Google has begun rolling out Gemini Spark to subscribers of its AI Pro tier in the United States, following an initial launch in May for AI Ultra users. The personal AI agent, designed to automate complex multi-step tasks, is now accessible via a dedicated tab on the web and a new page in the Gemini mobile app for Android and iOS. Google has indicated that the feature will soon be available to AI Pro members in other countries, though no specific timeline has been provided.

What Is Gemini Spark?

Gemini Spark represents Google's vision of a proactive, autonomous AI assistant that can execute workflows spanning multiple apps and services. Unlike the conversational Gemini chatbot, Spark is built to perform tasks with minimal user intervention. It leverages a suite of Connected Apps—including Google Workspace (Gmail, Calendar, Docs, Sheets, Slides, Drive, Keep, Tasks), Search, Personal Intelligence, a remote web browser, code execution capabilities, Canvas, and location data—to accomplish user-defined goals.

The system introduces three core components: Tasks, Skills, and Schedules. A Task is the high-level objective, such as planning a business trip or compiling a report. A Skill is a reusable set of instructions paired with additional context—for example, a "Travel Booking" skill that knows your preferred airlines and hotel chains. A Schedule is an automated trigger based on time or conditions, such as sending a notification when a flight is delayed and proposing an updated itinerary.

Users can have up to 15 tasks running concurrently, and Spark adheres to the same compute-based usage limits that apply to other Gemini features. This design ensures that even complex, multi-step operations remain manageable without overwhelming the system.

Expanded Workspace Integration

Since its debut with AI Ultra, Google has significantly enhanced Spark's integration with its productivity suite. The agent can now interact with Calendar, Gmail, Docs, Sheets, Slides, Drive, Keep, and Tasks in more sophisticated ways. Below are the key capabilities for each service.

Calendar

Spark can manage schedules by checking calendar availability, RSVPing to invitations, and scheduling new events. It can also reschedule meetings, change locations, or cancel events. A smart scheduling feature allows it to check participants' calendars and suggest open slots that fit everyone's availability, reducing the back-and-forth of finding a meeting time.

Gmail

The agent can search across email threads to find specific topics or summarize conversations. It can draft and reply to emails, forward messages, and organize the inbox using labels. This functionality is particularly useful for users who receive a high volume of correspondence and need quick summaries or automated responses.

Docs, Sheets, Slides, and Drive

Spark can search, read, and manage files in Google Drive. It can create, edit, and format documents, spreadsheets, and presentations. For shared documents, users must review and confirm edits before they are applied, ensuring collaborative integrity. The agent can also read comments on documents, spreadsheets, and presentations, and add images from prompts to documents or slides.

Important: To edit shared files, Spark requires explicit user confirmation before changes are made. This safeguard prevents accidental modifications to collaborative work.

Google Keep

The agent can find, retrieve, and create notes and lists within Google Keep. This allows users to quickly capture ideas or retrieve saved information without manually searching through the app.

Google Tasks

Spark can search for reminders and tasks, as well as add, update, and delete them. Notably, the agent can perform bulk actions on private tasks without requiring additional confirmation. Google advises users to carefully review their requests before submitting them to avoid unintended changes.

How Gemini Spark Differs from Other AI Agents

Google's approach with Spark is part of a broader industry trend toward autonomous AI agents. Unlike Microsoft's Copilot, which primarily assists within Office apps, Spark aims to orchestrate workflows across Google's ecosystem and beyond. Its use of Personal Intelligence—which can access user preferences, location, and browsing history—allows it to personalize tasks without constant prompting.

The remote web browser and code execution capabilities enable Spark to interact with websites and run scripts, making it suitable for data extraction, form filling, and other online tasks. The Canvas feature provides a visual workspace for planning and organizing tasks, while location data can trigger location-based automations, such as sending a reminder when leaving a certain area.

Practical Use Cases

Consider a user planning a business trip to London. With Spark, they can create a single Task: "Plan and manage my business trip to London." The agent then breaks this down into subtasks—booking flights, reserving a hotel, creating a schedule of meetings, and preparing travel documents. It can use a Travel Booking skill to search for flights and hotels according to stored preferences, and a Gmail Writing skill to draft confirmation emails. If a flight is delayed, a Schedule will trigger a notification and propose an updated itinerary.

Another example involves monthly reporting. A user can schedule a recurring Task that extracts data from Google Sheets, summarizes it in a Docs report, and emails the PDF to stakeholders—all without manual intervention. The ability to have 15 tasks running simultaneously means users can automate multiple workflows in parallel.

Availability and Future Outlook

Currently, Gemini Spark is available to AI Pro subscribers in the United States via the Gemini web app and mobile apps. Google has not announced pricing changes—Spark is included in the existing AI Pro subscription, which costs $19.99 per month and also provides access to AI Ultra for advanced reasoning. The company has promised broader international availability soon.

Analysts see Spark as a key differentiator for Google's AI offerings, especially as competition heats up with OpenAI's ChatGPT plugins and Microsoft Copilot. By tightly integrating with its own ecosystem and providing robust automation capabilities, Google aims to attract power users who rely heavily on Workspace products. However, the compute-based usage limits may constrain heavy users, and the requirement for explicit confirmation on shared document edits could slow down collaborative workflows.

Looking ahead, Google is expected to continue expanding Spark's skills library and supporting more third-party apps through the Connected Apps framework. The agent may also gain the ability to learn from user feedback and adapt its behavior over time, moving closer to a truly personalized AI assistant. For now, US-based AI Pro subscribers can start experimenting with Spark's task automation, while users in other regions await the global rollout.

Source: 9to5Google News