Find what you’re looking for
Browse everything
6,903 results in Threat Reports
Iranian APT Exploits PLC Vulnerabilities in U.S. Critical Infrastructure
In April 2026, Iranian-affiliated advanced persistent threat (APT) actors targeted internet-facing operational technology (OT) devices, specifically programmable logic controllers (PLCs) manufactured by Rockwell Automation/Allen-Bradley, across multiple U.S. critical infrastructure sectors. These attacks led to disruptions in energy, water, and government facilities by manipulating project files and tampering with human-machine interface (HMI) and supervisory control and data acquisition (SCADA) displays, resulting in operational disruptions and financial losses. ([databreaches.net](https://databreaches.net/2026/04/07/iranian-affiliated-cyber-actors-exploit-programmable-logic-controllers-across-us-critical-infrastructure/?utm_source=openai)) This incident underscores the escalating cyber threats from nation-state actors targeting critical infrastructure, highlighting the urgent need for enhanced cybersecurity measures and vigilance in protecting OT environments.
GrafanaGhost: Unveiling the AI Vulnerability Exposing Enterprise Data
In April 2026, a critical vulnerability named 'GrafanaGhost' was discovered in Grafana's AI components, allowing attackers to exfiltrate sensitive enterprise data through indirect prompt injection. By embedding malicious instructions within external web content, attackers could manipulate Grafana's AI to process these prompts as legitimate, leading to unauthorized data exposure without user interaction. This flaw was promptly patched by Grafana following responsible disclosure. The GrafanaGhost incident underscores the growing risks associated with integrating AI into enterprise systems. It highlights the necessity for robust security measures to prevent AI-specific vulnerabilities, as attackers increasingly exploit such weaknesses to access sensitive information.
Latin American Banks Confront 155% Surge in Social Engineering Scams in 2025
In 2025, Latin American financial institutions experienced a 155% increase in social engineering scams, with fraud attempts utilizing remote-access tools surging fivefold and malware attacks rising by 225%. This escalation underscores a shift in fraudsters' tactics, moving from basic phishing to sophisticated methods that exploit human behavior and technological vulnerabilities. The surge in fraud cases highlights the urgent need for enhanced security measures and collaborative efforts among financial institutions to combat evolving threats.
Understanding the Rise of Web Shell Attacks in 2026
In early 2026, a significant surge in web shell attacks was observed, targeting vulnerabilities in web servers and applications. Attackers exploited these weaknesses to deploy malicious scripts, enabling remote control over compromised systems. This escalation led to unauthorized data access, service disruptions, and substantial financial losses for affected organizations. The rapid deployment and stealthy nature of web shells posed significant challenges to traditional security measures. ([microsoft.com](https://www.microsoft.com/en-us/security/blog/2026/04/02/cookie-controlled-php-webshells-tradecraft-linux-hosting-environments/?utm_source=openai)) This trend underscores the evolving tactics of cyber adversaries, emphasizing the need for enhanced monitoring and proactive defense strategies. Organizations must prioritize the detection and mitigation of web shell threats to safeguard their digital assets and maintain operational integrity.
Mitsubishi Electric's 2025 Vulnerability: A Wake-Up Call for Industrial Security
In May 2025, Mitsubishi Electric disclosed a vulnerability (CVE-2025-0921) in their GENESIS64, MC Works64, and GENESIS products. This flaw allows local attackers to perform unauthorized writes to arbitrary files by exploiting symbolic links, potentially leading to denial-of-service conditions. The vulnerability affects all versions of GENESIS64 and MC Works64, as well as GENESIS version 11.00. Mitsubishi Electric has released patches and mitigation strategies to address this issue. ([mitsubishielectric.com](https://www.mitsubishielectric.com/psirt/vulnerability/pdf/2025-002_en.pdf?utm_source=openai)) This incident underscores the critical importance of securing industrial control systems against local privilege escalation attacks, which can disrupt essential operations. Organizations are urged to apply the provided patches promptly and review their security protocols to prevent similar vulnerabilities.
AI-Driven Ransomware Attack 2026: Lessons Learned
In early 2026, a sophisticated AI-driven ransomware attack targeted multiple organizations, exploiting vulnerabilities in AI systems to gain unauthorized access. The attackers utilized autonomous AI agents to conduct reconnaissance, escalate privileges, and deploy ransomware, significantly reducing the time from initial breach to full system encryption. This rapid progression left organizations with minimal time to detect and respond, resulting in substantial operational disruptions and financial losses. This incident underscores the escalating threat posed by AI-enhanced cyberattacks, highlighting the need for organizations to adopt advanced, AI-driven defense mechanisms. The convergence of AI and cybercrime necessitates a proactive approach to cybersecurity, emphasizing rapid detection, response, and recovery strategies to mitigate the impact of such attacks.
Iranian APT Exploits U.S. Critical Infrastructure PLCs in 2026
In April 2026, Iranian-affiliated advanced persistent threat (APT) actors exploited internet-facing operational technology (OT) devices, notably Rockwell Automation/Allen-Bradley programmable logic controllers (PLCs), across multiple U.S. critical infrastructure sectors. The attackers accessed these devices via default or weak credentials, leading to disruptions through malicious interactions with project files and manipulation of data on human-machine interface (HMI) and supervisory control and data acquisition (SCADA) displays, resulting in operational disruptions and financial losses. ([publicpower.org](https://www.publicpower.org/periodical/article/iranian-affiliated-cyber-actors-exploit-programmable-logic-controllers-across-us-critical?utm_source=openai)) This incident underscores the escalating threat posed by nation-state actors targeting critical infrastructure. The exploitation of OT devices highlights the urgent need for organizations to secure internet-facing systems, implement strong authentication measures, and regularly update and patch their systems to mitigate such risks.
LiteLLM Supply Chain Compromise: A Wake-Up Call for Open-Source Security
In March 2026, the LiteLLM Python package, widely used for routing large language model (LLM) API calls, was compromised through a supply chain attack. Malicious versions 1.82.7 and 1.82.8 were uploaded to the Python Package Index (PyPI) after attackers gained access to the maintainer's credentials via a compromised Trivy security scanner in LiteLLM's CI/CD pipeline. These versions contained a credential-stealing payload that executed automatically on Python startup, exfiltrating sensitive information such as SSH keys, cloud provider credentials, and Kubernetes secrets to an attacker-controlled server. The malicious packages were available for approximately three hours before being removed from PyPI. ([snyk.io](https://snyk.io/blog/poisoned-security-scanner-backdooring-litellm/?utm_source=openai)) This incident underscores the escalating threat of supply chain attacks targeting open-source ecosystems. The rapid propagation of malicious code through widely used packages highlights the need for enhanced security measures in software development pipelines, including stringent credential management, regular security audits, and the implementation of tools like Software Bill of Materials (SBOMs) and SigStore for verifying package integrity. ([ionix.io](https://www.ionix.io/threat-center/litellm-supply-chain-compromise-backdoored-pypi-packages-1-82-7-1-82-8/?utm_source=openai))
FBI's 2025 Internet Crime Report: A 26% Surge in Cybercrime Losses
In 2025, the FBI's Internet Crime Complaint Center (IC3) reported a significant surge in cybercrime, with total losses reaching $20.9 billion—a 26% increase from the previous year. The center received over one million complaints, marking a 17% rise compared to 2024. Investment-related fraud led to losses of nearly $8.65 billion, while business email compromise accounted for almost $3.05 billion. Phishing remained the most reported cybercrime, followed by extortion and personal data breaches. ([cyberscoop.com](https://cyberscoop.com/fbi-internet-crime-complaint-center-annual-cybercrime-report/?utm_source=openai)) This escalation underscores the growing sophistication of cybercriminals, who are increasingly leveraging artificial intelligence to enhance their attacks. The trend highlights the urgent need for organizations to bolster their cybersecurity measures and stay vigilant against evolving threats. ([forbes.com](https://www.forbes.com/sites/timkeary/2026/04/07/fbi-reports-208-billion-lost-to-cybercrime-as-hackers-turn-to-ai/?utm_source=openai))
Anthropic's Project Glasswing: Revolutionizing Cybersecurity with AI
In April 2026, Anthropic launched Project Glasswing, a collaborative initiative with major technology companies including Amazon, Apple, Microsoft, and Cisco, to enhance cybersecurity defenses using advanced AI. Central to this project is Claude Mythos Preview, an unreleased AI model that has identified thousands of previously undetected vulnerabilities across critical software systems, including a 27-year-old bug in OpenBSD and a 16-year-old flaw in FFmpeg. To mitigate potential misuse, Anthropic has restricted access to this powerful model to select partners and has committed significant resources to support open-source security organizations. ([anthropic.com](https://www.anthropic.com/project/glasswing?utm_source=openai)) This initiative underscores the growing importance of AI in cybersecurity, highlighting both its potential to fortify defenses and the risks associated with its misuse. As AI capabilities advance, the industry faces the dual challenge of leveraging these tools for protection while preventing their exploitation by malicious actors. ([cyberscoop.com](https://cyberscoop.com/project-glasswing-anthropic-ai-open-source-software-vulnerabilities/?utm_source=openai))
Forest Blizzard 2026: Unveiling and Neutralizing a Global Espionage Threat
In early 2026, the Russian state-sponsored group Forest Blizzard (APT28) compromised over 18,000 routers across 120 countries, exploiting known vulnerabilities in TP-Link and MikroTik devices. By hijacking DNS settings, they conducted adversary-in-the-middle attacks, intercepting credentials and tokens for services like Microsoft Outlook Web Access. This extensive espionage campaign targeted more than 200 organizations, including government agencies and critical infrastructure sectors. A collaborative effort led by the FBI, known as Operation Masquerade, successfully neutralized the threat by resetting DNS settings and preventing further exploitation. This incident underscores the persistent threat posed by state-sponsored cyber actors and highlights the critical need for robust network security measures. Organizations must remain vigilant, regularly update and patch network devices, and implement comprehensive monitoring to detect and mitigate such sophisticated attacks.
German Authorities Unmask Leaders Behind REvil and GandCrab Ransomware Attacks
In April 2026, Germany's Federal Criminal Police Office (BKA) identified Russian nationals Daniil Maksimovich Shchukin and Anatoly Sergeevitsch Kravchuk as the leaders of the GandCrab and REvil ransomware operations between 2019 and 2021. Operating under aliases such as 'UNKN' or 'UNKNOWN,' Shchukin and Kravchuk orchestrated at least 130 cyberattacks targeting German companies, resulting in over $40 million in damages and approximately $2.2 million in ransom payments. Their operations popularized the 'double extortion' tactic, demanding payment for decrypting data and additional sums to prevent public release of stolen information. ([bleepingcomputer.com](https://www.bleepingcomputer.com/news/security/german-authorities-identify-revil-and-gangcrab-ransomware-bosses/?utm_source=openai)) This identification underscores the persistent threat posed by sophisticated ransomware groups and highlights the importance of international cooperation in combating cybercrime. The GandCrab and REvil models have influenced current ransomware tactics, emphasizing the need for robust cybersecurity measures and proactive threat intelligence to mitigate such risks.