Products & Tools

Google Makes Tasks in Gemini Free Across Six Countries

Google's Tasks in Gemini went free on November 17, 2026 in six countries, with AI-placed calls and Google Pay checkout under user-set caps on Gemini 4 Hydrogen.

Google Makes Tasks in Gemini Free and Adds Calls and Payments
Google Makes Tasks in Gemini Free and Adds Calls and PaymentsZulfugar Karimov / unsplash
By Marcus Bennett5 min read

Why it matters

  • Google announced Tasks in Gemini free in six countries on November 17, 2026, running on Gemini 4 Hydrogen.
  • Free access covers users 18 and over in the US, Canada, Australia, India, Japan and Korea; Pro and Ultra users get access elsewhere.
  • U.S. Gemini can call businesses (blocked in Indiana, Louisiana, Minnesota, Montana and Nebraska) and pay via Google Pay with a hard spending cap.
  • Payments use the Agent Payments Protocol with Mastercard and Visa digital tokens; American Express support is forthcoming.
  • Google cautions tasks can fail or stall, so assigning work does not guarantee end-to-end automation.

Google announced on November 17, 2026 that Tasks in Gemini, its multi-step agent feature powered by Gemini 4 Hydrogen, is rolling out free of charge in six countries. The company describes the rollout as reaching hundreds of millions of people, and for the first time the agent can call businesses and complete purchases with Google Pay under a user-set spending cap.

The stakes are straightforward. Consumer AI agents that complete work end-to-end, rather than merely answering questions, are the current battleground for Google, its cloud rivals and its payments partners. Google's own documentation concedes the gap: tasks can fail or stall before completion, so assigning work to Gemini does not guarantee automation.

Who Gets Free Access and How It Works

Google's product page limits free access to users aged 18 and over in the United States, Canada, Australia, India, Japan and Korea. Pro and Ultra subscribers in selected other countries also receive access.

The interface change is minimal. Users switch the Gemini home-screen control from Chat to Tasks, and the agent then works in the background and notifies them when results are ready. Google positions this as the shift from a conversational assistant to a delegated worker.

What Can the Agent Actually Do?

The advanced features begin in the United States. According to Google, Gemini can now:

  • Call businesses on the user's behalf
  • Use Google Password Manager for approved account access
  • Compare products and check out with Google Pay
  • Accept assigned work through forwarded email or text messages

In Google's dealership example, Gemini used Maps to find nearby shops, called two dealerships simultaneously, checked Calendar availability and booked a service appointment. During that call, Gemini disclosed that it was an AI. Businesses can opt out of such calls entirely.

The calling feature is not available everywhere. Google states it is currently blocked in Indiana, Louisiana, Minnesota, Montana and Nebraska, a restriction that reflects the patchwork of state rules governing AI-placed calls.

How Payments Work and What Protections Exist

Purchases run on the Agent Payments Protocol, with digital-token support through Mastercard and Visa on supported cards. The architecture matters for trust. Users set a hard spending cap, and Gemini is designed to stay within it; receipts and tracking information flow into Google Wallet.

Google says supported cards retain their existing rewards and purchase protections, a detail that lowers the friction for cardholders wary of handing checkout to an agent. Two further commitments stand in the roadmap: American Express support is forthcoming, and Google plans to integrate the Universal Commerce Protocol to cover shopping from discovery to checkout.

New Input Channels: Email and Text

U.S. users can assign tasks without opening the app. They can forward email to [email protected], or verify a U.S. phone number on an eligible carrier to enable texting as an input channel.

Google's examples include:

  • Turning school emails into calendar events
  • Adding grocery items by text or photo to an Instacart order, before handing the user back to Instacart to complete the purchase

The Instacart case illustrates the current design philosophy. Gemini initiates and assembles, but the human closes the transaction. Native connections include Gmail, Calendar, Drive, Docs, Sheets, Slides, YouTube and Maps.

How Secure Is the Agent's Execution Environment?

Tasks run on Google Cloud's Agent Compute Engine inside a hardware-isolated virtual machine dedicated to the user's account. The architecture separates two components: the agent harness, which decides the next action, and the execution sandbox, which carries it out. Outbound traffic passes through mediated infrastructure, and persistent credentials and cloud keys stay outside the guest environment.

That separation is aimed at prompt injection (attacks in which malicious text hidden on a webpage instructs an AI agent to take unintended actions). Google's countermeasures include prompt-injection filters that scan search results and webpages, plus Safe Browsing restrictions that limit which links Gemini can open.

User controls are layered on top:

  • Users can add instructions while a task is running
  • Gemini requests confirmation or pre-approval before purchases and other critical steps
  • Routine actions can be set to always allow
  • Third-party connections are off by default
  • Users can disconnect apps, adjust task settings or delete chat history

What Are the Reliability Limits?

Google's own caution is the most consequential fact in the release. The company states that tasks can still fail to complete or miss their objective. On some websites, Gemini can execute most of a task before returning the screen to the user for a final click. The implication is blunt: assigning a task does not guarantee end-to-end automation, and the last-mile handoff remains a designed behavior, not a bug.

Where Did This Come From? Spark and Hydrogen

The November 17 release builds on two prior milestones. On May 19, 2026, Google introduced Gemini Spark, a cloud-based personal agent running on Gemini 3.5. Spark started with trusted testers, with a U.S. Google AI Ultra beta planned next. Its design already included Workspace integration, background operation after a laptop was closed or a phone locked, recurring tasks, and confirmation requests before high-stakes actions.

Five days before the free rollout, on November 12, 2026, Google introduced Hydrogen, a model tuned for rapid execution across agentic workflows, while positioning Carbon and Argon for deeper, long-horizon reasoning. Google says Hydrogen improves instruction following and tool selection in the Gemini app, reducing the back-and-forth needed to reach a reliable output. Hydrogen is the engine under Tasks; Spark was the design study that preceded it.

Why the Free Rollout Matters

The free tier gives Google a vastly wider base for its consumer agent than the paid Ultra track that Spark occupied, and it does so while payments networks such as Mastercard and Visa, with American Express pending, have signed on to tokenized agent checkout. The two-sided bet is clear: scale the user base now, and let the documented failure modes, the state-by-state calling gaps and the final-click handoffs define the honest boundary of what the agent can finish alone. Whether Gemini converts hundreds of millions of free users into delegators, or into skeptics burned by stalled tasks, will determine how quickly Google can close the distance between assigning work and receiving completed work.

Source: unite.ai

Share this article:

More from Marcus Bennett

Marcus Bennett

Show full bio

Senior reporter covering companies, deals and strategy in artificial intelligence and machine learning at AI In Context.

718 articles

Related articles

  1. Google turns Gemini into an agent that works like a co-worker
  2. Google's Gemini Will Now Make Phone Calls for You — With Limits
  3. Google rolls out Gemini Personal Intelligence to Android Auto
  4. Google Ships Gemini 3.5 Flash, Promises Pro Model Next Month
  5. Google’s Gemini for healthcare targets the paperwork behind care

« Previous articleNext article »