Google introduces business-focused agentic AI for Gemini

At a recent cloud event, Google announced a major update to its Gemini artificial intelligence platform. The company is moving past simple conversational responses to introduce a unified agent system designed to actively complete tasks on a user's behalf.

Moving toward agentic workflows

AI tools are shifting from passive chat interfaces to systems capable of taking ownership of assigned projects, generating code, scheduling meetings, and handling travel bookings. Chief Executive Officer Sundar Pichai noted that Gemini currently serves over one billion monthly active users, and nearly ninety percent of Fortune 100 businesses utilize Gemini Enterprise.

Because of this strong corporate adoption, Google is launching the new agent for businesses first. This initial focus allows the company to address complex challenges related to security, scale, and performance before releasing the feature to everyday consumers.

Objectives and integrations

Unlike traditional chatbots that simply follow strict instructions, the new Gemini agent accepts broad objectives. It can independently plan work, utilize custom tools, and connect directly to internal business systems to reach its goals.

The system automatically selects the best artificial intelligence model for a given task, but users can also manually choose models. This includes third-party options, starting with Anthropic's Claude models, with plans to add open-source and private models later.

To accomplish daily work, the agent integrates with a wide variety of data sources and productivity platforms, including Google Workspace, Microsoft 365, Slack, Jira, Confluence, Git, BigQuery, Databricks, Postgres, and Snowflake. It can also connect securely with any Model Context Protocol server, which standardizes how AI applications communicate with external data sources.