AI security startup AIR emerges from stealth, raises $50 million, and launches ScanAir to monitor and control enterprise AI agent skills and add-ons

Deal terms as reported in the coverage below.

Decision brief

What changed

Three AI security and governance startups announced new funding to expand enterprise products for AI oversight: AIR disclosed $50 million across two seed rounds, HiddenLayer raised a $100 million Series B, and AI Score raised a $5.4 million seed round. Based on the coverage, all three plan to use the capital to broaden product capabilities and go-to-market efforts focused on securing or governing enterprise AI systems, including agents and their tool use.

Why it matters

This signals that a distinct vendor category around AI agent security, runtime protection, and governance is maturing quickly enough to attract sizable capital, which gives enterprise buyers more options for controls beyond core model providers. For decision-makers deploying AI agents, the coverage suggests vendors are targeting practical risk points such as plugin vetting, prompt injection, agent manipulation, malicious tool use, and governance controls, which can affect procurement, architecture, and operating policies. The funding also implies these vendors may scale sales capacity and product depth soon, potentially accelerating competitive pressure to formalize enterprise AI control frameworks.

Affected roles

CEO COO CTO CISO

Evidence

The event is supported by three separate outlet reports: two from TechCrunch AI covering AIR and HiddenLayer, and one from Tech.eu covering AI Score. Across the reports, the funding amounts and intended uses are consistent in describing product expansion and go-to-market growth for enterprise AI security or governance platforms focused on agents, models, workflows, and tool usage.

What remains uncertain

The coverage does not verify product effectiveness, customer adoption, revenue traction, deployment scale, or whether these platforms integrate cleanly with existing enterprise AI stacks. It also does not establish whether the reported demand reflects broad enterprise standardization on third-party AI security tools versus early market experimentation, so any procurement urgency beyond the announced raises remains an assumption.

Monitor next

Watch for disclosed enterprise customer wins, integrations with major AI platforms, or published efficacy results showing how these vendors prevent prompt injection, agent misuse, or risky tool interactions in production environments.

Analytical support, not advice — assumptions and open questions stated above.

Source coverage