# NeuralTrust raised $20M to police the AI agents companies can’t even count > Source: > Published: 2026-06-18 11:56:33+00:00 NeuralTrust, a Barcelona startup, has raised $20mn to secure the AI agents that big companies are now deploying faster than they can keep track of. The seed round, worth €17.2mn, was led by Munich’s Alstin Capital, with VentureFriends, Seaya, Kibo Ventures, Banc Sabadell, the EA Ventures Plug and Play Fund and Finaves, plus public money from the European Innovation Council and Spain’s research agency. NeuralTrust calls it the largest cybersecurity seed round ever raised by an EU company, a claim worth a pinch of salt, but not an outlandish one. ## The problem: agents nobody is counting The problem NeuralTrust sells against is sprawl. Inside a large company, AI agents multiply without anyone keeping a tally. One team builds a customer-service agent on one model, another automates back-office work on a second, and vendor software arrives with agents of its own. Each one connects to internal systems, databases and outside tools, and acts on its own. Most security teams cannot say how many are running, [what each is allowed to do](https://thenextweb.com/news/arcade-dev-60-million-series-a-ai-agent-authorization), or whether one has been tricked into leaking data. “If you connect AI to your email system and it sends emails to outside addresses, leaking internal information, that’s a disaster,” chief executive Joan Vendrell told Tech Funding News. NeuralTrust’s answer is a single control layer, sold as three products: TrustGate, a gateway that every model, tool and agent call passes through; TrustGuard, a runtime engine meant to catch and stop attacks live; and TrustLens, which maps every agent and watches what it does. The company says it inspects millions of agent interactions a day, and that roughly 1.2 per cent, about one in 80, are malicious. ## The pitch underneath the pitch: ‘not American’ The more interesting part of the story is geography. NeuralTrust’s main rivals are American, and it is leaning hard into not being. “There are real issues now regarding technological and defence sovereignty in the EU,” Vendrell said. “Some customers, especially governments, are very sensitive about where the technology they buy comes from.” It is a well-timed argument. With the EU AI Act tightening and Washington’s recent move to cut foreign access to its most capable models, “made in Europe” has become a selling point in its own right, and European institutions have already started [swapping US security vendors for home-grown ones](https://thenextweb.com/news/frances-intelligence-service-is-dropping-palantir-for-a-homegrown-rival). ## A category, if it arrives before the incident Behind the round is a bet that securing AI agents becomes its own category, the way endpoint and cloud security did before it. There is evidence for that: Palo Alto Networks bought the startup Protect AI for a reported $500mn in 2025, a sign the giants now treat agent security as core, not a bolt-on. Gartner, which NeuralTrust quotes liberally, predicts 40 per cent of enterprises will pull back autonomous agents by 2027 over governance gaps spotted only after something breaks. The open question, as NeuralTrust’s own lead investor concedes, is timing: whether companies buy this before a serious agent-related incident forces them to. NeuralTrust is betting they will, and that when they do, plenty of them will want a European name on the invoice. ## Get the TNW newsletter Get the most important tech news in your inbox each week.