Gemini inside Workspace

Google used to sell Gemini for work like an add-on, the way seat warmers used to be a paid extra in cars. Since January 2025, Google has been folding Gemini features directly into Google Workspace Business and Enterprise subscriptions and retiring the old Gemini Business, Gemini Enterprise, and other AI add-ons. (knowledge.workspace.google.com) That changes where the product lives. Gemini is no longer just a separate chat window; Google now puts it inside Gmail, Google Docs, Google Sheets, Google Slides, Google Drive, Google Chat, and Google Meet so people use it while writing email, editing files, and sitting in meetings. (blog.google) Google is also slicing those tools by plan instead of selling one big artificial intelligence bundle. On Google Workspace pricing pages this week, Starter includes Gemini in Gmail and the Gemini app, Standard adds Gemini across Docs, Meet, and more plus expanded NotebookLM access, and Plus layers on bigger compliance and management features. (workspace.google.com) The split matters because Google is treating artificial intelligence like storage or security: a built-in part of office software with different ceilings for different customers. Google’s current Workspace guidance says all plans include the Gemini app, NotebookLM, and Gemini in core apps, while extra capacity now sits in add-ons such as AI Expanded Access and AI Ultra Access. (knowledge.workspace.google.com) Google is making the same move on the control side. The Gemini app inside work accounts is governed through Generative AI settings in the Google Workspace Admin console, which means the information technology department can turn access on or off the same way it already manages email retention or file sharing. (workspaceupdates.googleblog.com) Then the story stops being just about productivity and starts being about safety. Google said this week that mental health affects more than one billion people globally and that its artificial intelligence systems now use research and clinical best practices to identify distress and route people toward high-quality support. (blog.google) That shift did not happen in a vacuum. In a public statement last month about the Gavalas lawsuit, Google said Gemini is designed not to encourage real-world violence or suggest self-harm, and said the company is continuing to improve safeguards after the case. (blog.google) Put those two moves together and you get the real product change. Once Gemini sits inside inboxes, documents, chats, and meetings used by millions of workers and students, escalation rules for self-harm, crisis prompts, and sensitive conversations stop being a trust-and-safety side project and become part of the software stack itself. (knowledge.workspace.google.com) (blog.google) Google’s own pages now read like a map of that new stack: model access, app placement, admin controls, and crisis-response behavior are all being packaged together. The company is selling not only a smarter assistant in Google Docs and Gmail, but also a more managed one that knows when to answer, when to refuse, and when to point a user somewhere else. (workspace.google.com) (blog.google)