Google adds screen selection tool to Chrome's Gemini panel, integrates computer use into Gemini 3.5 Flash API

Google has added a screen selection tool to Chrome 149's Gemini panel that allows users to capture text or images from their current tab for prompts. Separately, the company integrated computer use capabilities directly into the Gemini 3.5 Flash model API, replacing the standalone Gemini 2.5 Computer Use model.

Chrome feature: Select from screen

The "Select from screen" tool appears in the Gemini panel's plus menu in Chrome 149. When activated, it highlights the current browser tab and prompts users to "Select any text or image to ask Gemini." The selected content is automatically added to the prompt box.

The feature is rolling out now with Chrome 149. Users who don't see it immediately can restart their browser to trigger the update.

Gemini 3.5 Flash gains native computer use

Google announced that Gemini 3.5 Flash now includes built-in computer use capabilities through the Gemini API. This native integration replaces the separate Gemini 2.5 Computer Use model that was previously available.

According to Google, developers can use the functionality to "build custom agents that can see, reason and take action across browser, mobile and desktop environments." The company claims improved performance for long-horizon and enterprise automation tasks, including continuous software testing and knowledge work across professional applications.

Google provided an example where 3.5 Flash uses computer use to "analyze the Gemini app and return a categorized list of features."

Safety controls for enterprise

Google has implemented safety features for enterprise customers:

• Ability to require explicit user confirmation for sensitive or irreversible actions
• Automatic task termination if an indirect prompt injection is detected

The computer use capabilities join existing Search and Maps grounding features in the Gemini API.

Availability

Gemini 3.5 Flash with computer use is available today through:

• The Gemini API
• A demo environment hosted by Browserbase
• Documentation and reference implementation via the Gemini Enterprise Agent Platform

Pricing for the computer use capabilities was not disclosed.

What this means

The Chrome screen selection tool streamlines multimodal prompting by eliminating the need to manually screenshot and upload images. The integration of computer use into Gemini 3.5 Flash consolidates Google's agentic capabilities into its primary fast model, suggesting the company views browser/desktop automation as a core feature rather than a specialized use case. The safety controls around prompt injection and user confirmation indicate enterprise deployment concerns around autonomous agent actions.