At Microsoft Build 2025, Microsoft unveiled a new generation of AI-powered capabilities in Microsoft 365 Copilot Studio, designed to transform how businesses build, automate, and scale intelligent solutions.
Key highlights include:
- Multi-agent orchestration, enabling AI agents to collaborate across departments like HR, IT, and Marketing to handle complex tasks together.
- The new Computer Use functionality, allowing agents to interact with any desktop or web application’s UI—just like a human—without relying on APIs.
- Model Context Protocol (MCP), which connects Copilot Studio agents directly to external knowledge servers and APIs, allowing for real-time data access and simplified maintenance.
- Copilot Tuning, a low-code feature that allows organizations to train AI models using their own internal data, workflows, and processes.
From low-code development to cross-agent collaboration and advanced automation, Microsoft is setting the stage for a new era of enterprise AI—one that is more accessible, secure, and deeply customizable.
Multi-agent orchestration
Multi-agent orchestration in Copilot Studio public preview. enables agents to exchange data, collaborate on tasks, and divide their work based on each agent’s expertise. For example, multiple agents can collaborate across HR, IT, and marketing to help onboard a new employee.
▶️ Watch video
Copilot Studio now supports the integration of custom models through Azure AI Foundry, enabling the use of over 1,900 pre-built models, including those optimized for specific industries. This allows organizations to create agents with highly relevant, context-aware responses tailored to their unique business needs. By combining both pro-code and low-code capabilities, Copilot Studio is evolving into a central platform for building intelligent, low-code agents. This feature is currently available in public preview.
Agents have become first-class business entities, and now you can manage them with the same tool you use to manage organizational identity and access: Microsoft Entra. With Microsoft Entra Agent ID, an agent created through Copilot Studio or Azure AI Foundry is automatically assigned an identity with no additional work required from developers, giving security admins visibility and control. In public preview.
To strengthen data security and prevent potential leaks, Microsoft Purview Information Protection is now extended to Copilot Studio agents that leverage Microsoft Dataverse. This integration enables organizations to automatically classify and protect sensitive information at scale, ensuring that AI-driven solutions align with enterprise compliance and governance requirements.
Automating complex business Orchestrating Generative AI Agents
Discover our latest solution that orchestrates generative AI agents to handle complex business processes — from legal compliance to operational setup — all through an intelligent, real-time chatbot assistant.
In this video, discover how we automate every step of international expansion using cutting-edge technologies like Azure OpenAI, Semantic Kernel, and Agent Collaboration. From regulations to local execution, everything is managed efficiently and intelligently. Let AI handle the complexity, so your team can focus on what truly matters: scaling your business.
New computer use in Microsoft Copilot Studio for UI automation
This new capability allows your Copilot Studio agents to treat websites and desktop applications as tools. With computer use, agents can now interact with any system that has a graphical user interface.
Computer use enables agents to interact with websites and desktop apps by clicking buttons, selecting menus, and typing into fields on the screen. This allows agents to handle tasks even when there is no API available to connect to the system directly. If a person can use the app, the agent can too.
Computer use adapts to changes in apps and websites automatically. It adjusts in real time using built-in reasoning to fix issues on its own, so work continues without interruption. It is also built on Copilot Studio’s robust security measures and governance frameworks, to help ensure compliance with organizational and industry standards.
With Computer Use in Copilot Studio, it’s now possible to build agents that automate tasks across user interfaces on both desktop and browser applications, including Edge, Chrome, and Firefox. This capability runs on Microsoft-hosted infrastructure, eliminating the need for organizations to manage their own servers. Enterprise data remains within Microsoft Cloud boundaries and is not used to train the Frontier model. This approach accelerates deployment, reduces maintenance, and helps lower infrastructure costs.
In this video, you’ll see how the Computer Use Agents model can fill out a web form based on a natural language request from the user — understanding exactly which fields to complete and how to interact with each section.
Model Context Protocol (MCP) in Copilot Studio
Model Context Protocol (MCP) enables direct integration with existing knowledge servers and APIs from within Copilot Studio. When connected to an MCP server, actions and knowledge are automatically incorporated into the agent and updated as capabilities evolve. This streamlines the agent development process and reduces the time required for ongoing maintenance.
MCP servers are made available to Copilot Studio using connector infrastructure. This means they can employ enterprise security and governance controls such as Virtual Network integration, Data Loss Prevention controls, multiple authentication methods—all of which are available in this release—while supporting real-time data access for AI-powered agents.
MCP enables our customers to:
- Easily connect to data sources: Whether you have a custom internal API or external data providers, the MCP protocol enables smooth and reliable integration into Copilot Studio.
- Access the marketplace of existing servers: In addition to custom connectors and integrations, users can now tap into a growing library of pre-built, MCP-enabled connectors available in the marketplace. This capability gives you more ways to connect with other tools and makes using them faster and easier.
- Flexible action capabilities: MCP servers can dynamically provide tools and data to agents. This enables greater flexibility while reducing maintenance and integration costs.
Copilot Tuning
With Copilot Tuning, you can use your company data, workflows, and processes to train models and create agents that perform domain-specific tasks with a high degree of accuracy using Copilot Studio.
For example, a legal firm can create an agent to reflect its unique style and expertise—automating document creation and even drafting arguments that blend institutional knowledge with client-specific context to help build the strongest case possible. Or a consulting company can tune agents for specific industries based on subject-matter expertise. Your agents and data operate securely within the Microsoft 365 service boundary, and Microsoft does not use customer data to train foundation models.
At Bravent, we’re ready to help you unlock the full potential of Microsoft Copilot Studio and tailor AI agents to your business needs. Whether you’re looking to streamline operations, enhance customer service, or accelerate innovation, our experts can guide you every step of the way.
Contact us today to explore how these AI advancements can empower your organization to work smarter and faster.
For more details, you can contact us at Info@bravent.net
