
Google is rolling out its background Tasks feature in Gemini free of charge to hundreds of millions of users across six countries, allowing the AI assistant to handle multi-step projects, recurring routines and research assignments without requiring users to remain in an active conversation.
Announced on November 17, the expansion covers the United States, India, Canada, Australia, Japan and South Korea. The feature runs on Google’s Hydrogen 4 model and is designed to carry out work in the background, using connected services and information sources to complete assignments and notify users when results are ready.
Tasks in Gemini can handle activities such as preparing daily briefings from emails and calendar entries, researching topics, organizing information and planning trips. Users can also ask it to summarize or archive newsletters, unsubscribe from mailing lists, and monitor flight and hotel prices against a specified budget. For research assignments, Gemini can collect cited sources and package its findings into downloadable documents, PDFs, spreadsheets and presentations.
Users access the feature by switching the Gemini interface from Chat to Task and describing the work they want completed. Unlike an ordinary conversation, a task represents a larger objective that Gemini can manage through multiple steps. Users can provide additional instructions while it is running and inspect its progress through the task interface.
Google also supports schedules that allow these assignments to run automatically. Google AI Pro and Ultra subscribers can schedule tasks hourly, daily, weekly, monthly or yearly. They can also create Gmail monitors that respond to incoming messages matching specified filters, and topic monitors that trigger work when relevant developments occur in areas such as news, finance, sports and local events.
Users without a Google AI subscription, as well as those on Google AI Plus, can schedule tasks to run up to once daily within a general overnight, daytime or evening window. The more advanced scheduling options require a Google AI Pro or Ultra subscription. Google cautions that event monitors are not intended for time-critical alerts because execution may not happen immediately.
The expansion does not include every market where Gemini operates. Google’s Tasks availability requirements list Australia, Canada, India, Japan, South Korea and the United States as countries where the feature is available without a subscription. Elsewhere, Google AI Pro or Ultra is required, except in the European Economic Area, Nigeria, Switzerland and the United Kingdom. Nigerian users therefore cannot access Tasks in Gemini through a paid subscription at this stage.
Access also requires users to be at least 18 years old, have a personal Google Account and keep Gemini Apps Activity enabled. The feature is available through the Gemini mobile app, the Gemini web app and the Gemini app for Mac. Work and school Google accounts are not currently supported.
For website-based assignments, Gemini can use a remote browser hosted in the cloud, allowing work to continue after users close their devices. It can also use a local Chrome browser on desktop, although that option requires the computer and browser to remain running. Tasks involving certain actions, such as signing into a website, can pause and wait for users to take control.
Google says tasks run in hardware-isolated virtual machines dedicated to individual accounts through its Agent Compute Engine. The company also uses sandboxing, prompt-injection filters and Google Safe Browsing protections. Gemini is designed to request confirmation before critical actions, including purchases, although Google acknowledges that tasks can still fail or produce incorrect results.
The November 17 rollout also introduces additional ways to delegate work in the United States. Users can forward emails and attachments to gemini@google.com to initiate tasks, or communicate with Gemini through Google Messages and iMessage after completing phone verification. Google’s new Google Pay integration can monitor product prices and restocks, prepare purchases within a spending limit and complete checkout after users provide explicit approval. These additional features have separate eligibility requirements and are not part of the free rollout in all six countries.
Discover more from Aree Blog
Subscribe now to keep reading and get access to the full archive.


