. It features a Go-based implant that provides persistent remote access with capabilities including credential dumping, keystroke streaming, RDP session hijacking, process hollowing, and deployment of screen-locking ransomware demanding cryptocurrency payments. The implant uniquely integrates the OpenAI Chat Completions API, allowing threat actors to analyze harvested data using AI in real time within the compromised session. This integration collapses traditional offline data processing delays, complicating detection and response efforts. The implant communicates over encrypted channels with resilient fallback mechanisms and has been observed in multiple evolving versions compiled no earlier than late 2024. Active infrastructure associated with NIK C2 has been identified in the wild, confirming its operational status.
This content describes deploying a local Ghidra Server to support malware analysis labs by providing project persistence, versioning, collaboration, and confidentiality. It explains how Ghidra Server enables storing analysis projects outside virtual machines to avoid data loss when reverting snapshots. The server is lightweight, runs on Linux with Java, and uses Java RMI for communication. Proper network configuration and user authentication are important. The server hosts raw malware samples but does not execute them, so it should be isolated within the lab network and never exposed to the Internet.
Cisco Talos identified UAT-11473, an advanced persistent threat actor exploiting CVE-2026-20127, an authentication bypass in Cisco SD-WAN Catalyst Controller instances. The attacker deployed a custom malware framework named KeyDrift, designed to persistently access and exfiltrate sensitive keying material from SD-WAN control-plane processes. KeyDrift includes components such as PolyFlame, which injects into legitimate controller processes to intercept decrypted control traffic, and LittleFilch, which collects and stages keying material for exfiltration. The malware operates stealthily within the controller environment, leveraging modified service scripts and runit service paths for persistence and execution. Targeting spans space and satellite services, defense, aerospace, and policy research sectors. The threat actor's infrastructure overlaps with the Volt Typhoon group’s covert network.
JNISleuth-MCP is an open-source tool designed to assist analysts in unpacking and analyzing stripped Android native (.so) malware libraries. It automates the recovery of JNI RegisterNatives bindings, deobfuscates ARM64 OLLVM indirect branches, and bypasses anti-Frida tracing techniques using a stealth Dobby agent. The tool integrates with reverse engineering platforms like Ghidra and IDA Pro to provide unified cross-boundary analysis from Java to native code, facilitating malware triage and research.
Check Point Research identified a malware campaign named Yoba that abuses the popularity of AI tools like ChatGPT and Claude by creating fake GitHub projects impersonating AI desktop applications such as GPT-5.6 Sol, GPT-6 Astra, and Claude Opus. These projects host malicious installers containing a dropper and loader that fetch and execute malware. The campaign uses DLL sideloading with legitimate signed binaries to evade detection and employs strong operational security by avoiding traditional command-and-control infrastructure, instead using Telegram, Tor, Catbox, and DNS over HTTPS. The threat actor shows indicators of Russian origin and operates primarily in the UTC-4 timezone. The malicious releases are hosted on GitHub without password protection, leveraging SEO poisoning to appear in top Google search results and lure victims.
This report summarizes multiple cyber threats and incidents observed in late October 2026, including large-scale breaches, supply-chain attacks, AI-enabled cybercrime tools, and critical software vulnerabilities. Notable incidents include a cyberattack on Japanese cloud provider Onamae Desktop Cloud affecting over 26,000 servers, a data breach at India's State Bank of India exposing nearly 3,900 customer records, and a supply-chain compromise of the LiteLLM AI gateway. AI-enabled tools are facilitating large-scale fraud and ransomware attacks targeting critical sectors in the United States. Critical PHP vulnerabilities enabling remote code execution are being widely exploited, with most instances unpatched. Oracle and GitLab have released patches for numerous critical vulnerabilities, including a critical remote code execution flaw in GitLab and a privilege escalation vulnerability in Arista VeloCloud Orchestrator that has been exploited in the wild.
A long-running malware campaign has infected 953 online stores, primarily Magento-based, with loader code that selectively serves malicious payloads to real visitors to evade detection. The campaign began in July 2025 and has used multiple domains for its JavaScript loader, with recent activity observed as late as October 2026. The loader performs visitor fingerprinting and has been delivered through exploitation of known Magento vulnerabilities and backdoors. Store owners are advised to check for loader code, vulnerabilities, backdoors, rogue admin accounts, and to apply the latest patches and security measures.
This report analyzes Hermes, an active malware-as-a-service platform operating in the Russian cybercrime market, along with its PyMapper distribution chain. The analysis is based on a detailed technical report authored by a researcher and publicly available. The platform facilitates malware distribution and service offerings typical of malware-as-a-service ecosystems. No specific software vulnerabilities or exploits are detailed in the provided information.
This content describes a technology used by Kaspersky to detect and block email spam domains generated by domain generation algorithms (DGAs). Spammers use DGAs to create many similar domain names automatically, making it difficult to block spam by individual domain names. Kaspersky's solution clusters similar domains and creates rules to block entire batches of these generated domains, improving spam filtering efficiency. The technology is part of Kaspersky's corporate email security solutions and helps reduce spam, phishing, and malware emails.
Aokigahara is a newly observed infostealer malware that targets a wide range of sensitive data on Windows systems. It steals traditional credentials such as browser passwords, cookies, cryptocurrency wallets, VPN and FTP client data, email profiles, and password manager files. Uniquely, it also targets AI applications and developer tools including Claude, ChatGPT Desktop, Codex, Cursor, Windsurf, GitHub Copilot, and Gemini CLI, collecting authentication files, session tokens, and configuration data. The malware additionally targets cloud environment credentials for AWS, Azure, and Google Cloud, as well as developer environment secrets like Git credentials and SSH private keys. Stolen AI and cloud credentials could enable unauthorized access to paid AI services, cloud resources, and source code repositories. The malware includes capabilities such as screenshot capture, clipboard theft, and a clipper component for cryptocurrency manipulation. There is no confirmed information on how the stolen data is monetized. No patch or remediation guidance is provided.