Cybersecurity researchers at Mindguard have disclosed a prompt injection vulnerability in Amazon Kiro, an AI-powered agentic integrated development environment (IDE), that could…
Aikido Security has published research recreating an Australian gym-booking incident in a synthetic environment, finding that Claude Opus 4.6, running on the…
Oasis Security has disclosed a vulnerability in NVIDIA's NemoClaw, an open-source reference stack for running AI agents, that could allow an attacker-controlled…
The U.S. government has issued a joint advisory warning of an active threat targeting critical infrastructure organizations using AI-generated exploit scripts. The…
OpenAI has announced a temporary pause in reinforcement learning (RL) training for its most advanced AI models to strengthen safety and security…
Researchers at Anthropic and Switzerland's EPFL have demonstrated that self-propagating payloads, dubbed "mind viruses," can spread between AI agents through editable system…
AgentWorm is a self-replicating worm targeting OpenClaw, reported with a 63% aggregate attack success rate across five model backends. Originally named ClawWorm,…
A new attack named MemGhost demonstrates how a single email can inject persistent false memories into AI personal assistants, manipulating their future…
Security researcher Chinmohan Nayak has detailed a WhatsApp-to-host attack chain leveraging three now-patched vulnerabilities in the OpenClaw personal AI assistant. The flaws,…
Cyera previously disclosed the Claw Chain vulnerabilities in OpenClaw, which required prior foothold for exploitation.