Cybersecurity & Exploits

Global: Study Shows Feasibility of Eclipse Attack on Ethereum 2.0 NodesResearchers have demonstrated the first end‑to‑end eclipse attack against Ethereum 2.0 execution‑layer nodes, showing that a malicious actor can fully…

Global: ProveRAG Demonstrates High Accuracy in LLM-Assisted Vulnerability Analysis Rising Volume of Vulnerabilities Challenges Analysts A recent study published on arXiv introduces ProveRAG, an LLM-powered system designed to assist cybersecurity…

Global: Static Detection of Core Structures in Tigress Virtualization-Based Obfuscation Using an LLVM PassResearchers Sangjun An, Seoksu Lee, and Eun‑Sun Cho announced on arXiv that they have developed a static…

Global: Study Proposes Cognitive Control Architecture to Counter Indirect Prompt Injection in Autonomous LLM Agents A recent arXiv preprint details a novel defense framework aimed at protecting autonomous large language…

Global: Multi-Targeted Backdoor Attacks Demonstrated on Graph Neural Networks Researchers have presented the first multi-targeted backdoor attack designed for graph classification tasks, enabling multiple distinct triggers to simultaneously redirect model…

Global: Security Risks Identified in Approximate Caching for Diffusion ModelsResearchers have published a comprehensive assessment indicating that approximate caching—a technique employed to reduce the computational load of diffusion‑model services—creates multiple…

Global: Researchers Introduce PhishLumos Framework for Proactive Phishing Campaign MitigationA new adaptive multi‑agent framework aims to shift phishing defense from reactive URL blocking to proactive campaign disruption. The system, named…

Global: Efficient Database-Driven Adversarial Prompting Reduces LLM Jailbreak CostsA team of AI security researchers announced a new adversarial prompting technique on arXiv in January 2026, aiming to lower the computational…

Global: Researchers Demonstrate Wireless Man-in-the-Middle Attacks on EV Charging ProtocolsA team of researchers published a study in January 2026 describing how they used a real‑time software‑defined radio (SDR) to intercept…

Global: Physical Adversarial RF Perturbations Reduce Drone Detector AccuracyResearchers Omer Gazit, Yael Itzhakev, Yuval Elovici, and Asaf Shabtai reported a new over‑the‑air (OTA) attack that lowers the detection rate of…