NeuralTrust Raises $20M to Secure Enterprise AI Agents at Scale

A Confident Bet on Defending the Agent Era

Good news for anyone deploying AI agents at work: the defenders are gearing up. On June 17, 2026, Barcelona-based NeuralTrust announced a $20M seed round — described as the largest cybersecurity seed financing by an EU company to date. The round was led by Alstin Capital, with VentureFriends, Seaya, Kibo Ventures, Banc Sabadell, and others, plus backing from the European Innovation Council.

That's the confirmed news. And it's a reassuring signal: as enterprises wire AI agents into real workflows, serious capital is flowing toward keeping those systems safe rather than leaving security as an afterthought.

The Problem, Explained Simply

Here's the situation in plain terms. An AI agent isn't just a chatbot — it's software that can take actions: calling tools, querying systems, and chaining requests together. When you have one agent, that's manageable. When you have hundreds, each making its own LLM, MCP, and tool calls, you need a way to see and govern all of it. Enterprise AI agent security is fundamentally about visibility and control at that scale.

NeuralTrust, founded in 2024, builds exactly that kind of defensive platform. Rather than treating agents as a black box, it gives security teams the tooling to watch, broker, and protect every interaction.

Three Layers of Defense

The platform centers on three products, and the design is refreshingly logical:

- TrustGate is an agent gateway that brokers every LLM, MCP, and tool call. Think of it as a single, well-guarded checkpoint that every request must pass through — no side doors.

- TrustGuard is a runtime engine that detects and stops threats as they happen, defending agents while they're actually working rather than only at design time.

- TrustLens handles posture management, inventorying every agent so teams always know what they're running. You can't protect what you can't see, and TrustLens is the seeing part.

Together they form a clean defense-in-depth model: inventory it, route it through a gateway, and watch it at runtime. That's enterprise AI agent security done methodically.

Real Customers, Real Confidence

This isn't theoretical. NeuralTrust already counts Iberia, Air Europa, Abanca, and Banc Sabadell among its customers — names from aviation and banking, two industries that take operational risk seriously. When regulated, high-stakes enterprises adopt a defensive platform early, it's a strong vote of confidence in the approach.

The Bigger Picture

Here's my analysis, clearly separated from the facts above: the AI-agent era is arriving fast, and the encouraging part of this story is that defensive infrastructure is maturing alongside the technology. A well-funded gateway-plus-runtime-plus-inventory model is precisely the kind of foundation enterprises need to adopt agents with confidence.

The takeaway is genuinely optimistic. As agents proliferate, you don't have to choose between innovation and safety. Companies like NeuralTrust are building the guardrails that let organizations move forward boldly — knowing every agent is accounted for, every call is brokered, and every runtime threat has someone watching the gate.

Sources: PR Newswire — "NeuralTrust raises $20M to secure the growing swarm of AI agents in the enterprise" — June 17, 2026; TNW — "NeuralTrust raises $20M to secure enterprise AI agents" — June 17, 2026.