Microsoft Launches Agent 365 to Manage AI Agents Like Human Workers
Key Points
- Agent 365 acts as a control plane for AI agents, offering dashboards, telemetry, and alerts.
- Integrates with Microsoft Entra registry and Microsoft 365 apps for unified security and access control.
- Supports third‑party agents from Adobe, Nvidia, ServiceNow, and Workday.
- Provides real‑time monitoring to prevent unexpected or unsafe agent behavior.
- Available now through Microsoft’s Frontier early‑access program for IT admins.
- IDC forecasts 1.3 billion AI agents by 2028, highlighting the need for management tools.
Microsoft introduced Agent 365, a control plane that lets businesses register, monitor, and secure AI agents across the Microsoft ecosystem. The platform offers dashboards, telemetry, and alerts, and integrates with Microsoft Entra, Microsoft 365 apps, and third‑party services such as Adobe, Nvidia, ServiceNow, and Workday. Available through the Frontier early‑access program, Agent 365 aims to give IT admins the tools to treat AI agents as managed coworkers, ensuring they act predictably and safely.
Agent 365 Provides a Centralized Control Plane for AI Agents
Microsoft is expanding its AI strategy with Agent 365, a platform designed to let organizations manage AI agents the same way they manage human employees. The service acts as a control plane, offering a set of dashboards that display real‑time telemetry, alerts, and connections between agents, people, and data. By providing visibility into how agents operate, Microsoft hopes to prevent unexpected behavior and enhance security.
Integration with Existing Microsoft Infrastructure
Agent 365 works with Microsoft Entra registry to register AI agents and define the resources they can access. The platform also integrates tightly with Microsoft 365 applications, allowing agents to interact with familiar productivity tools while respecting the same security policies applied to human users.
Support for a Broad Ecosystem of Third‑Party Agents
Beyond Microsoft‑built agents, the platform is built to accommodate agents from partners such as Adobe, Nvidia, ServiceNow, and Workday. This openness creates a larger “agent factory” where businesses can select the best‑fit AI solutions while still managing them through a unified console.
Security, Telemetry, and Alerting Features
Key capabilities of Agent 365 include telemetry collection, real‑time alerts, and the ability to set limits on what agents can do. These features are intended to protect organizations from both external threats and internal misbehavior, ensuring that AI coworkers act within defined boundaries.
Early‑Access Through the Frontier Program
Microsoft is rolling out Agent 365 via its Frontier early‑access program, which targets IT administrators who want to test and evaluate AI‑management scenarios before broader deployment. This approach allows feedback from early adopters to shape the final product.
Industry Outlook
Jared Spataro, Microsoft’s chief marketing officer of AI at work, noted that IDC predicts there will be 1.3 billion agents by 2028. Microsoft’s Agent 365 is positioned as the infrastructure needed to manage that growing population of AI assistants safely and efficiently.