Google's Gemini app for Mac can now click through apps and websites to finish tasks for you
Google announced on Nov. 19 that the Gemini app for Mac now has "computer use," a feature that lets the AI operate the computer for you: it opens apps and websites, clicks, types and fills in forms to finish multi-step tasks you describe in plain words. Google says Gemini does this in a separate background window with a live preview in the chat, so you can keep using your screen while it works. Google says you choose which apps Gemini can access, can watch each step and pause at any time, and approve key actions before they happen. One example Google gives is pulling details such as addresses out of PDFs saved on the computer and typing them into a multi-box online form. Google says the feature is rolling out in English in the U.S. It requires an Apple Silicon Mac running macOS Sequoia 15.0 or later, and it is available for personal Google accounts (age 18 and up) on the paid Google AI Plus, Pro and Ultra plans. Google's announcement does not list business (Google Workspace) accounts. These are Google's descriptions, and we have not tested the feature. What this means for you: a tool that can read local files and type into websites can also reach client documents, so before anyone at your firm turns this on, decide which apps and folders it may touch, keep it away from client files on personal accounts that sit outside your firm's business agreements, and review any form or filing it fills in before it is submitted. This is general information, not legal advice.
Source: blog.google (opens in a new tab)