idGenius launches Matrix to secure AI agents with relationship-based identity controls

idGenius announced Matrix v2, an AI agent identity security platform built to connect agents to owners, projects, tools and policies in real time. The company also became an independent operating company as enterprise adoption of AI agents raises new governance, access and cost-control issues.

Why it matters: - AI agents are already operating across enterprise applications and cloud platforms, where they can access data, call tools and delegate work. - A simple access record does not show the full business context behind an agent’s actions. - Matrix is designed to give enterprises control over who owns an agent, why it exists, what it can use and what depends on it. - The platform aims to reduce risk, contain misuse and identify unnecessary spend tied to agent activity.

What happened: - idGenius announced Matrix, version 2 of its AI agent identity security platform. - The company also said it is now an independent operating company. - Anomalix will continue separately with advisory, implementation and managed services. - The announcement was made Oct. 2, 2026, in Chicago.

The details: - Matrix uses a relationship ontology to connect AI agents with owners, organizations, business purposes, projects, credentials, tools, data, policies and other agents. - idGenius uses that context to govern actions in real time. - The idGenius agent harness lets other agents request policy decisions before taking actions. - Matrix uses Open Policy Agent and its Rego language to evaluate rules informed by the relationship ontology. - The platform delivers policy decisions through integrations at Model Context Protocol tool boundaries. - An agent’s proposed tool call can be evaluated against identity, owner, purpose, permitted resources and current risk before it proceeds. - Matrix includes an agent kill switch. - When containment is needed, idGenius can deny further actions at connected enforcement points. - The system also shows the owner, resources and dependent workflows affected by that decision. - Matrix brings together six capabilities: discover, learn, govern, enforce, observe and audit. - The platform also connects agent activity and token usage to business purpose. - Enterprises can use that data to measure changes in risk, capital expenditure and operating expense. - idGenius says the platform is built for AI agents and other nonhuman identities.

Between the lines: - The pitch reflects a shift from basic permissioning to context-aware governance for autonomous software. - The focus on business purpose suggests enterprises want to control not just access, but whether an agent’s action still matches its original mandate. - The kill-switch design points to a growing need for rapid containment when an agent behaves outside policy or changes behavior. - The separation of idGenius and Anomalix suggests the company wants a cleaner product focus around Matrix.

What’s next: - idGenius will focus on developing and delivering Matrix as an independent company. - Anomalix will continue its separate consulting and services business. - Enterprises adopting AI agents will likely face more pressure to prove governance, auditability and cost control as deployments expand.

The bottom line: - idGenius is betting that AI agent security needs shared business context, not just access control.