If you’ve been following Microsoft Ignite this year, you may have noticed something big quietly land among all the announcements. It’s called Agent 365, and it’s shaping up to be the missing piece we’ve all been waiting for in the world of Microsoft AI governance and enterprise-ready agents.

For a while now, organisations have been experimenting with copilots, automations, and custom AI tools, but there has been a bit of a gap. We could build powerful agents, but we couldn’t manage them with the same confidence and end to end visibility we expect from our Microsoft enterprise systems. That’s exactly the gap Agent 365 closes.

What is Agent 365?

Agent 365 is a unified control plane for managing AI agents across your enterprise. Think of it as a central home for identity, observability, lifecycle management, security, and compliance for every agent you build or deploy.

It connects directly into the Microsoft 365 ecosystem, which means your agents operate with the same structure and guardrails your people do. They have identity. They have permissions. You can see what they’re doing. You can govern them properly.

Microsoft’s description sums it up well:

Agent 365 brings together identity, observability, and interoperability so that businesses can manage AI agents with clarity and control as they scale.

Source: Microsoft 365 Blog – “Microsoft Agent 365: The Control Plane for AI Agents”

Why this matters for businesses

Most organisations are now beyond experimenting with AI. They’re moving into production, scaling agents that automate meetings, handle documents, write code, or triage workloads. The challenge is no longer “can we build an agent” but “can we trust this works the way we need it to.”

Agent 365 meets that need in a few key ways:

1. Agents get their own identity

An AI agent now signs in just like a user or service principal. That means transparent permissions, audits, and clear accountability for every action it takes. This alone changes how enterprises can adopt AI.

2. Real observability

You can trace what the agent did, when, and why. Defender and Purview provide monitoring and security insights that sit on top of the agent’s activity so risks can be caught early and managed properly.

3. Integrated governance

Admins can apply policies, enforce access controls, and maintain standards without needing a whole new stack of tools. Agent 365 aligns with existing Microsoft 365 governance models.

4. Interoperability across the organisation

Agents can participate in email threads, update documents, collaborate through chat, or plug directly into workflows. They don’t sit off to the side. They work through the same channels your people do.

5. Technology teams get a proper toolkit

Through the Agent 365 SDK, developers can instrument their agents, define how they authenticate, package blueprints for deployment, and use CLI tools to validate and monitor behaviour. It brings real engineering discipline to the AI space.

What was shown at Ignite

If you want to see Agent 365 in action, the Ignite breakout session “Build Adaptive Intelligence with Agent 365 and Dataverse” is worth your time. It’s a forty-three minute deep dive that demonstrates:

• How developers instrument agents with the SDK
• How blueprints manage identity and lifecycle
• Managed connectivity with Microsoft apps and external systems
• Real examples like Telstra’s Meeting Prep Agent
• How agents move through documents, chat, and email as part of normal work
• How security tools like Defender and Purview surface insights

👉 Watch the full Ignite session: https://ignite.microsoft.com/en-US/sessions/BRK305?source=sessions

This session was one of the clearest signals that Agent 365 isn’t just another feature. It is the foundation for how enterprises will build and govern agents moving forward.

A closer look at how partners are already using it

Our team at Velrada has been hands-on with Agent 365 leading up to Ignite. We recently shared a walkthrough of how this all works in practice and what it unlocks for organisations starting their AI journey.

Here’s the short video from my colleagues that breaks it down in a very real, practical way:

👉 Watch the Velrada overview video: https://youtu.be/lUnRrRHn2dE

It shows how Agent 365 gives organisations visibility, secure access, and confidence when deploying agents into environments where structure and safety actually matter.

Why I think this changes everything

For years, we’ve been building incredible solutions with Power Platform, Microsoft 365, and now AI. But we’ve all felt that governance has been the hardest part. How do you scale responsibly? How do you keep everyone safe? How do you maintain visibility when AI starts to automate more of the work?

Agent 365 finally provides an answer. It brings the same maturity and reliability we expect from enterprise platforms to the world of AI agents.

This is the moment where AI stops being an experiment and becomes operational.

If you haven’t explored it yet, the blog, the Ignite session, and the SDK documentation are great starting points. The sooner organisations understand what Agent 365 enables, the sooner they can build AI solutions that are both innovative and governed.

If you’d like help understanding how Agent 365 fits into your Microsoft ecosystem or how to begin preparing for agent-driven workflows, feel free to reach out. Happy to chat.