Section
Security
Safety, cyber risk, misuse, and defensive systems around AI.
Related Coverage
- Microsoft Unveils MDASH Agentic System That Found 16 Windows Vulnerabilities
On May 12, 2026, Microsoft Security announced codename MDASH—a multi-model agentic scanning harness of 100+ specialized agents that discover, validate, and help remediate software bugs—after using it to find 16 Windows vulnerabilities (including four critical RCEs) shipped in the same day’s Patch Tuesday, while topping industry CyberGym-style benchmarks in limited private preview.
- Anthropic: DeepSeek, Moonshot, and MiniMax Ran Industrial-Scale Claude Distillation
On February 23, 2026, Anthropic published “Detecting and preventing distillation attacks,” alleging three Chinese AI labs—DeepSeek, Moonshot, and MiniMax—used roughly 24,000 fraudulent accounts to generate more than 16 million Claude exchanges for illicit capability extraction. The company framed the campaigns as ToS and regional-access violations and detailed detection and defense investments against industrial-scale distillation.