Cybersecurity & Exploits

Global: Researchers Demonstrate Physical-World Distance-Pulling Attacks on Autonomous Tracking Drones New Attack Vector Identified Researchers have introduced a novel class of threats known as distance-pulling attacks (DPAs) that target autonomous…

Global: Malicious Image Patches Exploit OS AgentsResearchers have uncovered a new attack vector that targets operating‑system (OS) agents powered by vision‑language models (VLMs). The study, posted to arXiv in March…

Global: CNN‑Based IoT Device Identification Study Highlights Trade‑offs in Accuracy and SpeedResearchers at Aalto University have evaluated two deep‑learning approaches for identifying Internet‑of‑Things (IoT) devices, releasing findings in a study…

Global: Study Highlights Gaps in Industrial Control System Threat Information SharingResearchers examining cyber threats to critical infrastructure reported in December 2025 that current mechanisms for sharing threat information among private…

Global: Study Maps Security Risks of Urban UAVs and AI-Driven Detection StrategiesA recent arXiv preprint synthesizes a decade of literature on unmanned aerial vehicle (UAV) deployments in smart cities, highlighting…

Global: Study Maps Threat Landscape for MLOps PipelinesA new study posted on arXiv provides a systematic assessment of adversarial threats targeting Machine Learning Operations (MLOps) ecosystems, highlighting risks that span…

Global: HASTE Framework Boosts Prompt Injection Detection for Large Language ModelsResearchers have introduced HASTE (Hard-negative Attack Sample Training Engine), a systematic framework designed to improve the detection of prompt injection…

Global: Study Identifies Indirect Targeted Poisoning Threats in Chain-of-Thought Language ModelsA team of researchers announced a new class of attacks on large language models that employ chain-of-thought (CoT) reasoning, describing…

Global: Researchers Identify Four Bypass Techniques Against CHERI Compartmentalisation on Linux and BSDIn a recent arXiv preprint, a team of security researchers detailed four distinct methods for bypassing the compartmentalisation…

Global: New Multi-Agent Framework Aims to Mitigate LLM Resource Exhaustion AttacksA team of researchers led by Nirhoshan Sivaroopan and including Kanchana Thilakarathna, Albert Zomaya, Manu, Yi Guo, Jo Plested, Tim…