Microsoft’s Copilot Vision packs the promise of a truly multimodal assistant: point a camera or share a window, and the AI reads, summarizes, translates, highlights UI elements, and even talks back — a combination of visual comprehension and conversational voice that changes what “help” on a PC or phone can look like. The practical walkthrough in the PCMag UK piece captures that promise in everyday tasks — translating menus in Paris, identifying objects, summarizing manuals and web pages...