NeuralTrust Raises $20M Seed to Secure Enterprise AI Agents

The problem with AI agent security is that it doesn't map cleanly to existing tooling. A network firewall doesn't know whether an agent is being manipulated into leaking secrets through a clever multiturn conversation; a traditional SIEM wasn't built to track whether a tool-calling workflow just had its controls bypassed. NeuralTrust's $20M seed raise, as reported by tech.eu, is essentially a bet that enterprises moving AI agents into production have created a category-defining gap in their security posture.

The company, founded by CEO Joan Vendrell, CTO Victor Garcia, and COO Alejandro Domingo, offers three linked products: TrustGate for managing agent traffic across models and tools, TrustGuard for runtime security, and TrustLens for visibility into deployment and behavior. The platform reportedly inspects millions of AI agent interactions daily, with approximately 1.2% requiring intervention for issues including attempted data extraction, tool manipulation, and control bypassing. That figure is small enough to seem routine and large enough to matter at enterprise scale.

The round was led by Alstin Capital, with VentureFriends, Seaya, Kibo Ventures, Banc Sabadell, EA Ventures Plug and Play Fund, and Finaves participating. The European Innovation Council and Spain's State Research Agency also provide support, which tends to open doors in regulated industries and government procurement. Vendrell's stated aim, according to the report, is to "turn AI security into a strategic advantage for the enterprises that will define the next decade."

NeuralTrust has also contributed to the broader threat taxonomy: the company identified "Echo Chamber," a multiturn jailbreak technique, and "Semantic Chaining," a multimodal attack vector, both now incorporated into the OWASP AI Security Project taxonomy. Contributions like these position a company as the authoritative voice in an emerging threat category, which matters for credibility and standards-driven enterprise procurement. The honest caveat is that the article doesn't disclose revenue, customer count, or deal size, so commercial traction is hard to assess externally. What the reporting also doesn't address is how TrustGate and TrustGuard integrate with existing enterprise security stacks -- whether this sits alongside SIEMs as an overlay or eventually displaces them is a genuinely open question for buyers.

The clearest near-term beneficiaries are enterprises in regulated sectors, particularly in the EU, where AI Act compliance requirements create immediate demand for agent governance documentation and monitoring. The European Innovation Council's involvement is not incidental to that positioning.