AIR raises $50M to build a firewall for AI agents before they go rogue

As companies race to deploy AI agents that can browse the web, execute code, and make decisions autonomously, a new Israeli startup is betting $50 million that nobody is watching what those agents actually do once they’re running. AIR Security emerged from stealth on September 1 with a $50M funding round led by Sequoia Capital and Greenoaks, with additional backing from Swish Ventures, Netz, and a roster of angel investors from the cybersecurity and AI worlds. The company’s pitch: a real-time firewall purpose-built for AI agents, capable of discovering which agents are running inside a company, continuously vetting their skills and plugins, and blocking anything that looks like trouble. The problem AIR is trying to solve AIR’s own research paints a fairly alarming picture. The company found over 17,800 public AI add-ons linked to untrusted external sources, collectively accounting for roughly 6.7 million installations. Even more concerning: AIR identified cases of brand impersonation where AI skills masqueraded as products from well-known companies like Anthropic and OpenAI. The goal was to slip past security reviews and potentially execute arbitrary code. The platform works as an ...