Google is introducing a new agentic AI experience for Gemini Enterprise, designed to move beyond answering prompts and help businesses complete multi-step work from a unified interface.
Unlike conventional chat tools, the new agent can receive an objective, create a plan, select relevant tools and skills, and work across connected company systems. Google will begin the rollout with enterprise users, focusing on the security, performance and governance requirements of large organizations before expanding availability further.
From prompts to coordinated work
The agent can work with business platforms including Google Workspace, Microsoft 365, Slack, Jira, Confluence, Git, BigQuery, Databricks, Postgres and Snowflake. It can also interact with compatible Model Context Protocol servers, enabling organizations to connect approved internal and external tools.
Gemini can automatically select an AI model for a task, while users may also choose a preferred option. Initial third-party model support includes Anthropic's Claude, with plans to add open-source and private models over time.
A dedicated task inbox lets users follow the agent's progress, including delegated subtasks, activated skills and code-related work. The system is designed to provide clearer oversight as AI takes on increasingly complex workflows.
An AI teammate with an enterprise identity
Each agent receives its own Workspace identity, including an email address and organizational context. It can recognize team structures, time zones, calendars and approval processes, then be invited into workflows through email, mentions, file sharing or group chats.
Actions completed by the agent are recorded in an audit trail linked to the AI identity, helping organizations distinguish automated work from human activity. The service will be available across mobile devices, desktop environments, command-line tools and key workplace applications.
Google says Gemini now reaches more than 1 billion monthly active users, while nearly 90% of Fortune 100 companies use Gemini Enterprise. Early business testing has included organizations such as On, Shopify and PayPal.
This shift toward goal-oriented AI could reshape digital work by turning enterprise assistants into more transparent, adaptable collaborators for everyday operations.