Microsoft Empowers Copilot Studio with New ‘Computer Use’ Automation Feature
Microsoft has introduced a powerful new feature to its Copilot Studio platform that allows AI agents to independently interact with websites and desktop applications – without needing traditional API access.
Announced this week, the new “computer use” capability lets businesses build AI agents that can navigate user interfaces by clicking buttons, selecting menu options, and entering data into fields – essentially mimicking human interaction with software. The move brings Microsoft’s offering in line with similar features recently launched by OpenAI and Anthropic.
“Computer use enables agents to interact with websites and desktop apps by clicking buttons, selecting menus, and typing into fields on the screen,” said Charles Lamanna, Corporate Vice President of Microsoft’s Business & Industry Copilot division. “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.”
With this upgrade, Copilot Studio users can deploy AI agents to automate tasks such as data entry, market research, or invoice processing – streamlining workflows across platforms that were previously inaccessible to automation tools. Microsoft says the AI agents are designed to adapt to changes in app layouts or website elements, ensuring stability even as user interfaces evolve.
Earlier this month, Microsoft rolled out a similar feature called “Actions” to its consumer-facing Copilot. That tool allows the assistant to carry out tasks like booking reservations and online shopping in the background. However, Actions is currently limited to select partners, while the new Copilot Studio functionality is expected to offer broader compatibility across more websites and desktop applications.
The update marks a significant leap for enterprise users seeking seamless automation across their digital operations.
