Microsoft’s AI assistant feature, Copilot Vision, is now available for free in the Edge browser. Announced by Mustafa Suleyman, CEO of Microsoft AI, on Bluesky, this feature allows you to interact with your screen using voice commands, creating what Microsoft terms a “talk-based experience,” where you speak and wait for Copilot to respond.
By opting into this feature, Copilot Vision can view what’s on your screen, as Suleyman mentions. For instance, it can assist with recipe guidance while cooking or help decode job descriptions for tailored interview preparation and cover letter ideas. However, using AI for your resume might not be advisable. As per a Microsoft support page, Copilot Vision can highlight areas on your screen to point you to relevant information but won’t interact with links or perform actions for you.
To experiment with Copilot Vision, open the designated link to Microsoft’s website in the Edge browser. You’ll receive a prompt to enable the feature; once permission is granted, access the Copilot sidebar, click the microphone icon, and your Vision session will commence, indicated by a chime and a change in browser hue.
However, the process may not always be seamless. Personally, it took multiple attempts before Edge offered the opt-in prompt. Even after initiating a Vision session, the controls sometimes didn’t display, leaving me with a “One moment…” message at the bottom of my screen. This could be due to using an older, less powerful laptop, so your experience may be different.
According to Microsoft, while Copilot’s responses to you are logged, your inputs, images, or page content aren’t collected during a Copilot Vision session. To discontinue sharing your screen with Copilot, you can end the session or close the browser window.
