A Birmingham, Michigan startup just raised $10 million to make sure the AI agents companies are rushing to deploy don't go off the rails. Ultra Security, operating out of an office at 333 Park St., emerged from stealth this week with a seed round aimed at giving businesses visibility into what their autonomous AI systems are actually doing behind the scenes.
The company, founded in 2025 by Chase Lee, the former founder and CEO of Trustpage, built what it calls an agentic security platform, according to VentureBeat. The $10 million seed round and the company's Birmingham headquarters were also detailed in Crain's Detroit Business, which reported on the launch from the Park Street office, pictured in a photo credited to CoStar Group Inc.
Ultra's pitch is narrow but increasingly urgent: as companies hand more autonomy to AI agents, those agents need to reach outside tools, databases and connectors to get work done, and that's exactly where things can go wrong. The platform sits between AI agents and the MCP servers and connectors they call, per VentureBeat's reporting. Every agent-to-tool interaction is logged, traced and made queryable, giving security teams a record of what an agent accessed and why.
Guardrails Built for a System Companies Don't Fully Trust Yet
The idea is that businesses shouldn't have to take an AI agent's word for it. Ultra offers guardrails, governance, observability and connectivity, and the platform runs locally by default, according to a report from Dealroom. That local-first design appears aimed at giving companies control over sensitive agent activity without routing everything through outside servers.
Ultra says it has already lined up a stack of compliance credentials ahead of its public launch, claiming SOC 2 Type II, ISO 27001 and ISO 42001 certification along with AARM Extended conformance, per VentureBeat's report. The company has also completed a FedRAMP 20x Class B audit with a third-party assessment organization and is in the process of pursuing full FedRAMP certification, according to a separate account from EIN Presswire.
Backing the round are Ludlow Ventures, Hidden Capital, Entrée Capital, Detroit Venture Partners, firedrop and FDVC, the venture outlet reported. The company intends to use the funds to expand operations and its development efforts, according to a report from FinSMEs.
A Crowded, Fast-Growing Corner of Cybersecurity
Ultra isn't alone in racing to capitalize on companies' anxiety over autonomous AI. The scramble to secure businesses against swarms of rogue artificial intelligence agents is fueling demand for new security tools industry-wide, according to CNBC. That outlet reported that Dallas-based Island, an enterprise browser security startup, recently raised $400 million at a $6.4 billion valuation, and that the AI security market now includes established players like Palo Alto Networks alongside a growing roster of startups competing for a slice of the business.
The money flowing into the sector overall tells a similarly aggressive story. Cybersecurity and privacy-related startups raised $10.6 billion globally across funding stages in the first half of 2026, per data from Crunchbase News, which also noted that second-quarter funding alone totaled $4.4 billion, a roughly 30% drop from the prior quarter and the year-ago period, even as the quarter produced eight rounds of $100 million or more, including a $600 million raise by Cyera at a $12 billion valuation.
Seed-stage investment specifically in AI-focused security startups is bucking that broader slowdown, with companies in that category raising $855 million across more than 150 reported seed rounds so far in 2026, putting the category on track for an all-time high, according to a separate Crunchbase analysis. Ultra's $10 million raise lands squarely in that surge, betting that as more Michigan and national businesses lean on AI agents to handle real work, someone needs to be watching what those agents touch.
