GoKawiilStarting today, Gemini users on Mac will be able to talk to the AI chatbot from any window on their computer simply by long-pressing the "fn" button on the bottom left corner of their keyboard. They can use the feature to transcribe spoken words at the cursor, so the chatbot can take notes as they read sources or brainstorm out loud. Google says Gemini will return a polished transcript by automatically removing "ums" and "ahs."
If users opt into the chatbot's screen-aware reasoning capability, Gemini will be able to see and understand what's on the screen and identify open apps to execute complex tasks. Users can, for instance, highlight text in an open PDF document and then press "fn" to tell Gemini to create a summary for it. Or, they can highlight text from their notes and then tell Gemini to draft an email from the information in those notes if they have access to Gemini's Spark agentic AI. Users can also get Gemini to generate images for them, say, by vocally describing what they want to see or getting the chatbot to look at ideas open on their screen.
Google launched its native Gemini app for macOS in April, a day after it released the native Windows app. In June, Google added its Spark agentic AI assistant to the app, giving it the ability to do complex tasks, such as asking it to create new spreadsheets or documents. Now, it's rolling out this feature to all Gemini users on macOS around the world. While it only supports English for now, the company says support for more languages is coming later this year.