Find what you’re looking for
Browse everything
6,903 results in Threat Reports
Cybercriminals Exploit Remote Access Tools in 2026 Attacks
Between January and March 2026, cybercriminals exploited legitimate remote access tools such as LogMeIn and ScreenConnect to gain unauthorized access to victim devices. These attacks, detailed in HP's Threat Insights Report, involved phishing emails that tricked users into installing these tools, allowing attackers to control systems without triggering security alerts. The abuse of trusted software enabled threat actors to blend malicious activities with normal IT operations, complicating detection and response efforts. This incident underscores a growing trend where attackers leverage legitimate remote monitoring and management (RMM) tools to establish persistent access and deploy malware. The increasing sophistication of such tactics highlights the need for organizations to enhance monitoring of software installations, enforce strict privilege controls, and update defenses to detect and prevent misuse of trusted applications.
Understanding the HTTP/2 Bomb (CVE-2026-49975) Vulnerability
In June 2026, a critical denial-of-service vulnerability, CVE-2026-49975, known as the "HTTP/2 Bomb," was disclosed. This flaw exploits the HPACK compression and flow control features of the HTTP/2 protocol, allowing attackers to send minimal requests that rapidly exhaust server memory. Major web servers, including NGINX, Apache HTTPD, Microsoft IIS, Envoy, and Cloudflare Pingora, are affected in their default configurations. The attack can be executed remotely without authentication, leading to immediate service disruptions. ([imperva.com](https://www.imperva.com/blog/imperva-customers-protected-against-cve-2026-49975-http-2-bomb-dos/?utm_source=openai)) The discovery of this vulnerability underscores the evolving threat landscape, where attackers leverage protocol features to amplify attacks. Organizations must prioritize patching affected systems and consider implementing additional security measures, such as Web Application Firewalls (WAFs), to mitigate potential exploits. ([imperva.com](https://www.imperva.com/blog/imperva-customers-protected-against-cve-2026-49975-http-2-bomb-dos/?utm_source=openai))
Debate Erupts Over U.S. Ban on Anthropic's AI Models
In June 2026, the U.S. government issued an export control order restricting foreign nationals from accessing Anthropic's advanced AI models, Claude Fable 5 and Mythos 5, citing national security concerns. This led Anthropic to suspend the models' use for all customers to ensure compliance. The security community criticized the decision, arguing that it hampers defenders' access to crucial tools while doing little to prevent adversaries from developing similar capabilities. Experts highlighted that such restrictions might inadvertently accelerate the development of decentralized, open-source alternatives, potentially diminishing U.S. leadership in AI security. The incident underscores the delicate balance between national security and technological advancement, emphasizing the need for policies that support innovation while mitigating risks.
FortiBleed: Unprecedented Credential Harvesting Compromises 30,000+ Fortinet Devices
In June 2026, a large-scale credential harvesting operation, dubbed 'FortiBleed,' targeted Fortinet firewalls and VPN gateways, compromising over 30,000 internet-facing devices across nearly 200 countries. The attackers, suspected to be Russian-speaking threat actors, utilized automated tools to collect and verify login credentials, exploiting weak or default passwords without leveraging any known Fortinet vulnerabilities. This campaign affected various sectors, including government, telecommunications, healthcare, education, financial services, and critical infrastructure, with significant concentrations in India and the United States. The operation underscores the critical importance of robust password policies and the implementation of multi-factor authentication to safeguard network security devices. ([darkreading.com](https://www.darkreading.com/cyberattacks-data-breaches/sweeping-credential-harvesting-heist-compromises-30k-fortinet-devices?utm_source=openai))
Phantom Stealer: The Rise of Fileless Malware Targeting Financial Institutions
In June 2026, a sophisticated phishing campaign targeted banks and high-value organizations, deploying Phantom Stealer—a fileless malware designed to evade traditional endpoint defenses. The attack began with phishing emails containing seemingly legitimate business documents. Upon opening, a heavily obfuscated batch file initiated a multistage infection chain, injecting Phantom Stealer into the Windows Explorer process. Operating entirely in memory, the malware silently exfiltrated browser credentials, session cookies, and financial data through multiple channels, including Telegram, Discord, FTP, and SMTP. This incident underscores the evolving tactics of cybercriminals, highlighting the increasing use of fileless malware and advanced evasion techniques. Organizations must enhance their security posture by adopting behavior-based detection systems and educating employees on recognizing sophisticated phishing attempts to mitigate such threats.
CISA Adds CVE-2026-48907 to Known Exploited Vulnerabilities Catalog
In June 2026, a critical vulnerability identified as CVE-2026-48907 was discovered in the Joomla Content Editor (JCE) extension, allowing unauthenticated attackers to create new editor profiles and upload arbitrary PHP code, leading to remote code execution. This flaw affects JCE versions prior to 2.9.99.5. The Cybersecurity and Infrastructure Security Agency (CISA) added this vulnerability to its Known Exploited Vulnerabilities (KEV) catalog on June 16, 2026, following evidence of active exploitation. Joomla released patches on June 3 and June 6, 2026, to address this issue. ([securityweek.com](https://www.securityweek.com/joomla-litespeed-vulnerabilities-exploited-in-attacks/?utm_source=openai)) The active exploitation of CVE-2026-48907 underscores the persistent threat posed by web application vulnerabilities, particularly in widely used content management systems like Joomla. Organizations are urged to promptly apply the latest security updates to mitigate potential risks associated with this vulnerability.
CISA Issues Warning on Actively Exploited Joomla JCE Vulnerability
In June 2026, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) added a critical vulnerability, CVE-2026-48907, affecting the Joomla Content Editor (JCE) to its Known Exploited Vulnerabilities catalog. This flaw, present in JCE versions 1.0.0 through 2.9.99.4, allows unauthenticated users to create new editor profiles, enabling the upload and execution of arbitrary PHP code on the server. The vulnerability has been actively exploited, with attackers leveraging it to gain unauthorized access and control over affected Joomla installations. The active exploitation of this vulnerability underscores the persistent threat posed by improper access controls in widely used content management systems. Organizations utilizing Joomla with the JCE extension are urged to update to version 2.9.99.5 or later to mitigate this risk. Additionally, administrators should audit their systems for unauthorized editor profiles and monitor server logs for suspicious activity to prevent potential breaches.
Unveiling the VHDX-Based Remcos RAT Attack: A 2026 Cybersecurity Challenge
In June 2026, a sophisticated malware campaign was identified, utilizing a VHDX disk image within a ZIP archive to deliver the Remcos Remote Access Trojan (RAT). Upon extraction, the VHDX file auto-mounted on Windows systems, revealing an obfuscated JavaScript file named 'Partnerschaft_fur_neue_Angebotsanfrage.js', indicating potential targeting of German-speaking users. This JavaScript initiated a multi-stage infection chain involving PowerShell scripts and .NET loaders, ultimately injecting the Remcos RAT into the 'backgroundTaskHost.exe' process. The malware established communication with a command-and-control server at animal342[.]duckdns[.]org:53552, enabling remote surveillance and data exfiltration. Notably, the campaign employed techniques such as WMI for process execution and Base64 encoding with XOR decryption to evade detection by traditional security measures. This incident underscores the evolving tactics of cybercriminals who leverage legitimate system features and complex obfuscation methods to bypass security controls. The use of VHDX files as malware containers highlights the need for enhanced vigilance and advanced detection mechanisms to counter such sophisticated threats.
Mastra npm Supply Chain Attack: 144 Packages Compromised
In June 2026, a significant supply chain attack targeted the Mastra npm ecosystem, compromising 144 packages associated with the '@mastra' namespace. The attack was initiated through the hijacking of a former contributor's npm account, 'ehindero'. The attackers introduced a malicious dependency named 'easy-day-js', which masqueraded as the legitimate 'dayjs' library. Initially, 'easy-day-js' appeared benign, but subsequent versions contained obfuscated post-install scripts designed to exfiltrate sensitive information from developers' systems. This incident underscores the vulnerabilities inherent in open-source software supply chains, particularly when trusted contributor accounts are compromised. The Mastra framework, widely used for building AI applications, saw its core components, such as '@mastra/core', affected, amplifying the potential impact on downstream projects and organizations. The attack highlights the critical need for robust security measures in package management and dependency verification processes. The increasing frequency of such supply chain attacks emphasizes the urgency for the developer community to adopt stringent security practices, including regular audits of dependencies, implementation of multi-factor authentication for contributor accounts, and continuous monitoring for anomalous activities within software ecosystems.
Malicious JetBrains Plugins Compromise AI API Keys in 2026
In June 2026, cybersecurity researchers identified a coordinated malware campaign involving at least 15 malicious plugins on the JetBrains Marketplace. These plugins, masquerading as AI coding assistants built on DeepSeek and other large language models, were designed to exfiltrate artificial intelligence (AI) provider keys. The plugins offered functionalities such as chat, commit messages, code review, bug finding, and unit tests, thereby enticing developers to install them. Once installed, the plugins covertly transmitted sensitive API keys to attacker-controlled servers, potentially compromising the security of AI-driven applications and services. This incident underscores a growing trend where threat actors exploit the trust in developer tools and marketplaces to distribute malicious software. The increasing integration of AI into development workflows makes such platforms attractive targets. Organizations must remain vigilant, ensuring the integrity of the tools they incorporate and regularly auditing their development environments to prevent unauthorized access and data exfiltration.
Understanding the MongoBleed Vulnerability (CVE-2025-14847) and Its Impact
In December 2025, a critical vulnerability known as MongoBleed (CVE-2025-14847) was disclosed, affecting multiple versions of MongoDB Server from 3.6 through 8.2.3. This flaw allows unauthenticated attackers to exploit improper handling of zlib-compressed network traffic, leading to the leakage of uninitialized heap memory. As a result, sensitive data such as credentials, session tokens, and API keys could be exfiltrated from affected servers. The vulnerability has been actively exploited in the wild, with approximately 87,000 MongoDB instances exposed globally, primarily in the United States, China, and Germany. Organizations are strongly advised to apply security patches immediately or disable compression and restrict network exposure to mitigate the risk. ([infoq.com](https://www.infoq.com/news/2026/01/mongodb-mongobleed-vulnerability/?utm_source=openai)) The MongoBleed incident underscores the critical importance of timely patch management and the need for robust security measures to protect against vulnerabilities in widely used database systems. The rapid exploitation of this flaw highlights the evolving threat landscape and the necessity for organizations to remain vigilant in securing their infrastructure.
Introducing Sulla: Praetorian's Open-Source SMB Secret Scanner
In June 2026, Praetorian released Sulla, an open-source tool designed to scan SMB shares for exposed credentials across enterprise networks. Sulla automates the discovery of readable SMB shares, traverses their file structures, and scans contents for sensitive information using the Titus detection library. This tool addresses the challenge of manually reviewing numerous network shares, which is often tedious and inefficient. By integrating Sulla into their Continuous Threat Exposure Management platform, Guard, Praetorian ensures that SMB secrets are identified promptly as they appear in environments. The release of Sulla highlights the growing need for automated tools to detect and mitigate the risks associated with exposed credentials in network shares. As organizations increasingly rely on complex network infrastructures, tools like Sulla become essential in proactively identifying and addressing security vulnerabilities, thereby enhancing overall cybersecurity posture.