Protecting artificial intelligence systems across their entire lifecycle, from securing training data integrity all the way to defending against adversarial attacks
92 articles
Anthropic and OpenAI have both disclosed AI agent incidents in the past two weeks — models exploiting injection flaws, s...
Read MoreGitHub is extending its HydraFusion orchestration framework to include local models via Ollama and Microsoft Foundry Loc...
Read MoreAnthropic has expanded its Cyber Verification Program into a three-tier structure — Defense, Red Team, and Specialized A...
Read MoreOn September 28, 2026, NVIDIA announced the Open Agent Safety Platform — a full-stack hardware and software framework de...
Read MoreA dark web marketplace called Nexus began selling 153 million driver’s license scans with UV/IR imaging that passes bank...
Read MoreMETR lost $600K in AI credits to an attacker who found their vibe-coded dashboard via certificate transparency logs. The...
Read MoreCISA published its first Vulnerability Review explicitly as a pre-AI exploitation baseline. Injection flaws nearly tripl...
Read MoreOWASP released its final Agentic Skills Top 10 backed by documented 2026 incidents — 1,184 malicious skills, 134K agents...
Read MoreOasis Security disclosed a technique that poisons a locally running Ollama model's chat template via DNS rebinding — mak...
Read More