Google adds computer-use capability to Gemini 3.5 Flash

Google said computer use is now built into Gemini 3.5 Flash.

The capability allows developers to build agents that can interact with browser, mobile, and desktop environments. Google said computer use had previously been available through a standalone Gemini 2.5 computer-use model and is now integrated into Gemini 3.5 Flash.

Google said the feature is intended for agentic computer-use tasks, including software testing and work across professional applications. Developers and enterprises can access it through the Gemini API and the Gemini Enterprise Agent Platform.

Google also described safety measures for the feature. The company said it used targeted adversarial training for computer-use behavior and is offering optional safeguards that can require user confirmation for sensitive or irreversible actions and stop tasks when indirect prompt injection is detected.

Google recommends using the feature with sandboxing, human review, and access controls.

Source: Google — https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-computer-use-gemini-3-5-flash/