# Tasks in Gemini Is Free: Countries, Setup and Approval Limits

How to start a Gemini Task, distinguish the six-country free rollout from US-only features, and set useful limits for calls, purchases and recurring work.

By Paras Madan | 2026-11-18


Tasks in Gemini is rolling out free of charge in six countries. That does not mean every new calling, payment and messaging feature is available in all six. Check that distinction if something announced this week is missing from your account.

For a first attempt, I would hand off a small research task with a clear stopping point. It is easier to judge a comparison you can inspect than a purchase or appointment you need to undo.

## Where the free rollout applies

Google's [November 17 announcement](https://blog.google/innovation-and-ai/products/gemini-app/tasks-in-gemini-free-new-features/) names the US, India, Canada, Australia, Japan and Korea. Tasks use Gemini 4 Hydrogen, while the additional advanced features begin rolling out in the US.

The basic entry point is the Gemini home-screen toggle: change Chat to Task and describe what you want completed. This is task mode inside Gemini, not simply an item added to a to-do list.

Treat rollout language literally. A country appearing in the announcement does not guarantee that every account has received every control. Check the feature's own eligibility rather than assuming a subscription purchase will unlock it.

## Separate Tasks from the US-only additions

Google's [release notes](https://gemini.google/release-notes/) specify English, US availability, personal Google accounts and age 18 or older for the new Google Pay checkout and email-forwarding features. Texting also requires a verified US phone number and an eligible carrier.

Checkout presents an itemized summary and requires explicit one-tap and device approval before purchase. Setting a maximum price helps define a task, but it is not a reason to expect an unattended payment.

Calling has another geographic restriction: Google's launch post excludes Indiana, Louisiana, Minnesota, Montana and Nebraska. Someone eligible for free Tasks may still lack this capability.

## Write a task that has a clear finish

Give Gemini the outcome, constraints and permission boundary together. For example: “Compare three desk lamps available for delivery this week. Keep the total under $60, include shipping, and show the return policy. Do not buy anything.”

That is more useful than “Find me a good lamp” because it defines both a usable result and where the task must stop. The same pattern works for research and household administration: say what matters and what requires your decision.

If no option qualifies, ask it to report that rather than quietly relaxing your requirements.

## Review calls and purchases before they happen

For business calls, the release notes describe reviewing the call plan before dialing and receiving a summary plus transcript afterward. Google says Gemini identifies itself as an AI and does not share payment or sensitive account details over the phone.

Check the business, purpose and information that can be shared before approving. A request to compare appointment availability is different from permission to book any available slot.

For shopping, review the actual item, seller, total and delivery details. Two products can meet the same price cap while differing on compatibility, returns or delivery timing.

## Add schedules after a useful first run

Google's [scheduling guide](https://support.google.com/gemini/answer/17094710) distinguishes task schedules from schedules in ordinary chats. Task triggers can be based on time or conditions, such as an arriving email, and can be described conversationally in a task thread.

I would run the task once and inspect the result before scheduling it. Then specify a notification rule: report only a relevant change, provide a weekly summary, or stop after a deadline. Otherwise, a recurring assistant can become another source of messages to sort through.

Include an end condition. “Check until Friday afternoon” is easier to manage than a request whose usefulness expires silently.

## Know when a task needs you back

Google acknowledges that some websites still require handing control back to the user for the final step. A prepared form or checkout page is not a completed action. Look for the actual confirmation.

The [task-management guide](https://support.google.com/gemini/answer/17094507) also distinguishes local-browser work: the device and Chrome must stay awake and running. Background work does not automatically mean every task continues after the laptop closes.

Start with something bounded and easy to verify. Expand permissions when the workflow earns that trust, not merely because the feature is now free.

I'm Paras. I work at the intersection of AI, engineering and distribution.

I help AI teams turn product launches into practical explanations people can use.

[Book a 30-minute call](https://cal.com/paras-madan/30min).

