Cybersecurity & Exploits

Global: Comprehensive Survey Maps Challenges in Fuzzing Open-Source Operating SystemsResearchers from a multinational team have released a detailed survey that examines fuzzing techniques applied to open-source operating systems. The work,…

Global: Study Examines Security Risks of Machine Learning Workloads on Serverless Platforms A new research paper released on arXiv on January 2026 provides a comprehensive security analysis of machine learning…

Global: Timing Variability in Macro-Fused Conditional Jumps Across Modern CPUs Researchers have measured significant timing variability in conditional jump instructions that can be macro‑fused with a preceding operation, revealing that…

Global: Study Finds Widespread Security Gaps in Model-Sharing PlatformsStudy OverviewA recent large‑scale empirical investigation evaluated custom model loading practices across five leading model‑sharing platforms, uncovering significant security concerns tied to…

Global: Optimization Model Targets Hospital Cyber Resilience in GermanyA new defender‑attacker‑defender optimization model is designed to help hospital networks improve their cyber resilience, according to a recent preprint posted on…

Global: Commercial LLM Agents Often Bypass Safety Checks in User-Mediated Attack ScenariosIn January 2026, researchers released a study examining how twelve commercially available large‑language‑model (LLM) agents respond when unwitting users…

Global: New Benchmark Evaluates Indirect Prompt Injection Risks in Retrieval-Augmented Generation SystemsOn January 16, 2026, researchers Haoze Guo and Ziqi Wei submitted a paper to arXiv introducing OpenRAG‑Soc, a benchmark…

Global: New Multi‑Turn Economic Denial‑of‑Service Attack Targets LLM Agent‑Tool InterfacesResearchers have disclosed a novel denial‑of‑service (DoS) technique that exploits the interaction loop between large language model (LLM) agents and external…

Global: New LLM Framework Aims to Enhance Security Operations Center ReportingResearchers from an international team have introduced a large‑language‑model (LLM) based system called MESSALA to improve the creation and evaluation…

Global: Researchers unveil AgentGuardian framework to enforce access control for AI agentsFramework OverviewA new security framework designed to govern AI agent behavior has been introduced by a team of researchers…