Companies can use Copilot Studio to build AI agents for themselves without needing to code.

Microsoft has announced a new feature for Copilot Studio called “computer use” so AI agents can access websites and applications on their own. Companies can use Copilot Studio to build AI agents for themselves without needing to code.

Users just need to type in what they want their agent to do in natural language, and then test and fine-tune the prompt to see the agent perform in a simulated mode.

“With computer use in Copilot Studio, makers can build agents that automate tasks on user interfaces across both desktop and browser applications, including Edge, Chrome, and Firefox. Additionally, computer use runs on Microsoft-hosted infrastructure, meaning organizations don’t need to manage their own servers. Enterprise data stays within Microsoft Cloud boundaries and is not used to train the Frontier model,” Charles Lamanna, CVP of Business and Industry Copilot, said in a blog post.

The AI agent interacts with websites and desktop apps by clicking buttons, selecting menus and typing into fields on the screen. This means that agents can even connect with a system where there’s no API available.

“If a person can use the app, the agent can too,” Lamanna added.

The AI agents can perform tasks like automating data entry, doing market research and processing invoices.

Microsoft also added that they can automatically adapt to changes in the interface as they’re able to “see” the screen without user intervention.

Copilot Studio users can sign in to get early access to the feature.