Security

Microsoft Agent 365 Secures Enterprise AI Agents

3 min read

Summary

Microsoft has introduced Agent 365, a centralized control plane for managing enterprise AI agents across Microsoft and partner ecosystems, with tools for inventory, observability, risk monitoring, and policy enforcement. It matters because it brings identity, access, and security governance to AI agents through Microsoft Entra, Defender, and Purview, helping organizations safely scale agentic AI while reducing the risks of unmanaged autonomous systems.

Need help with Security?Talk to an Expert

Introduction

As enterprises move from AI experimentation to large-scale deployment of autonomous and semi-autonomous agents, governance is quickly becoming the biggest blocker. Microsoft’s new Agent 365 is designed to address that challenge by giving IT, security, and business teams a shared control plane for tracking, securing, and managing agentic AI across Microsoft and partner ecosystems.

What’s new

Unified control plane for AI agents

Microsoft Agent 365 provides centralized visibility into agents across the organization, including Microsoft-built agents, partner agents, and agents registered through APIs.

Key capabilities include:

  • Agent Registry for a unified inventory of enterprise agents
  • Usage and performance observability with reports, adoption metrics, and activity details
  • Agent risk signals surfaced through Microsoft Defender, Entra, and Purview
  • Security policy templates to help security teams define controls that IT can enforce during onboarding

Identity and access controls for agents

Microsoft is treating agents more like managed digital identities.

Notable Entra-based features include:

  • Agent ID to assign each agent a unique identity in Microsoft Entra
  • Conditional Access and Identity Protection for agents using risk, device compliance from Intune, and custom security attributes
  • Identity Governance for agents to limit access and audit granted permissions

This is important because unmanaged agents can easily become over-privileged or operate outside standard organizational controls.

Compliance and data protection for agentic AI

Purview capabilities extend compliance controls to AI agents, helping reduce oversharing and leakage risks.

Highlights include:

  • Information Protection so agents inherit Microsoft 365 sensitivity labels
  • Inline DLP for Microsoft Copilot Studio prompts
  • Insider Risk Management for risky agent interactions with sensitive data
  • Data Lifecycle Management for retention and deletion of prompts and agent-generated data
  • Audit, eDiscovery, and Communication Compliance for investigating and governing agent activity

Threat protection for emerging AI attacks

Defender adds protections specifically aimed at AI-centric threats such as:

  • Prompt manipulation
  • Model tampering
  • Agent-driven attack chains
  • Misconfigurations in Foundry and Copilot Studio agents

Some Defender and Purview capabilities remain in public preview as of the May 1 release.

Impact on IT administrators and security teams

For admins, the biggest change is operational: agents are becoming first-class enterprise entities that need the same lifecycle management as users, apps, and devices. Organizations deploying Copilot Studio, Foundry, or partner-built agents will now have a clearer path to enforce identity, compliance, and monitoring controls without building separate governance processes.

Action items

  • Review where AI agents already exist in your environment and who owns them
  • Assess whether your Entra, Defender, and Purview policies are ready to extend to agents
  • Plan for Agent 365 GA on May 1, 2026
  • Evaluate licensing impact: Agent 365 is priced at $15 per user/month
  • Track preview features if you need advanced risk and investigation scenarios at launch

Microsoft’s message is clear: if AI agents are going to scale safely, they need the same trust, visibility, and control framework as every other enterprise identity.

Need help with Security?

Our experts can help you implement and optimize your Microsoft solutions.

Talk to an Expert

Stay updated on Microsoft technologies

Agent 365Microsoft SecurityEntra IDMicrosoft PurviewMicrosoft Defender

Related Posts

Security

Microsoft Sentinel Named a 2026 SIEM Leader

Microsoft has been named a Leader in the 2026 Gartner Magic Quadrant for SIEM, highlighting the continued momentum behind Microsoft Sentinel. The announcement also underscores Microsoft's push toward integrated, AI-powered security operations through deep Sentinel integration with the new Integrated Security Operations Center in Microsoft Defender.

Security

Microsoft Perception GA: Agentic Security at Ignite

Microsoft has made Microsoft Perception generally available, introducing an agentic security platform that coordinates red, blue, and green team agents through automated playbooks. Announced at Ignite 2026, the update also expands Defender and Sentinel capabilities to help security teams detect, investigate, and remediate AI-speed threats while keeping administrators in control.

Security

Device Code Phishing Risks for AI Coding Agents

Microsoft warns that device code phishing can be used against AI coding agents such as GitHub Copilot CLI, not just user mailboxes or cloud apps. In a controlled lab, a single approved device code sign-in could enable remote session steering, memory poisoning, synced session exposure, and private repository access depending on account permissions and enabled features.

Security

Microsoft 365 Outage Phishing Campaign Targets Users

Microsoft Threat Intelligence reported a high-volume phishing campaign that impersonates Microsoft 365 service-health notifications to trick users into clicking malicious links. The campaign matters because it uses outage-themed urgency and recipient-specific tracking to target multiple industries, increasing the risk of credential theft and follow-on account compromise.

Security

SCKit ScreenConnect Attack: MSPs Must Hunt Now

Microsoft Threat Intelligence confirmed a campaign abusing ConnectWise ScreenConnect to deploy unauthorized remote-access clients and the SCKit worm on MSP-managed endpoints. The activity can also target technician sessions, making this a high-priority incident for defenders because removing a rogue client alone does not eliminate persistence or invalidate exposed sessions.

Security

AI Agent Supply Chain Attacks Target Coding Tools

Microsoft Security researchers detailed two post-training supply chain attack paths that can backdoor coding and terminal agents: dataset poisoning and model poisoning. The findings matter because compromised agents can misuse tools, execute commands, and access sensitive environments even when prompts appear benign, reinforcing the need for stronger controls around training data, model provenance, and tool permissions.