CrowdStrike Unveils SafeMind Agent System
Article excerpt
CrowdStrike unveils SafeMind system with Nvidia. U.S. cybersecurity firm CrowdStrike has introduced the SafeMind system, developed in collaboration with Nvidia. It integrates offensive and defensive AI models in a closed loop, where one seeks attack methods and the other creates and tests detection rules. The announcement was made on September 1 at the Fal.Con 2026 conference in Las Vegas. SafeMind will be natively integrated into the CrowdStrike Falcon platform. Initially, the system includes two specialized models: The models operate with agentic wrappers - software frameworks that provide AI with context, tools, and rules for executing multi-step tasks. In Nvidia’s description, the system is compared to a continuous red and blue team confrontation. Red Tempest conducts attacks in an isolated environment, Falcon captures telemetry, and the defensive framework identifies gaps and creates new detection rules. The attack is then repeated with the updated defenses. The SafeMind defensive framework is built on Nvidia Nemotron models with open weights, further trained on CrowdStrike’s cyber data. Nemotron 3 Ultra orchestrates defensive agents: reconstructing attack sequences, planning actions, and invoking necessary tools. The fine-tuned Nemotron 3 Super is used as a specialized model for writing and correcting detection rules. To configure it, CrowdStrike employed additional...
Keep reading with a free account
The rest of this article, and every signal for CrowdStrike, is in your free account.
Extracted from this sentence
CrowdStrike also introduced the Cyber Superintelligence Lab, a research division bringing together the company’s AI, offensive security, and incident response experts.
