Google is preparing to introduce a new “Auto Browse” capability for Gemini, now appearing under the tools section with a cursor icon but not yet available publicly. The design and placement of this feature suggest it is aimed at enabling Gemini to autonomously browse the web, manage tabs, and directly interact with Chrome on the user’s behalf. This approach closely follows agentic trends established by platforms like Perplexity Comet and OpenAI Atlas, where users can delegate complex browsing and information retrieval tasks to AI agents. The presence of checks verifying whether Gemini is running inside a Chrome extension further points to a sidebar integration, allowing for seamless control without leaving the Gemini interface.
The Auto Browse tool was recently added to Gemini’s internal toolset, indicating accelerated development following the initial release of Chrome agentic features last September, which at that stage reached only a limited audience in the US. These moves align with Google’s broader strategy to close the feature gap with competing AI-powered assistants and to extend Gemini’s reach beyond conventional chat or search. For users, this could mean managing browsing sessions, researching topics, or executing workflows within Chrome, all orchestrated by Gemini.
Code references indicate that Auto Browse may become exclusive to the Gemini Ultra plan, positioning it as a premium capability. If rolled out as expected, this feature will likely serve professionals and power users who need to automate research or web-based tasks, marking a notable step towards a more agent-driven web experience. Google’s timing and incremental rollout also reflect the company’s intent to test and refine these features before expanding access, which suggests a wider release may follow soon.
