Your Employees Already Run AI Agents You Cannot See

Coding assistants, desktop copilots and agents wired to company tools through MCP servers now run on ordinary laptops, with the user's own access. On November 5, Microsoft shipped two features that treat them as a security problem: Defender can now discover them, and a new OS-level layer can contain them. Here is what changed, why it matters, and a six-step plan for any security team, Microsoft shop or not.

For information only. This article is based on Microsoft's official announcement, "What's new in Microsoft Security: November 2026" (November 5, 2026). TechVernia has not tested these features and has no commercial relationship with Microsoft.

A New Kind of Shadow IT

For twenty years, security teams fought shadow IT: apps and cloud services employees adopted without asking, quietly holding company data. The fix was always the same. Find them, then decide what to allow.

Local AI agents are the next wave, and they are harder. A coding assistant, a desktop copilot or an agent connected to company tools through Model Context Protocol (MCP) servers installs in minutes. Once running, it can read files, run commands, open network connections and act with the user's own permissions.

The difference with classic shadow IT is simple: an unapproved app stores data. An unapproved agent takes actions.

50+ Local AI agents Defender can now discover
2 Platforms covered by discovery: Windows and macOS
GA Both features generally available, November 5, 2026
4 Areas MXC restricts: files, network, processes, user interactions

What Microsoft Shipped

01
Discovery

Defender Finds Local AI Agents

Microsoft Defender now discovers more than 50 supported local AI agents on Windows and macOS by observing their activity. It also maps the MCP servers each agent is configured to use, and links together the agent, the device and the user account. For a security team, that is the first real inventory of which agents run where, and what they are plugged into.

02
Containment

Microsoft Execution Containers (MXC)

MXC is an operating system-enforced containment layer for AI agents on Windows, designed to limit what an agent can access. It applies least-privilege controls to files, network, processes and user interactions, with stronger isolation for higher-risk scenarios. Teams can tune and validate policies before enforcing them broadly. Integration with Intune, Entra and Agent 365 is announced as upcoming.

In the same announcement, Microsoft described where it wants to go next: dynamic permissions that give an agent access scoped to its current task, and only for as long as the task lasts; a dynamic risk score that checks whether an agent's authority still matches its behavior; and conditional access that can narrow permissions or require human approval for higher-impact actions. Microsoft did not state availability dates for these.

Why It Matters

An agent running with a user's permissions is, in practice, an insider that never sleeps. It can be wrong, and it can be manipulated. If a web page, an email or a document injects instructions into it (prompt injection), the attacker inherits everything the agent can touch: files, tokens, internal tools behind MCP servers.

The order matters. Discovery comes first, containment second. Blocking agents you have not inventoried only pushes them further into the shadows.

A Six-Step Plan for Security Teams

StepWhat to do
1. InventoryList which AI agents run on your endpoints and which MCP servers they connect to. Defender customers can start with the new discovery view; others can begin with software inventory, process telemetry and config files.
2. Assign ownersEvery approved agent needs a named human who answers for it, its permissions and its connections.
3. Least privilegeNo agent should hold standing admin rights or broad file system access. Scope access to the task, not to the user.
4. Watch the networkBaseline where agents talk. A local agent opening outbound connections to unknown hosts deserves an alert.
5. Human approvalRequire a person to confirm high-impact actions: deleting data, sending email, moving money, changing configurations.
6. Test before blockingRun containment policies in audit mode first, measure what would break, then enforce.

The Fine Print

Early days. Microsoft has not published the full list of the 50+ agents Defender detects. The Intune, Entra and Agent 365 integrations for MXC are announced, not shipped. MXC is a Windows feature, and both tools only help organizations running Microsoft's security stack.

The bigger point stands regardless of vendor: local agents are now endpoints with privileges, and they need the same lifecycle as any other privileged software.

On November 5, 2026, Microsoft made two features generally available: Defender discovery of more than 50 local AI agents and their MCP servers on Windows and macOS, and Microsoft Execution Containers (MXC), an OS-enforced layer that limits what AI agents can access on Windows. Together they treat local agents as a new form of shadow IT that must be discovered, owned and contained.

Frequently Asked Questions

What is a shadow AI agent?
An AI agent running on a company device without the security team's knowledge or approval, such as a coding assistant or desktop copilot. Unlike a regular unapproved app, it can take actions with the user's permissions.
What is Microsoft Execution Containers (MXC)?
An operating system-enforced containment layer for AI agents on Windows, generally available since November 5, 2026. It applies least-privilege limits to an agent's access to files, network, processes and user interactions.
Which AI agents can Defender discover?
Microsoft says Defender discovers more than 50 supported local AI agents on Windows and macOS by observing their activity, along with their configured MCP servers. It has not published the full list.
Why do MCP servers matter for security?
MCP servers connect an agent to tools and data, such as file systems, ticketing systems or internal APIs. Each one extends what the agent, or an attacker who manipulates it, can reach.
What if my company does not use Microsoft Defender?
The same plan applies: inventory agents with your existing endpoint tools, assign owners, enforce least privilege, monitor outbound traffic and require human approval for high-impact actions.

My Take

The real news here is not a product launch. It is that the largest security vendor on Windows now treats local AI agents as endpoints to discover and contain. That is a signal for every security policy, whatever stack you run.

As a network security engineer, I would start with the network. Agents need to talk to models, tools and MCP servers, and that traffic is visible. Knowing which machines talk to which AI services is the cheapest inventory you can build this week, and it does not depend on any vendor.

Do you know how many AI agents are running in your company right now?

Sources:

This article is published for information purposes only, to help readers understand emerging AI security risks.

Related Articles:

Kodjo Apedoh

About the Author

Kodjo Apedoh

Network Engineer & AI Entrepreneur

Founder of TechVernia & SankaraShield. Certified Network Security Engineer with 4+ years of experience specializing in network automation (Python), AI tools research, and advanced security implementations. Also builds iOS and Android applications. Holds certifications from Palo Alto Networks, Fortinet, and Cisco. Based in Arlington, Virginia.

Connect on LinkedIn →