Google Chrome Gets Gemini 3 Auto Browse AI Assistant
Google adds Gemini 3 to Chrome with an auto browse feature that completes multi-step web tasks, turning the browser into an AI assistant.
Google has introduced new Gemini 3 capabilities inside Chrome, unveiling an auto browse feature that can carry out multi step tasks on the web on a user’s behalf. Announced on 28 January 2026, the update reframes Chrome as an active assistant, not only a window to websites, according to the company.
What is changing inside Chrome?
As described by Google, the latest Chrome release brings Gemini into a side panel that stays available while a user continues to browse. The company says this experience is designed to help users summarise long pages, compare information across multiple sites and reference details from other Google services without switching tabs repeatedly. The stated aim is to reduce context switching and make routine online tasks, from research to planning, more fluid.
How does auto browse work in practice?
Google explains that auto browse is an opt in, agentic capability that can perform a sequence of steps to complete everyday web tasks. With permission, it can navigate across pages, gather information, fill forms and prepare actions such as applying voucher codes or assembling a shortlist.
Before any sensitive action, for example completing a payment or posting on a social account, the system is designed to pause and seek explicit confirmation. The company says users can review, edit or decline suggested steps at any time.
- Persistent side panel access to Gemini for questions while keeping the current page open.
- Ability to summarise pages, compare options and draft or refine text relevant to what is on screen.
- Optional connections to selected Google services so that relevant information can be surfaced contextually.
- Human in the loop confirmations for high risk actions to preserve user control.
Privacy, safety and controls
Google emphasises that the features are opt in and come with additional safeguards. According to the company, users choose which services to connect, can disconnect them at any time and can inspect what the agent plans to do before it executes. The firm also notes that Chrome’s existing protections against malware, phishing and harmful downloads continue to apply, with new defences rolling out to address emerging threat patterns.
Google states that the upgraded Gemini experience is starting to roll out, with auto browse available as a preview to a select set of users. Wider availability is expected in phases. The company has not specified detailed timelines for additional countries. Users will see prompts to enable the new features and can opt in from Chrome settings when the update reaches their devices.


