Google Expands Gemini Automation to Mac

Google now allows its Gemini application on Mac to independently execute repetitive tasks, a feature already available on PC. Users can describe their requests in natural language to automate work involving multiple steps across different software and devices.
From Sales Listings to 3D Models
To illustrate possible use cases, Google cites eBay: Gemini can prepare a batch of listings for sale. In the example of clothing for sale, the assistant scans a folder of photographs stored on the computer, searches for prices, notes the characteristics of the items, and prepares the listings.
Google also mentions the quick filling of PDF files present on the machine, particularly shipping declarations. For creative activities, the company presents this capability as a way to speed up processes and gives the example of a pencil drawing automatically converted into a 3D model.
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisissez votre rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Gemini Operates Within Authorized Software
These operations rely on the "Computer Use" feature. Gemini observes the display through screenshots and can perform clicks, enter text, scroll content, and navigate through software or websites. Its access remains limited to programs authorized by the user, for which it requests permission for key actions.
Execution takes place in a background window, while the conversation allows users to follow the operations in real-time. The assistant does not take control of the screen or cursor: the Mac remains fully usable during this time.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.