Google Launches Gemini Agent to Run Work Across Workplace Apps

by Ben Voss
Google Launches Gemini Agent to Run Work Across Workplace Apps

Google Launches Gemini Agent to Run Work Across Workplace Apps

Google Cloud introduced the Gemini agent on October 8, 2026, describing it as a single AI system that can answer questions, plan and complete tasks, create content, and write and run code across business software. The launch moves workplace AI beyond chat toward delegated work performed across connected applications.

Reuters reported that the agent plans work, uses tools, connects to company systems, and returns finished results in documents, email and developer environments. It works inside Google Workspace applications including Gmail, Docs, Sheets and Calendar, and can also be used through Microsoft 365 and Slack.

One agent, multiple models and systems

Google says the Gemini agent uses organizational context, reusable skills and business-tool connections to complete tasks from a single prompt. The platform can connect to services including Confluence, Microsoft Office, Teams, Slack, Salesforce, ServiceNow, Git, Jira, BigQuery, Databricks, PostgreSQL and Snowflake. It can also connect to Model Context Protocol servers inside or outside an organization’s network.

The system is not limited to Google’s own models. Reuters reported that it currently uses Google’s Gemini models and Anthropic’s Claude models, selecting a model for each task. Google said additional models may be added later.

Organizations can create “coworker agents” that act as team members with their own email addresses and access limited to the information provided to them.

Industry versions are already in preview

Google Cloud said specialized versions for financial services and legal work are in preview. Government, healthcare and retail versions are planned. The financial-services version includes skills for investment research, credit analysis and risk modeling. Google says it also provides confidence scores, methodologies, data lineage and source citations for regulated workflows.

Google says nearly 90% of Fortune 100 companies use Gemini Enterprise. That is a company-reported figure and does not independently measure how extensively employees use the tools.

How organizations can start

  1. Confirm access and governance. Work with your Google Cloud administrator to determine whether Gemini Enterprise and the required connectors are enabled.
  2. Connect a narrow set of systems. Begin with approved tools such as Workspace, Slack, Microsoft 365, Git or Jira, and apply least-privilege permissions.
  3. Create one repeatable skill. Start with a bounded workflow, such as preparing a project-status summary or retrieving approved internal documentation.
  4. Review every result. Keep a human approval step for financial, legal, security and customer-facing work, and monitor the agent’s tool use and data access.

Google’s launch puts it in direct competition with OpenAI, Anthropic and Microsoft to make AI agents a new operating layer for office work. The practical test will be whether companies can give these systems enough context to be useful without giving them more access or authority than employees can safely supervise.

References

Google Cloud. (2026, October 8). Welcome to Gemini at Work 2026: Introducing the Gemini agent. Google Cloud Blog. https://cloud.google.com/blog/products/ai-machine-learning/welcome-to-gemini-at-work-2026

Reuters. (2026, October 8). Google Cloud introduces Gemini agent for work as AI race heats up. Investing.com. https://www.investing.com/news/stock-market-news/google-cloud-introduces-gemini-agent-for-work-as-ai-race-heats-up-4939419