When you use Copilot Vision on a system with two or more monitors, the screen capture logic does not always behave as expected. The AI may analyze the wrong display or miss content you intended to share. This behavior is not random — it follows a deterministic active capture rule based on how Windows manages display focus. This article explains why Copilot Vision captures a specific monitor and how the active window selection works in a multi-monitor setup.
Copilot Vision is a feature in Microsoft Edge that lets the AI see and interpret the content of a web page you are viewing. When you activate Vision, the AI captures the visible area of the browser window and can answer questions about images, text, or layout. On multiple monitors, the capture logic depends on which monitor holds the Edge window that initiated the Vision session. The AI does not scan all displays — it only reads the content of the active Edge window.
This article covers the technical capture logic, the role of Windows display focus, and how to control which monitor Copilot Vision reads. It also explains common capture failures and how to avoid them.