Emerging from stealth, AIR Security has secured $50 million to launch a specialized firewall designed to protect AI agents from malicious plugins, unauthorized instructions, and supply chain risks.
- AIR Security raised $50 million in funding led by Sequoia Capital and Greenoaks.
- The startup provides a specialized 'AI Firewall' to protect AI agents and their ecosystem.
- Research revealed 17,800+ public AI add-ons rely on untrusted external instruction sources.
- The platform monitors AI skills, plugins, and MCP servers for malicious behaviors.
As artificial intelligence transitions from a novelty to a core component of enterprise operations, a new frontier of cybersecurity has emerged. AIR Security has officially emerged from stealth mode, announcing a massive $50 million funding round to address the vulnerabilities inherent in the growing AI agent ecosystem.
The funding, led by heavyweights Sequoia Capital and Greenoaks, comes at a critical juncture. As AI agents gain the autonomy to browse websites, access sensitive files, and execute tasks on behalf of employees, they become prime targets for adversaries using poisoned content or direct compromises to facilitate data theft and fraud.
The Growing Threat Landscape
AIR Security’s foundational research highlights a staggering vulnerability in the current market. The firm discovered that over 17,800 public AI add-ons—representing approximately 6.7 million installations—rely on untrusted external instruction sources. Furthermore, malicious 'AI Skills' have been identified impersonating industry leaders like Anthropic and OpenAI to bypass security reviews and execute arbitrary code.
"Every enterprise has a firewall protecting its network. Now they need one protecting their AI agents," says Yair Saban, co-founder and CEO of AIR.
The rise of coding agents such as Claude Code, Cursor, and Codex has accelerated this risk. While these tools boost productivity, they operate with a level of autonomy that traditional security measures are not equipped to monitor, creating a significant gap in organizational visibility.
Why This Matters
BozokMedia analysis shows that the shift toward 'Agentic AI' requires a paradigm shift in cybersecurity. Traditional firewalls protect networks, but AIR Security’s firewall protects the context. By evaluating every skill, plugin, and MCP server across an organization’s AI supply chain, the platform ensures that malicious updates or compromised integrations are automatically revoked.
| Feature | Traditional Firewall | AIR Security Firewall |
|---|---|---|
| Primary Focus | Network Traffic/IPs | AI Agent Context/Instructions |
| Threat Target | Malware/Hackers | Malicious Plugins/Poisoned Prompts |
| Monitoring Scope | Data Packets | AI Skills, MCP Servers, & Add-ons |
Frequently Asked Questions
1. What exactly is an AI Agent Firewall?
It is a security layer that evaluates the safety of AI skills, plugins, and instructions before they can interact with an enterprise's sensitive data.
2. Who is behind AIR Security?
The company was co-founded by CEO Yair Saban and CTO Niv Hoffman, with strategic leadership from former Disney and Costco CISO Ryan Knisley.
