Google Cloud announces Gemini agent for enterprise work, now in private preview

At its Gemini at Work 2026 event on Thursday, Google Cloud announced the Gemini agent, which the company describes as ‘a universal agent for work’. Google says the tool is meant to let work begin in one prompt box, drawing on an organization’s business context to answer questions, handle knowledge work, create content and write code. According to Google, the agent plans tasks, uses skills and tools, connects to the business systems of Cloud customers and returns finished results in the documents, inbox and developer environments where people already work.

The Verge reports that the agent will be offered within the Gemini Enterprise app, where users can chat with it and hand it tasks from one interface. It can also work directly inside Workspace apps such as Gmail, Drive, Docs, Sheets and Calendar, and can be reached from phones, desktop computers and the web as well as through third-party apps such as Slack or Microsoft 365. Because it runs in the cloud, The Verge says, it keeps the same context across connected devices. 9to5Google adds that users can mention @Gemini in Gmail, Google Docs, Sheets, Slides and Chat to call it up, and that an API lets it be built into third-party applications as a headless agent that does not require its own user interface.

Google’s pitch, as quoted by 9to5Google, is that users should ‘give it objectives, not instructions’. According to 9to5Google, the agent runs each job on whichever model fits best; today that includes the Gemini and Claude model families, with support for other private and open models planned. It relies on sub-agents for multi-step work, coordinating steps in parallel or in sequence over periods that can stretch to hours or days, and Google describes four kinds of memory: session, semantic, procedural and episodic. The Verge reports that Gemini can also act as a coworker agent with a dedicated identity and its own email address.

Google also gives examples of how the agent could fit into daily work. 9to5Google cites a scenario in which Workspace Intelligence spots a manager’s emailed request for a project update as a slide deck and offers a single-click option to pass it to Gemini. In another, a user asks for a meeting with a familiar group of colleagues without naming anyone, and Gemini works out the participants from a chat space and an earlier thread, checks calendars and starts an email thread to find a time. Beyond acting as a personal assistant, Google says the agent can work on behalf of a group or a specific role, citing a project manager within a team or an analyst in a finance department.

The Gemini agent is currently available only to enterprise customers in a private preview. According to 9to5Google, it will become widely available soon for Workspace customers on select Business and Enterprise plans. The Verge notes that Google launched a consumer agent, Gemini Spark, in May, which subscribers use through a separate interface within the Gemini app, and suggests the new agent may hint at Google’s broader vision of an assistant that answers questions and does work from a single window.