Microsoft introduces Agent 365 control center and a pack of new AI agents for enterprise users
Microsoft Corp. is leaning more into enterprise artificial intelligence agentic workflow support with a new command center for AI agents: Agent 365.
Announced today during Microsoft Ignite 2025 in San Francisco, Agent 365 provides centralized visibility and management for AI agents across an enterprise. It acts as a control plane that brings together telemetry, dashboards and alerts into one place.
“As agents multiply in numbers and sophistication, companies face a new kind of challenge: how to manage and govern agents responsibly and at scale, without rebuilding the trusted systems they rely on,” said Charles Lamanna, president of business apps and agents at Microsoft.
He added that the best way to manage agents in the enterprise is the same way management handles people.
To tackle the problem, Agent 365 brings together five core elements: registration, access control, visualization, interoperability and security.
Building on Microsoft Entra, a comprehensive family of identity and access management products, information technology teams gain a single source of truth for every agent operating in their organization. It provides a single registry to track all agents and their relevant roles across the enterprise: information technology , development, security and business operations.
All agents in the system are required to have an agent ID, which allows them to manage and limit access similar to how user roles are handled. IT can pre-program guardrails for creating, onboarding and managing agents. Policy settings in Entra will enforce adaptive, risk-based roles that respond in real time to context and risk, allowing rogue agents to be detected and curtailed.
On the front end, users will get a visual dashboard filled with advanced analytics to provide a map of connections between agents, users and resources in use. This allows engineers and IT teams to dive into tailored metrics to understand how the complex web of software and people works together. It includes agent metrics for performance, speed and quality alongside detailed logging, reporting and discovery to assist with detecting and logging agent interactions for security and ethics audits.
The company said the new control center is agent supplier-agnostic and will work with agents built using Microsoft platforms, open-source frameworks or third-party platforms across Azure and partner clouds.
All of this is designed to work together with Microsoft Defender, the company’s cybersecurity solution, and Purview, a unified data governance and compliance platform that provides visibility and control for both agents and data.
Purview handles potential AI-related data exposure risks and prevents agents from processing sensitive information by applying policy settings. Defender allows organizations to detect and prevent known exploits and emerging threats that target agents and Entra acts to stop attacks in real-time.
Managing and securing Copilot
As more users adopt Copilot, organizations find themselves needing to monitor and secure its use. To this end, Microsoft released Baseline Security Mode today in general availability.
This helps organizations secure their Microsoft 365 environment using recommended configurations. After a guided experience, IT and security teams can identify gaps in policy, simulate changes and deploy protections for common Microsoft apps, including Office, SharePoint, Teams, Entra and more.
The company also released an additional Agent Dashboard. In preview mode, the dashboard offers centralized reporting on agent adoption across an organization. IT and leadership can use it to analyze trends, group adoption and highlight individual agent actions. The goal is to enable organizational administration and management to understand adoption, aiding in decision-making regarding governance and impact.
Purview has been updated with Data Loss Prevention for Microsoft 365 Copilot, now in preview, which helps ensure that sensitive information stays protected. If a prompt sent to Copilot includes confidential or sensitive information — such as credit card details or personal data — the new prompt protection blocks Copilot and agents from responding, preventing the data from being used for grounding or web search.
Also in preview, Purview Data Security Posture Management for SharePoint can now assist IT administrators in discovering and fixing links that are shared too widely at scale, reducing potential exposure of sensitive information. Purview AI Observability in DSPM, also in preview, will allow IT admins to see everything happening with agents and help security teams make informed decisions to mitigate risks.

New agents for business users and IT admins
Along with the agent control center with Agent 365, Microsoft introduced several new AI agents to assist employees in scaling and using their knowledge.
Work IQ is the intelligence layer behind Microsoft 365 Copilot, the AI chatbot tool integrated into Microsoft products, and the company’s AI agents. Tapping into Work IQ, Microsoft released three new agents designed to simplify workforce management, assist with upskilling and strengthen employee connections.
These new agents include a Workforce Insights Agent designed to provide management with real-time insights into workforce roles, tenure and location to enable data-driven team planning decisions. The new People Agent helps users find the appropriate person to connect with, given their role or skill, based on previous interactions, helping connect employees. The Learning Agent will provide personalized micro-lessons, tips and curated courses to help employees build their skills and AI literacy.
Sales teams will get access to the Sales Development Agent, new in preview. This agent acts as a fully autonomous sales agent, capable of doing research, qualifying and engaging leads during after-hours to provide additional coverage for salespeople. It can research prospects, reach out to them, and follow up without immediate human oversight. However, it will still transfer leads to human workers when it cannot handle a question or encounters an issue.
In Teams, the company’s collaboration platform that combines chat, video meetings and file sharing, a new Teams Admin Agent, now in preview, can help automate tedious recurring tasks. It will be able to securely execute tasks in Teams, such as meeting monitoring or managing user accounts.
A new SharePoint Admin Agent, now in preview within the collaboration platform, will provide IT administrators with AI-driven insights and automation capabilities. Microsoft said it will assist with monitoring inactive and ownerless sites, track overshared content and permission-sprawl, and implement policies to archive or lock down roles as needed. Admin teams can also use it to better understand increasing AI agent usage as Copilot adoption accelerates.
Images: Microsoft
A message from John Furrier, co-founder of SiliconANGLE:
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
- 15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more
- 11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network.
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.





