Google said its Gemini desktop app for Macs can now carry out agentic tasks, controlling a user’s system to fill out forms, execute repetitive tasks and interact with online content.
The “Computer Use” feature comes after Anthropic, Meta, OpenAI and others have introduced similar agentic features for their desktop apps.
Google said the feature could be used for carrying out repeated tasks on a batch of files, organising files into folders based on specific criteria, creating slideshows from files or information in online dashboards.
Desktop actions
Apple is developing similar capabilities in an autonomous AI agent, and has begun hiring staff for positions related to the agent, Bloomberg reported.
Apple developing AI agent
One of the listings reportedly sought a senior engineering manager for a proactive, agentic AI assistant to write code, browse the web, control apps and navigate devices on a user’s behalf.
Google this week began more broadly rolling out the “Tasks” feature in its Gemini AI assistant that can carry out multi-step online tasks, such as making online purchases or booking appointments.