Nvidia and CrowdStrike are significantly expanding their collaboration to advance AI in cybersecurity, focusing on the development and deployment of autonomous AI agents. This partnership aims to enhance real-time threat detection and response, providing machine-speed defense for critical infrastructure across various environments, including cloud, data centers, and edge locations. The expanded partnership was announced in late October 2025.
The collaboration leverages CrowdStrike's Charlotte AI AgentWorks, a no-code platform allowing security teams to build and deploy AI agents using plain-language prompts. This platform integrates with NVIDIA's AI technologies, including Nemotron open models, NeMo Data Designer for synthetic data generation, the NeMo Agent Toolkit, and NIM Microservices for AI model deployment. Customers can fine-tune Nemotron models with CrowdStrike data using NeMo Data Designer for their specific AI agents.
A primary focus of this joint effort is to enhance the defense of critical infrastructure by deploying continuously learning autonomous AI agents closer to where data is generated, improving local inference and enabling faster detection and response. This strategy extends protection to data centers and controlled environments, with CrowdStrike's Falcon platform providing extended protection and governance for AI agents developed with NVIDIA NeMo tools and third-party ecosystems. The companies are working to build an "agentic ecosystem" for secure development and scaling of these agents.
The cybersecurity landscape is rapidly evolving with the rise of autonomous AI agents, transforming security from manual processes to proactive, real-time defense. This shift has led to an "agentic AI arms race" where both defenders and attackers utilize AI. Notable developments include OpenAI's Aardvark, an autonomous agent for scanning and patching software vulnerabilities, and Google's CodeMender for detecting and rewriting vulnerable code. Researchers have also identified "agent-aware cloaking" as a new attack vector exploiting how AI search tools retrieve web content. The general trend for 2026 anticipates AI agents moving towards fully autonomous systems capable of complex reasoning and task execution across various industries, including cybersecurity.
[link] [comments]