Skip to main content
Google Gemini AI agent on a smartphone screen, displaying "Tasks" and "Spark" logos, with a shopping cart icon.

Editorial illustration for Google’s Spark AI agent becomes Tasks in Gemini, now free for calls, purchases

Google's Spark AI Becomes Free Tasks in Gemini

• 4 min read

Google is dropping the Spark name. The AI agent it introduced in May, previously locked behind Google AI Pro and Ultra subscriptions, is now called Tasks and runs on Gemini 4 Hydrogen. It's also free, rolling out to everyone in the US, Canada, Japan, Korea, Australia, and India.

The pitch is an agent that actually does things, not just chats. Google says it can build meal plans, manage a calendar, and comb through emails looking for recurring charges a user might have forgotten about. To use it, open Gemini and flip the homepage toggle from "Chat" to "Tasks."

US users get the fuller version. That includes letting the agent call businesses directly to compare prices or check whether something's in stock, pay for purchases through Google Pay, and log into accounts protected by Google Password Manager. Tasks can also be assigned remotely, by emailing [email protected] or texting the service, no app required.

The timing matters. Agentic AI tools built to handle real-world errands have piled up fast this year, from Anthropic's Claude to OpenAI and Meta's own entries, and Google is now pushing its version past the paywall.

Google is rebranding its Spark AI agent as “Tasks” and making it free to use. The feature, which is powered by Google’s new Gemini 4 Hydrogen AI model, can do things like create meal plans, manage your calendar, and dig through your emails to find recurring expenses you might not need.

Why this matters

Google handing an AI agent the ability to call businesses and spend your money through Google Pay, for free, is a bigger bet than another rebrand headline suggests. Tasks moves Gemini from "answer my question" territory into "act on my behalf" territory, and that shift changes what builders and founders need to design for. If an agent can compare prices, check inventory, and complete a purchase without a human confirming each step, the interesting engineering problem stops being accuracy and becomes accountability: who's liable when the wrong item gets bought, or a business gets called at the wrong number.

Google giving this away free, rather than gating it behind Gemini Advanced, also tells us where the company sees the real competitive fight, not chatbot quality, but who controls the default agent sitting between consumers and transactions. Anyone building agentic products should watch adoption numbers here closely. A free, Google Pay-connected calling agent sets the bar other assistants now have to clear, and it puts pressure on smaller teams building similar task-automation tools to find a niche Google hasn't already absorbed.

Common Questions Answered

What is Google Tasks and how does it differ from the previous Spark AI agent?

Google Tasks is the rebranded version of Spark, Google's AI agent that was previously available only to Google AI Pro and Ultra subscribers. Tasks now runs on Gemini 4 Hydrogen and has been made completely free to all users in the US, Canada, Japan, Korea, Australia, and India, making it accessible to a much broader audience.

What specific capabilities can Google Tasks perform beyond answering questions?

Google Tasks can build meal plans, manage your calendar, and search through emails to identify recurring charges you may have forgotten about. The agent can also make calls to businesses, check prices, verify inventory, and complete purchases through Google Pay, moving beyond simple question-answering into taking actions on your behalf.

Why does Google's decision to make Tasks free and action-capable represent a significant shift for Gemini?

Tasks moves Gemini from a conversational AI that answers questions into an autonomous agent that can act on your behalf without requiring human confirmation for each step. This shift from 'answer my question' to 'act on my behalf' fundamentally changes how developers and builders need to design applications and services to work with AI agents.

Which AI model powers the new Tasks feature?

Google Tasks is powered by Gemini 4 Hydrogen, Google's latest AI model that enables the agent to perform complex actions like making calls, comparing prices, and completing transactions. This model represents a significant upgrade from the previous Spark implementation in terms of capability and reliability.

What regions currently have access to Google Tasks?

Google Tasks is rolling out for free to users in the United States, Canada, Japan, Korea, Australia, and India. The feature is being made available to everyone in these regions regardless of subscription level.

LIVE05:12California Takes OpenAI to Court in New Lawsuit Claiming Private Ownership