
Google is expanding access to Gemini Tasks, an AI assistant feature that can make phone calls, book appointments, manage schedules, and complete online purchases on behalf of users. Announced on November 17, 2026, the service is now rolling out for free to hundreds of millions of eligible users across six countries, powered by the company’s Gemini 4 Hydrogen model.
According to Google’s official announcement, Tasks allows users to delegate activities that previously required navigating multiple apps or websites. The feature is launching in the United States, Japan, South Korea, Canada, Australia, and India, with additional capabilities initially available in the U.S.
Gemini Tasks Automates Calls, Bookings and Purchases
Unlike Gemini’s standard Chat mode, which primarily responds to questions, Tasks can perform actions across connected services. Users can activate it by switching the Gemini interface from Chat to Task and describing what they need completed.
The assistant can contact businesses, compare availability, schedule appointments, and manage information across Google services. It can also access supported websites through Google Password Manager with user authorization and complete purchases using Google Pay.
Josh Woodward, Google’s vice president overseeing Gemini, described using Tasks to arrange a vehicle service appointment by contacting two dealerships simultaneously. The assistant identified an available appointment, checked his Google Calendar, and scheduled the visit in less than five minutes.
Woodward also described using Gemini to examine a $389 electricity bill by accessing historical statements and utility account information. The assistant identified increased electricity consumption during peak pricing hours as a contributing factor.
Google Adds Payment Controls and App Integrations
Gemini Tasks supports purchases through Google’s Agent Payments Protocol, which uses digital payment tokens with supported Mastercard and Visa cards. Users can establish spending limits, while the assistant requests approval before completing purchases.
The feature also supports interactions through connected applications, including grocery shopping services and Google Workspace tools. Google said Gemini currently connects with more than 60 applications and can accept task instructions through text messages and forwarded emails in supported markets.
To protect sensitive information, Google runs tasks in isolated virtual environments and applies security filters to websites and external content. Users can adjust permissions, disconnect applications, and require confirmation before sensitive actions.
Gemini Tasks Expands Beyond Paid Subscribers
The technology was initially introduced to selected testers in May 2026 under the name Gemini Spark before becoming available to Google AI Ultra subscribers. It was subsequently renamed Task Mode ahead of the wider rollout.
The latest expansion removes the subscription requirement for eligible users in six countries, although certain advanced capabilities remain geographically restricted. Google also acknowledged that the assistant may occasionally fail to complete tasks or require users to finish actions manually on unsupported websites.
Featured image credits: Wikimedia Commons
For more stories like it, click the +Follow button at the top of this page to follow us.
