Google has announced what it calls its biggest Gmail update in over 20 years, bringing Gemini 3–driven AI capabilities directly into the inbox. The new features are designed to make Gmail behave more like a proactive personal assistant, helping users manage, write, and respond to emails more efficiently.
According to the company, the rollout began on Thursday and will gradually reach Gmail’s roughly 3 billion users, starting with English-language users in the United States.
At the heart of the update is AI Inbox, a reworked Gmail layout that moves away from a simple time-based email list. Instead, it presents a tailored overview that surfaces priority messages, reminders, and clear next steps.
Users are guided toward suggested actions such as paying a bill or replying to a time-sensitive message, while related emails are grouped into concise topic summaries based on importance. Speaking about the change, Blake Barnes, Google’s vice president of product for Gmail, said the goal is to make Gmail more proactive and genuinely supportive in managing everyday tasks.
Gmail is opening up several of its AI-powered features to everyone at no charge. This includes Help Me Write, which can generate full email drafts based on short prompts, and Suggested Replies, an enhanced take on Smart Replies that adapts to a user’s tone and writing style. Google is also making thread summaries free, allowing users to quickly grasp the main points of lengthy email conversations without reading every message.
Users on Google’s paid AI plans get a deeper set of tools layered on top of the free features. Subscribers to Google AI Pro, which costs $19.99 per month, and AI Ultra, priced at $249.99 per month, can use advanced options such as AI Overviews in Search.
This allows them to ask natural questions, like checking last month’s utility bill, and receive quick answers sourced directly from their inbox. They also gain access to a Proofread tool that works much like Grammarly, offering real-time grammar checks and style improvements to polish email drafts.
Google says the new capabilities have been built with privacy at the core. The company stressed that personal emails are not used to train its future AI models and that all processing takes place securely inside each user’s account. According to Google, the data never leaves this protected environment, ensuring that analysis happens within clearly defined privacy safeguards.
Although many of the new tools are rolling out right away, AI Inbox is starting with a limited release for a small group of trusted testers. Google says a wider rollout is planned later this year, with support for more languages and additional regions to follow over the coming months.
- Google Brings Gemini AI Assistant Capabilities to Gmail - January 12, 2026
- Top 10 Agentic AI Use Cases for Business Automation This Year - January 11, 2026
- What Synthetic Media Is and How It’s Transforming Digital Content? - November 4, 2025
