Mintive Agent Identity Hits 2.5M Daily Calls as Security Grows

Mintive agent identity is moving from a niche security concern to core enterprise infrastructure. According to a company announcement reported by Business Wire on October 29, 2026, Mintive now powers 2.5 million agent calls every day across more than 50 paying enterprise customers. The company has covered more than 200 million calls in just seven months, a pace that shows how quickly AI agents have moved from experiments into production systems.

The announcement also includes new funding. Mintive raised 7 million dollars in a round led by Cowboy Ventures, bringing its total funding to 11 million dollars. Investors in the round include Coatue, Operator Collective, Burst Capital, Maven Ventures and Hustle Fund, alongside Okta co-founder Frederic Kerrest and WorkOS founder Michael Grinich. According to the announcement, the money will help Mintive meet demand from enterprises that need a trusted way to govern agent access.

Why Mintive agent identity matters now

The problem Mintive is solving is easy to state and hard to fix. Enterprises are deploying agents to scale their businesses, yet many of those agents arrive in production with broad access and no clear record of which agent took which action. That gap worries security and information technology teams, because an agent that can read internal tools, move data and trigger workflows needs the same discipline as an employee with system access.

According to Gartner figures cited in the announcement, 40 percent of enterprise applications will feature task-specific AI agents by the end of 2026, up from less than 5 percent a year earlier. By 2028, the announcement notes, the average global Fortune 500 enterprise could have more than 150,000 agents, compared with fewer than 15 in 2025. Whether those forecasts prove exact, the direction is clear: the number of agents inside a company is about to explode, and identity management is the control plane that keeps that growth safe.

Mintive treats each agent the way an enterprise has long treated an employee, with its own identity and only the access its work requires. Without that approach, each new agent platform brings another permissions model, another integration path and another set of controls for security teams to manage. Mintive aims to replace that sprawl with one layer of rules that applies across agents, models and tools.

How the Mintive platform governs agent access

According to the company, the Mintive platform combines several pieces that usually live in separate tools. Its MCP Gateway governs tool access, while its LLM Gateway governs AI model calls, so one set of rules applies whether a team uses Claude, ChatGPT, Copilot, Cursor or an internally built agent. Mint Guard screens calls for prompt injection, exposed secrets, personal data and other risks before they reach enterprise systems.

Traceability is a central part of the pitch. Each tool call can be traced back to the person who initiated it, the agent that acted, the server it reached and the model involved. That audit trail matters when something goes wrong, and it matters just as much when auditors or compliance teams ask who did what. The platform is SOC 2 Type II audited, according to the announcement.

The founders bring deep platform experience to the problem. Mintive was founded by Jiquan Ngiam, a former senior staff software engineer at Google Brain and a founding engineer at Coursera, and Vijay Vasudevan, who co-created TensorFlow at Google. According to the announcement, they started the company after seeing the first enterprise agents enter production with broad access and no clear record of their actions.

Real deployments show the model working

Customer examples in the announcement show Mintive agent identity in everyday use. At Modern Treasury, a bank operations agent built with Mintive Coworker agents searches across Sentry, Notion, Linear, the codebase and internal tools before a technical account manager steps in. According to the company, that workflow cuts 30 minutes from simpler cases and up to four hours from complex ones.

At Ro, Mintive operates in a healthcare environment where agent access and auditability carry security and compliance requirements alongside the work itself. At the University of Michigan, Mintive serves as an enterprise Model Context Protocol gateway, giving students, faculty and staff a common path to approved tools while central IT maintains one layer of access and control. Customers also include Coursera and Apax, according to the announcement.

Security leaders quoted in the announcement frame the choice as practical rather than ideological. Christian Burrows, Head of Security at Stability AI, said that simply blocking employee access to AI agents is not an effective strategy, because people will find ways around it. The better approach, in his view, is to make saying yes the easy path, giving employees freedom to use tools while security teams keep visibility and control over what agents can access and do.

Aileen Lee, founder and managing partner at Cowboy Ventures, said governance, oversight, identity management and security are the major gating factors in enterprise AI adoption. Mintive, in her view, offers the scalable, trusted third-party scaffolding businesses need to deliver agent identity, permissions and tracing regardless of the underlying models.

The broader context supports that reading. Recent coverage of the agent ecosystem has focused on alignment and evaluation, as seen in reports on tools that align enterprise agents with human judgment, while security teams are separately racing to govern what agents can do once they are inside the network. Mintive sits at the intersection of those two trends: it does not just watch agents, it gives each one an identity, a permission set and a traceable history.

For enterprises, the takeaway is straightforward. Agents are already working inside company systems, and each one needs its own identity and permissions, just as every employee does. Mintive co-founder and chief executive Jiquan Ngiam put it plainly in the announcement: with the right permissions and controls, agents can take on more responsibility and move faster, without a rogue agent moving out of a sandbox to compromise the enterprise. More detail is available from the Business Wire announcement and from Mintive.