Microsoft recently launched a new, free AI assistant feature called "Copilot Vision" in its Edge browser. This feature provides users with a more intelligent browsing experience, allowing them to interact with the AI assistant via voice commands to interpret on-screen content in real-time and assist with various tasks.

QQ_1744855284393.png

According to Mustafa Suleyman, CEO of Microsoft AI, Copilot Vision is a voice-based assistant. Users simply enable the feature, allowing Copilot Vision to "see" the content displayed on their screen. For example, during cooking, users can ask Copilot Vision for guidance following a recipe, or during a job search, they can have it analyze job descriptions and provide personalized interview preparation advice or cover letter drafts.

However, Microsoft's official support page notes that while Copilot Vision can highlight information on the screen to help users find relevant data, it doesn't perform specific clicks or actions on behalf of the user. Furthermore, the Copilot Vision feature in Edge currently has some limitations. For cross-application and system-level support, such as guidance within software like Photoshop, users need a Copilot Pro subscription.

To experience Copilot Vision, users can access a specific link provided by Microsoft through the Edge browser. After accessing the link, the browser will display an authorization request. Once users agree and authorize, they can launch the voice assistant by clicking the Copilot icon in the sidebar while browsing. Upon successful launch, the browser's color scheme will subtly change to indicate the session has begun. Microsoft states that during Copilot Vision sessions, the system records the assistant's responses to user queries but does not collect user input, images, or webpage content. Users can end screen sharing at any time by ending the session or closing the browser window.

Key Highlights:

🔍 Copilot Vision is a free AI assistant feature in Microsoft Edge, allowing voice interaction.

👨‍🍳 Users can leverage this feature for real-time screen content interpretation and guidance in scenarios like cooking and job searching.

🔒 While the basic feature is free, cross-application support requires a Copilot Pro subscription. User authorization is required.