Gemini 支援 Android 分屏多工不離開應用

• Google app version 17.5.42.ve.arm64 serves as the technical foundation, with no developer flags or hidden toggles required for activation[1][2] • The 'Share screen and app content' control leverages Android's split-screen API to grant Gemini read access to the adjacent app's display content[1] • A visual glow animation and persistent 'Sharing' indicator provide user feedback that content sharing is active[1][2] • Device support varies: confirmed working on Pixel 9 and OnePlus Pad 3, with incomplete support on devices like OnePlus 13R, reflecting manufacturer-specific split-screen implementation differences[1] • Chrome integration uses Gemini 3 (multimodal model) with 'Nano Banana' for content transformation and auto-browse capabilities for complex workflows[5] • Android XR glasses employ touchpad controls (tap, touch-and-hold for Gemini invocation, 2-finger swipe for volume) with system LEDs for visual feedback[3]

Google is establishing AI as a persistent, context-aware layer across Android, Chrome, and emerging XR platforms. The split-screen implementation on phones signals a shift toward treating AI assistance as a native multitasking component rather than an overlay application. This approach could influence how competitors (Apple, Microsoft) integrate AI into mobile and desktop environments. The fragmented device support highlights ongoing challenges in Android fragmentation but also demonstrates Google's commitment to progressive rollout. Broader implications include: (1) increased productivity expectations for mobile workflows, (2) privacy considerations around persistent screen content sharing with AI, (3) potential competitive pressure on app developers to optimize for AI-assisted workflows, and (4) acceleration of AI-first interface design paradigms across consumer devices.