Google is redesigning its Gemini Android app to integrate Gemini Spark, its personal AI agent, more closely with the main chat experience. The update introduces dedicated “Chat” and “Task” buttons above the prompt box, allowing users to switch between conventional conversations and tasks that an AI agent can execute. The redesign is part of Google’s broader effort to turn Gemini from a question-answering assistant into a platform capable of handling recurring activities, multi-step workflows and everyday digital tasks.
The update also introduces a new Scheduled page that brings task management into the main Gemini interface. Users can view active, paused and completed tasks, explore suggested workflows and filter their conversation history by task status. According to reports published on October 12 and 13, 2026, the redesign is rolling out through a server-side update on Android, while the same interface changes have not yet appeared broadly on iOS or the web. Gemini Spark continues to require a Google AI Pro or AI Ultra subscription.
Key Takeaways
- New Chat and Task buttons: Gemini now separates ordinary conversations from AI-agent tasks through two buttons above the prompt box.
- Spark integration: Google’s personal AI agent is integrated directly into the main Gemini experience rather than appearing as a separate navigation item.
- Scheduled page: Tasks are organized into Active, Paused, Completed and Suggested categories.
- Unified history: Chats and AI tasks appear together, with filters for All, Ready, Needs attention and In progress.
- Simplified navigation: Google has removed the standalone Spark entry and renamed Spark settings to Task settings.
- Android-first rollout: The redesigned interface is rolling out on Android, with broader iOS and web availability not yet confirmed.
- Paid subscription requirement: Gemini Spark remains available to Google AI Pro and AI Ultra subscribers.
- Strategic shift: The redesign highlights Google’s push to make Gemini a practical AI agent that can execute tasks, not just respond to prompts.
What Has Changed in the Google Gemini App?
The most visible change is the introduction of two buttons directly above Gemini’s prompt box. Users can select Chat for conventional interactions or Task to access the Spark-powered agent experience.
Previously, Spark was presented as a separate section within Gemini’s navigation. The new design brings task creation closer to the main conversation interface, reducing the need to navigate between different sections to access AI automation.
Google is also changing how conversations and tasks are organized. Instead of maintaining separate lists, the app combines chats and Spark tasks into one history view. A filter icon allows users to sort entries by status, including All, Ready, Needs attention and In progress.
The navigation drawer has also been updated. The standalone Spark menu entry disappears, while the account menu now uses the name Task settings instead of Gemini Spark Settings.
These changes primarily concern the interface and organization of existing capabilities. They do not represent the launch of a new underlying AI model, nor has Google announced a new pricing structure alongside the redesign.
| Feature | Previous experience | Redesigned experience |
|---|---|---|
| Starting a task | Access Spark through a separate section | Select Task above the prompt box |
| Standard conversations | Main Gemini chat | Select Chat above the prompt box |
| Task management | Separate Spark experience | Integrated into the main Gemini app |
| Scheduled activities | Existing scheduling interface | New Scheduled page |
| Conversation history | Chats and Spark tasks more separated | Combined history |
| History filters | Existing task navigation | All, Ready, Needs attention and In progress |
| Settings | Gemini Spark Settings | Task settings |
| Initial rollout | Existing Spark availability | Server-side Android rollout |
What Is Gemini Spark?
Gemini Spark is Google’s personal AI agent, introduced in May 2026. Unlike a conventional chatbot that primarily responds to individual prompts, Spark is designed to carry out tasks on a user’s behalf, under their direction.
The agent can work with supported Google services and connected applications to complete workflows that involve several steps. These capabilities include organizing information, managing recurring activities and working across documents and other digital resources.
For example, a user could ask Spark to review emails, identify upcoming deadlines and compile a summary. Another user might request a recurring report based on information collected from multiple sources.
Google has also expanded Spark’s ability to interact with websites through Chrome. In supported environments, the agent can help with activities such as researching flight options, arranging apartment viewings and beginning other online errands. Sensitive actions, including payments, can require the user to take over and complete the process.
The distinction is important because AI agents need more than language-generation capabilities. They must also interpret instructions, coordinate tools, track progress and respond appropriately when a task requires additional information or user approval.
The new Gemini interface makes these agent capabilities more prominent by placing Task alongside Chat, presenting automation as a central part of the product rather than a separate feature.
How the New Scheduled Page Works
The Scheduled page is designed to give users a single place to monitor tasks and scheduled activities.
According to reporting from 9to5Google, the page organizes items into four categories: Active, Paused, Completed and Suggested. This gives users a clearer view of which workflows are running, which have been stopped and which have already finished.
The categories also help distinguish ongoing automation from possible future activities. Suggested tasks can give users ideas for how to use Gemini, while the other sections focus on the status of tasks that have already been created.
| Scheduled page category | Purpose |
|---|---|
| Active | Shows tasks that are currently active |
| Paused | Groups tasks that have been temporarily stopped |
| Completed | Displays tasks that have finished |
| Suggested | Presents suggested tasks or workflows |
The Scheduled page builds on Google’s existing work with scheduled actions. In 2025, Google introduced scheduled actions that allowed eligible users to ask Gemini to perform tasks at a specified time or repeat them on a recurring basis.
For example, a user could schedule a weekly digest, request a summary of calendar events or receive recurring updates about a topic. Spark expands the broader idea of scheduled AI assistance by focusing on workflows that can involve multiple steps and connected tools.
The redesigned interface is intended to make those activities easier to find and manage. However, Google has not announced that every task can be scheduled in every region or account type. Availability depends on the feature, subscription and supported integrations.
Why Google Is Combining Chats and AI Tasks
Combining conversations and tasks reflects a change in how Google wants users to interact with Gemini.
In a traditional chatbot, the conversation is the primary product. Users submit a question, receive a response and decide what to do next. AI agents introduce a different interaction model because users may assign a goal and expect the system to carry out several steps before returning a result.
That creates new requirements for the interface. Users need to know whether they are asking for information or authorizing an agent to perform work. They also need to see which tasks are running, which have finished and where additional input may be required.
The Chat and Task buttons help establish that distinction. The unified history then provides a single place to return to both ordinary conversations and ongoing work.
The redesign may also make Spark easier to discover. When an agent is tucked away in a separate navigation section, users may not realize they can assign it tasks from the same app they already use for questions and research.
By integrating the two experiences, Google is attempting to reduce friction between deciding what needs to be done and asking AI to help complete it.
Gemini Spark’s Existing Automation Capabilities
The interface update is part of a larger expansion of Spark’s capabilities throughout 2026.
Google announced Spark in May as a personal AI agent intended to help users manage digital activities proactively. Subsequent updates expanded its integrations and added ways to interact with files, applications and websites.
Google Workspace integration
Spark can work with supported Google services, including Gmail, Docs and Sheets, depending on the available integrations and permissions. These connections can help users organize information, summarize material and prepare documents.
For instance, an AI agent could gather information from several messages and turn it into a structured document. A recurring workflow could also update a spreadsheet using new information from invoices or other permitted files.
Connected applications
Google has expanded Spark’s connections to additional services, including Google Tasks and Google Keep, alongside selected third-party applications. These integrations can support workflows involving notes, task lists, files, reservations and other activities.
The specific actions available depend on the connected service, regional availability and the permissions granted by the user.
Chrome browsing
Google also introduced Chrome browsing capabilities for Spark in July 2026. With permission, the agent can use supported browser sessions to help with multi-step web errands.
This could reduce the time required for repetitive activities, such as comparing travel options or researching services. However, actions involving sensitive information or transactions may require additional confirmation.
The October redesign does not introduce all these capabilities for the first time. Instead, it brings the task-oriented experience into the main Gemini interface so users can access and monitor it more conveniently.
Google Gemini Spark Availability and Subscription Requirements
The redesigned interface is rolling out through a server-side update on Android. That means the changes can appear for eligible users without requiring a conventional app update to introduce every interface element.
As of October 13, 2026, reporting indicates that the new Chat and Task buttons and the revised Scheduled page have not yet appeared broadly on iOS or the web. Google has not provided a definitive timeline for those platforms.
Gemini Spark requires a Google AI Pro or AI Ultra subscription. Access can also depend on account eligibility, region and the particular features involved.
| Availability detail | Status |
|---|---|
| Android redesign | Rolling out |
| iOS redesign | Not yet broadly reported |
| Web redesign | Not yet broadly reported |
| Google AI Pro | Spark access supported for eligible users |
| Google AI Ultra | Spark access supported for eligible users |
| New AI model announced | No |
| New pricing announced | No |
Users who do not see the new interface immediately may need to wait for the server-side rollout to reach their account. The absence of the new buttons does not necessarily mean that the account is ineligible for every existing Gemini feature.
Google’s official Spark documentation and in-app subscription information remain the appropriate places to check eligibility and current feature availability.
How the Redesign Could Affect Google’s AI Strategy
The update comes as major technology companies compete to build AI agents that can perform tasks across software applications.
Google faces competition from OpenAI, Anthropic, Microsoft and other companies developing assistants that can interact with files, websites, business tools and connected services. The competitive focus is gradually expanding beyond model quality to include how reliably an AI system can complete useful work.
Google has several potential advantages in this area. Its ecosystem includes Gmail, Google Drive, Docs, Sheets, Calendar and Chrome, providing a large set of services that can support task-oriented experiences.
Integrating Spark more closely into Gemini could make those capabilities easier to use. Instead of opening a separate agent interface, users can choose Task directly when they want Gemini to perform work.
The unified history could also help users manage the results of longer workflows. If an agent requires clarification or a task is paused, a centralized interface makes it easier to find the relevant activity and continue from there.
However, interface changes alone will not determine whether users adopt AI agents. Reliability, permissions, speed, subscription costs and the ability to complete tasks without excessive supervision will remain important factors.
Privacy and Control Over Automated Tasks
As Gemini becomes more capable of interacting with applications and websites, privacy and user control become increasingly important.
An agent may need access to email, files, calendar information or browser sessions to complete a task. The scope of that access determines what information it can inspect and which actions it can perform.
Users should review the permissions they grant to connected applications and understand whether a task will access personal files, use an authenticated browser session or interact with external services.
Sensitive actions also require care. Researching a purchase is different from completing a payment, and drafting an email is different from sending it. Systems that clearly identify when confirmation is required can help users retain control over consequential actions.
Google has described safeguards for its Chrome browsing features, including handing control back to the user for sensitive steps such as payments. The precise protections available can vary by feature and workflow.
A centralized task history may make it easier to monitor automated activity, but users should still review results and check the status of important tasks rather than assuming that every workflow completed successfully.
The Bigger Picture
Google’s Gemini redesign reflects the broader transition from conversational AI toward task-oriented agents. By placing Chat and Task side by side, combining conversations with task history and introducing a dedicated Scheduled page, Google is making automation a more visible part of the Gemini experience.
The update also builds on the company’s existing investments in Workspace integrations, Chrome browsing and scheduled actions. These capabilities could make Gemini more useful for people who want AI to help manage recurring activities and multi-step workflows. The ultimate value will depend on how reliably Spark completes tasks, how well its integrations work and how clearly users can control the actions it takes.
Looking Ahead
The immediate next step is a broader rollout of the redesigned interface across eligible Android accounts, followed by clarification on when the same experience will reach iOS and the web. Google may also refine the task history, filters and scheduling controls as it learns how users interact with the new interface. For now, the update is primarily a product-design change rather than the announcement of a new model or pricing tier.
Over time, the more important development will be whether Gemini can reliably turn user instructions into completed work across supported applications. A convenient Task button is only the starting point: users will expect agents to track progress, recover from errors, request approval when needed and provide clear results. If Google can deliver that experience consistently, Spark’s closer integration could help Gemini evolve from a chatbot into a more practical personal AI assistant.
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
