SECURE ACCESS FOR AI AGENTS
Microsoft Entra Agent ID Public Preview
Manage, protect, and govern agent identities and their access to resources--just as you do for your employees--with an enterprise-grade secure access solution.
OVERVIEW
Take charge of agent sprawl
Get observability and access controls for agents. Govern agent identity sponsorship, access, and lifecycle. Block access for risky agents and to risky resources.
- Accelerate AI adoption, get a complete inventory of your agent fleet, and ensure new agents are created with identity built-in and secure defaults.
- Keep your agent fleet under control with lifecycle management and IT-defined guardrails for both agents and the people who create and manage them.
- Reduce risk of breaches with adaptive conditional access that blocks agents with anomalous activity or risky users. Prevent agent access to unauthorized or malicious public resources with traffic inspection. Gain deep visibility into agent actions.
- Extend the infrastructure you use for protecting apps, devices, and people to agents. Learn more about Microsoft Agent 365, the control plane for agents, which includes Agent ID.
SCENARIOS
Secure access for agents with familiar controls
Extend Microsoft Entra capabilities in the admin center to manage agent sprawl, gain transparency to agent actions, and prevent attacks.
Unify agent identity provisioning, metadata, and visibility
Assign identities to agents, view in a consolidated registry across your org, and manage agent metadata including capabilities, tasks, logs, and standard protocols.
Ensure agents help maintain security and compliance posture
Enforce real-time policies for agent access to resources, automatically block risky agents, and define granular access control policies based on security attributes.
Bring agents into familiar identity governance processes
Automate governance from deployment to expiration, ensure sponsors are assigned and maintained, and enforce access assignments as intentional, auditable, and time bound.
Detect and block threats by flagging suspicious agent interactions
Detect and flag unusual or unauthorized activities, trace and block agents with compromised tokens, and automatically remediate compromised agents.
Get network visibility and unified access policies
Log agent network activity for audits and threat detection. Apply web categorization to APIs and MCP servers. Restrict file uploads and downloads. Automatically block malicious destinations.
Market landscape
Learn about the growth and impact of agents in large organizations
As organizations rapidly adopt agents for the workplace, expectations and pressures also increase. Key concerns include agent discovery, governance, authorization, and risk reduction.1
42%
of large organizations have deployed agents.
76%
of leaders expect employees to manage agents in 2 to 3 years.
55%
of the workforce accepts or actively embraces AI agents.
78%
of organizations are concerned about cybersecurity for agents.
Get started with Microsoft Entra Agent ID in Agent 365
Microsoft Entra Agent ID is included in Agent 365. To get started, join early access in Frontier, the Microsoft early access program for the latest AI innovations.
RESOURCES
Stay up to date
Discover essential resources for agent access management.
Related products
Explore Microsoft Entra
Secure access for any identity to any AI or resource.
Frequently asked questions
- Register, manage, govern, and protect agents and their access to resources just like you do with your employees. Our purpose-built, secure-by-default solution extends the same controls that you already use—conditional access, lifecycle management, access governance, and network traffic—so you can confidently scale agent adoption and enable secure collaboration of agents with your employees and other agents.
- Agents are digital teammates and AI-powered digital entities that can autonomously perform tasks, make decisions, and access resources. Microsoft Entra treats agents as first-class identities, enabling secure authentication, lifecycle governance, and access control to protect sensitive data and systems.
- AI agents need identity and access management (IAM) to ensure secure authentication, enforce least-privilege access, and maintain governance across their lifecycle. Without IAM, agents can become invisible, over-permissioned, or vulnerable to misuse, posing risks to sensitive data, systems, and compliance. Microsoft Entra enables centralized control and protection of agents for your organization.
Public Preview
Try Microsoft Entra Agent ID
Agent ID is available via Agent 365 early access in Frontier, the Microsoft early access program for the latest AI innovations.
- [1]Source: KPMG AI Quarterly Pulse Survey, September 2025. Survey results based on 130 U.S.-based C-suite and business leaders representing organizations with annual revenue of $1 billion or more.
Follow Microsoft Security