ARTICLE AD BOX
New Delhi: Recently, Google has announced that computer use is now built into tools in Gemini 3.5 Flash, by allowing developers to build AI agents that can interact across platforms, including browser, desktop, and mobile environments.
Earlier, it was available only as a standalone Gemini 2.5 computer use model; the capability is now integrated directly into the Gemini 3.5 Flash. Gemini already supports function calls and built-in tools such as Search and Maps grounding.
With the computer use now built into the main Gemini Flash model, developers can use Gemini 3.5 Flash to reliably build custom agents that can see, reason, and take such action across browser, mobile, and desktop environments.
The capability improves performance for long-horizon and enterprise automation tasks by including continuous software testing and knowledge work across professional apps. To help mitigate prompt injection risks for agents operating in live environments, Google uses adversarial training for computer use in Gemini 3.5 Flash.
Enterprises should take a defense-in-depth approach by combining these safeguards with secure sandboxing, human-in-the-loop verification, and strict across controls. In addition, information about safety measures is available via Google’s best practices documentation. Developers and enterprises can start by using the computer use on Gemini 3.5 Flash via the Gemini API and Gemini Enterprise Agent Platform.
English (US) ·