AIR's $50M Raise: A New Security Paradigm for AI Agents

Introduction: The Rise of AI Agents in Enterprise Systems

As companies increasingly delegate complex tasks to AI agents, a new frontier in cybersecurity has emerged—one that demands more than traditional endpoint protection. These autonomous tools, equipped with skills, plug-ins, and add-ons, are beginning to resemble operating systems, yet they lack the oversight mechanisms that protect us from malicious drivers or applications.

The Problem: AI Agent Supply Chains

When we think about software supply chains, we often picture traditional applications. But in the world of AI agents, this supply chain is rapidly forming around skills, plug-ins, MCP servers, and add-ons. These components allow agents to interact with the internet, access databases, and integrate with various enterprise systems. However, unlike conventional software that must be signed or verified, these tools are largely unregulated, leaving a gaping vulnerability.

"In the early 2000s, whenever you installed a driver, the driver didn't need to be signed. Today, every time you install a driver, you see a signature saying who signed it, because the driver is actually loading code into the kernel," said Yair Saban, CEO of AIR. "You don't have that with skills or plug-ins or MCPs, and it's a shame, because it's the same mechanism, it's the same lesson, but we haven't learned it."

Without these checks, AI agents can become vectors for cyberattacks, especially when attackers poison the content they consume instead of targeting the agents directly.

AIR's Solution: Visibility and Continuous Vetting

Founded by Yair Saban and Niv Hoffman—veterans of Israel's Unit 8200 intelligence corps—AIR aims to plug this gap. Their platform offers a three-pronged approach to agent security:

• Discovery: Identifies agents running inside companies, including those operating outside IT department approvals.
• Vetting: Continuously analyzes skills and add-ons used by agents against a vetted whitelist.
• Enforcement: Blocks unauthorized actions and interactions in real time.

AIR's platform filters out approximately 27% of the add-ons and skills it finds online, indicating its rigorous approach to risk management. This is not just about scanning—it's a continuous re-verification process that adapts as new threats emerge.

Market Context: Competition in AI Security

AIR is entering a crowded field. Competitors like Noma Security, Zenity, Astrix Security, and Operant AI are also developing tools to secure AI agents. In fact, the venture capital interest in this space has been substantial, with Zenity raising $125 million and Noma securing $100 million.

Saban emphasizes that while discovery and monitoring are easy to replicate, AIR's competitive advantage lies in its ability to continuously vet tools and add-ons—a process that is inherently difficult and resource-intensive. "This is not a scanning problem, it is a continuous re-verification problem," Saban said. "Inspecting every skill, plugin, MCP server and sub-agent an enterprise's agents touch, re-inspecting each one every time it changes, in real time and across an entire company's agent fleet, is an infrastructure problem long before it is a security problem."

Investment and Growth Strategy

AIR's $50 million funding round—$10 million led by Sequoia and $40 million led by Greenoaks—reflects strong investor confidence in the company's vision. The capital will be primarily used for hiring researchers and expanding sales efforts in the U.S. and Europe.

The startup currently has more than 20 customers, with a quarter of them being large enterprises. The demand is strongest in heavily regulated sectors like financial services and pharmaceuticals, where compliance and data integrity are paramount.

Leadership Legacy: From Intelligence to AI Security

Saban and Hoffman's backgrounds in cybersecurity offer a unique perspective on AI risk. Having worked on offensive security measures within Israel's military intelligence unit, they bring deep expertise in threat modeling and real-time response systems—skills directly applicable to protecting AI ecosystems.

Their approach reflects a broader trend: leadership from the defense sector is increasingly entering the commercial cybersecurity space, especially as AI agents begin to take on enterprise-critical roles. AIR's platform stands as a testament to this evolution, combining military-grade vigilance with practical business applications.

Looking Ahead: The Future of AI Agent Governance

As AI agents become more autonomous and integrated into workflows, the need for governance frameworks becomes critical. AIR's solution isn't just a tool—it's a new category of security infrastructure that must evolve with AI's growing complexity.

Ultimately, the goal is not to prevent AI agents from becoming powerful tools but to ensure they remain safe, controllable, and compliant. AIR's platform offers one path forward—one that recognizes the urgency of securing AI supply chains before they become the next frontier in cyber warfare.

Conclusion: A Security Infrastructure for the AI Age

The rise of AI agents is transforming enterprise operations. But with this transformation comes a new set of risks, particularly in how these agents interact with external resources and internal systems. AIR's approach to monitoring, vetting, and enforcing agent behavior offers a compelling solution to an urgent problem.

With significant funding behind them and a clear vision for the future, AIR is positioning itself at the forefront of AI security—a critical development in our journey toward a more intelligent but also more secure digital world.