Google Rolls Out Agentic AI in Gemini, Starting With Business Users

Meta Description: Google Cloud launched a unified Gemini agent for enterprise tasks, connecting systems, models, and workspaces while moving AI into an agentic workflow era.

Tags: Gemini Enterprise, AI agents, Google Cloud, enterprise AI, agent workflows


Google Cloud’s new unified Gemini agent is entering the agentic age by combining question answering with task completion in a single interface, a move that pushes enterprise AI from chat windows toward systems that can plan and execute work across business tools.

Background: The Shift From Chatbots to Work-owning AI

For much of the recent AI cycle, chat interfaces were the main product. Systems answered prompts, generated drafts, and guided users through one request at a time. But the direction of travel has changed: AI tools are moving beyond conversational experiences toward systems that can take ownership of assigned tasks, generate code, schedule meetings, book appointments and travel, and more.

That trend has already produced consumer-facing agents like Meta’s Muse and messaging-oriented agents such as Instinct and others, as well as the recent debut of ChatGPT’s Dots. Against that backdrop, Google Cloud said on Thursday that it is bringing its Gemini AI into the agentic age with a unified agent that can not only answer questions but also get things done on the user’s behalf, all from a single interface.

Development: Google Starts With Enterprises, Not Consumers

Google’s entry point is business software. At the event’s start, Sundar Pichai, Google’s CEO, outlined why the company has a good shot at achieving scale for its agentic efforts: Gemini today has over 1 billion monthly active users, and nearly 90% of Fortune 100 businesses now use Gemini Enterprise at work.

Given that corporate footprint, Google will initially focus on bringing the agent to businesses before later rolling it out to consumers. Pichai said the staged approach allows the company to solve the harder problems around security, scale, and performance, which come with launching powerful agents such as these.

Thomas Kurian, CEO of Google Cloud, described the new agent as a system that can be given “objectives, not just instructions.” In practice, that means it can plan work, use custom skills and tools, and connect to businesses’ internal systems to accomplish its goals. By default, the AI picks the best model for the task, but users can also choose a model, including third-party options, starting with Anthropic’s Claude models. Google said it will expand the model picker to include open source models and other private models in the future.

The workflow is designed for real enterprise requests, not only simple prompts. A request can include attachments, like files, folders, or other projects designed for specific workstreams, like those that combine files and skills. The agent can connect to business data and systems, including Google Workspace, Microsoft 365, Slack, Jira, Confluence, Git, BigQuery, Databricks, Postgres, Snowflake, and others. It can also connect and work securely with any Model Context Protocol, or MCP, server inside or outside the company’s network.

Latest Progress: A Coworker-Like Agent With Audits and Spend Controls

Google framed the agent less like a chatbot and more like a digital teammate. The AI will have its own Workspace account, as if it’s just another coworker. That means it has its own email address and its own context. Google said it knows who at the company is on what team, their time zones, who needs to approve items, what’s on people’s calendars, and more. Users can call the agent by tagging it, emailing it, sharing with it, or adding it to a group chat. As it takes action, it writes an audit trail attributed to the agent, instead of a person.

Visibility is handled through a “tasks inbox” interface, where users can see Gemini’s thinking process, delegation of tasks to subagents, loading of special skills, code, and progress. The agent will be accessible from iOS and Android mobile devices, Windows and Mac desktops, the command line interface, Google Workspace, Microsoft 365, ServiceNow, and Slack.

The enterprise angle also extends to early adoption and cost management. Google said early testers included sportswear brand On, Shopify, and PayPal, and noted that Gemini Enterprise customers include large businesses like BNP Paribas, Bradesco, Merck, Orange Spain, Santee Cooper, SOMPO, Ulta Beauty, and Wesfarmers.

New flexible spending options, including multi-model orchestration, smart routing, and real-time spend caps, will help businesses keep their enterprise AI costs under better control. That matters because agentic systems can touch many tools, run many steps, and use different models during a single objective, making cost visibility harder than in a simple chat window.

Where This Leaves Gemini Now

Google’s announcement places Gemini Enterprise at a specific stage: a business-first agent platform that is already being tested and used through existing enterprise relationships, while the consumer rollout remains a later phase. The next signals to watch are how quickly the model picker expands to open source and private models, how widely MCP servers are adopted across company networks, and whether the agent’s audit trail and spend controls become decisive features for organizations evaluating agentic AI.