crowdstrike

Nvidia and CrowdStrike have launched SafeMind, a new family of agentic AI models designed to identify and remediate cybersecurity attack paths for customers.

NVIDIA and CrowdStrike announced CrowdStrike SafeMind, an agentic cybersecurity system developed using NVIDIA Nemotron models, at the Fal.Con 2026 conference to enhance AI-powered defense against automated attacks.

CrowdStrike has discovered a worm that targets AI software supply chains, stealing credentials and performing destructive actions while evading detection by mimicking legitimate AI coding activities.

CrowdStrike's 2026 Global Threat Report declares that prompts are the new malware, highlighting how prompt injection attacks allow hackers to exploit AI systems using plain language instead of code, dramatically lowering the barrier to entry for cyberattacks.