AI Search
The implementation of Gemini 3 is supposed to efficiently process information. Through system-level integration, the model captures the context of open tabs and handles specific queries directly within the browser interface. As a result, experimental approaches from Project Disco, in which open tabs are converted into customized web apps, are getting integrated into Google Chrome.
In short:
In supported regions like the US, access is provided through the current stable versions of Google Chrome. Eligible users can enable the AI features via the settings under experimental features or through a Gemini icon in the address bar. Manual installation of extensions is not required.
The auto browse feature detects the structure of a webpage as it loads and immediately provides a summary. The sidebar can also be used to extract tables or create comparisons between different tabs. The system adjusts the level of detail based on the respective context, for example for technical specifications or instructions. A specific launch date for the European market has not yet been announced. However, it is expected that the rollout will be delayed due to regulatory adjustments related to the DMA and the AI Act.
Google Blog Image source: Google
