NeuralTrust launches runtime security product for AI agents
NeuralTrust said its new runtime security product is designed to inspect AI agent traffic directly and work across different agents and clouds through its Agent Gateway.
NeuralTrust said Tuesday that it launched a runtime security product for AI agents, designed to extend across every agent without custom integration for each one.
The company said the offering is built on its Agent Gateway, which it described as a gateway with native, built-in security. NeuralTrust said the system inspects agentic traffic directly, regardless of which agent or cloud produced it, and tracks requests and actions through a full loop.
In the release, NeuralTrust said alerts, audit logs and analytics run across all six stages in real time. It also said the system can integrate with a company’s SIEM, or security information and event management system, in one click.
Joan Vendrell, NeuralTrust’s co-founder and CEO, said earlier agent security products were point solutions built for one coding assistant, framework or vendor. He said NeuralTrust’s runtime security plugs in once through the gateway and covers every agent that runs through it.
The company also said it had been named a Representative Vendor in two Gartner Market Guides and three Gartner Hype Cycles, and a Leader in KuppingerCole’s 2025 Leadership Compass for Generative AI Defense.
NeuralTrust said the platform is intended to help enterprises identify, secure and govern AI agents. It also said clients include global banks, airlines, energy companies and government agencies.
The release was datelined Las Vegas and dated Aug. 5, 2026. NeuralTrust said it is exhibiting at Black Hat USA 2026, booth 8106, through Aug. 6, and said technical documentation on TrustGate and TrustGuard is available at docs.neuraltrust.ai.
Sources
Editorial note: This article was prepared with AI assistance and includes source attribution for reader review.
Comments ()