Microsoft Expands Copilot Studio with Autonomous Computer Use and AI-Powered Research Tools
More details are expected to be revealed during the Microsoft Build conference in May
Microsoft is deepening its push into autonomous computing with a new feature in Copilot Studio that allows AI agents to independently operate user interfaces. Now available via an early access research preview, the feature enables the AI to perform actions like clicking buttons, navigating menus, and entering text—mirroring how a human would interact with applications.
“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,” Microsoft stated. “If a person can use the app, the [AI] agent can too.”
More details are expected to be revealed during the Microsoft Build conference in May.
Last year, AI research lab Anthropic also introduced autonomous computer use capabilities . The company recently announced that its Claude 3.5 Sonnet large language model (LLM) can now interact with computers—viewing screens, moving cursors, clicking buttons, and typing text. This enables it to carry out tasks such as
During the same period last year, Microsoft also introduced Copilot Vision, a feature that allows users to browse the web with AI support. With user permission, it can read and analyze web pages, simplify complex information, and extract key insights. Copilot Vision is now being rolled out to Windows Insiders for broader testing.
Last month, Microsoft unveiled two new AI-powered sales agents for Microsoft 365 Copilot, designed to enhance productivity for sales teams by streamlining lead management and deal closure.
With a public preview scheduled for May, the new Sales Agent and Sales Chat tools will integrate seamlessly with Microsoft Dynamics 365 and Salesforce, providing sales professionals with cutting-edge AI support.
