Google has natively integrated the Gemini 3.5 Flash model into its latest released Chrome 149 browser and introduced a new "Select from screen" interaction feature.
This new feature allows users to directly select a portion of the screen and have targeted conversations with AI. Developers and regular users no longer need to switch between different models or complex tools, significantly improving AI interaction efficiency when browsing web pages.
Leave behind global uploads, precisely lock on to local web information
According to Google's official update announcement, when users use Gemini in the Chrome browser, they can use the cursor to precisely outline the specific area they want to ask about. The selected image or text fragment will automatically be attached and combined in real time with the user's input prompt.
This operational logic is very similar to the well-known Google Lens, but its core advantage lies in not redirecting to traditional web search result pages. Instead, it sends the selected local content directly into the conversational AI architecture, allowing the AI agent to provide more targeted answers to details.
Empowering three scenarios, bringing comprehensive innovation to edge-side experiences
The launch of this feature will bring an intuitive experience upgrade for users, such as selecting a few pairs of shoes on an e-commerce website to let AI analyze which one suits their sports style better. Additionally, users can also select complex charts for clearer explanations, or choose a design element on a webpage for creative reshaping.
At the same time, the Gemini 3.5 Flash model has also integrated native computer operation tools. This enables the AI agent to perform complex cross-platform tasks such as accessing websites, filling out long forms, clicking buttons, and collecting data more smoothly in various environments.
