Ultra Security, a Birmingham, Michigan company founded in 2025, emerged from stealth on October 21, 2026, announcing a $10 million seed round. The funding supports its platform that provides visibility, guardrails, and governance for autonomous AI agents operating inside enterprise systems.
The round drew participation from Ludlow Ventures, Hidden Capital, Entrée Capital, Detroit Venture Partners, firedrop, and FDVC, along with angels from security and SaaS backgrounds. No single lead investor was identified in the announcement.
Platform Capabilities for Enterprise AI Agents
The Ultra Security platform sits between AI agents and Model Context Protocol servers. It intercepts every tool call, data read, and prompt request to enforce policies before execution. Users define rules in plain language that block credential exposure, redact PII and PHI, impose rate limits, and prevent unauthorized data flows across connectors.
Additional controls let administrators manage access by user or team and review logs with AI-assisted anomaly detection. The system supports any model or vendor through a universal connectivity layer. Deployment defaults to local execution on-device, though a remote-hosted option is available for organizations that require it.
Certifications already in place include SOC 2 Type II, ISO 27001, and ISO 42001. The platform also meets AARM Extended requirements and aligns with the NIST AI Risk Management Framework. A FedRAMP 20x Class B audit has been completed, with full certification in progress. Pricing includes a free tier, usage-based plans, and enterprise contracts.
Founding Background and Compliance Standards
Chase Lee, who previously founded Trustpage and later served as VP of New Products at Vanta after its acquisition, leads the company. The founding team includes alumni from Vanta and Secureframe. The group spent roughly one year in stealth developing the product with design partners before the October 21 announcement.
Traditional security tools such as web application firewalls and API gateways lack visibility into live AI agent sessions. Ultra Security addresses this gap by operating at the MCP layer where agents gain direct access to email, code repositories, and internal data stores. The local-first architecture and existing compliance attestations reduce deployment friction for security and compliance teams evaluating the platform.
Market Context and Next Steps
Interest in AI agent security has grown rapidly, with peer companies attracting hundreds of millions of dollars in funding over the past five months. Ultra Security positions its offering as a practical layer that enterprises can adopt without replacing existing model providers or connectors.
Access requests are open at ultra.security. The company plans to use the seed capital to expand engineering and customer success teams while broadening support for additional connectors and policy templates. Organizations evaluating the platform can begin with the free tier to test policy enforcement before scaling to paid usage.
Frequently asked questions
What does Ultra Security's platform do?
It intercepts tool calls and data requests from AI agents in real time, applying policy guardrails and logging activity across any model or connector.
When did Ultra Security raise its seed round?
The $10 million round closed on October 21, 2026, coinciding with the company's emergence from stealth.
Who founded Ultra Security?
Chase Lee, former founder of Trustpage and VP at Vanta, leads the company with a team that includes alumni from Vanta and Secureframe.
Where is Ultra Security based?
The company is headquartered in Birmingham, Michigan, near Detroit.
By YasTech Editorial
Comments 5