Microsoft launches 'Computer Use' feature in Copilot Studio for smarter AI automation

Microsoft has introduced a powerful new feature called "Computer Use" in its Copilot Studio platform, designed to make AI-powered task automation even more flexible and capable.

With this feature, AI agents can now directly interact with both websites and desktop applications — clicking buttons, selecting menu items, and typing into fields on the screen, just like a human user would. This means AI can automate tasks even when no APIs are available to connect with a system.

Charles Lamanna, Corporate Vice President of Microsoft Business & Industry Copilot, explains:

"If a person can use the app, the AI agent can too."

How It Works

The "Computer Use" feature lets businesses create AI agents that can perform tasks across various user interfaces, including desktop applications and popular browsers like Edge, Chrome, and Firefox.

What’s more — it runs on Microsoft-hosted infrastructure, so organisations don’t need to set up or maintain their own servers. All enterprise data stays within Microsoft Cloud boundaries, and it isn’t used to train external AI models like Frontier, ensuring privacy and security.

Key Use Cases

Here’s what the new feature can automate:

Data Entry: Automatically inputting data from multiple sources into a central system, cutting down manual work and errors.

Market Research: Gathering and compiling market data from the web for easy analysis, eliminating time-consuming manual searches.

Invoice Processing: Extracting details from invoices and entering them into accounting systems, streamlining workflows and reducing errors.

With this move, Microsoft is helping businesses accelerate AI deployment, reduce maintenance needs, and cut infrastructure costs, giving AI a more hands-on role in everyday operations.