CrowdStrike (CRWD) launches AI Cyber Superintelligence Lab and new defense tools

CrowdStrike Holdings (NASDAQ: CRWD) announced the launch of its Cyber Superintelligence Lab, described as the industry's first AI research organization focused on cyber defense and safety.

The company introduced SafeMind AI security models, designed as agentic, autonomous systems to address emerging AI-driven cyber risks. It also unveiled Falcon Guardian for AI runtime detection, aimed at protecting enterprise AI agents as they operate in real time.

The initiatives are being developed with direct collaboration from NVIDIA, combining offensive and defensive AI approaches to help stop breaches and manage AI security risks. According to the source, SafeMind is anchored on NVIDIA Nemotron models, while Guardian is being embedded into hyperscalers such as Google Cloud.

CrowdStrike is a US-based cybersecurity company that provides software to help businesses detect and respond to digital threats across their networks and endpoints. The new tools broaden its reach from endpoint security into AI agent protection, AI governance and frontier AI risk services.