Introducing computer use in Gemini 3.5 Flash

Gemini 3.5 Flash now includes built-in computer use capabilities, allowing the model to interact directly with desktop, mobile, and browser environments. By integrating this functionality directly into the main model, developers can create agents that see, reason, and perform actions across various software platforms. This advancement is designed to support complex, long-horizon tasks such as continuous software testing and automated knowledge work.

The model is available to developers and enterprises through the Gemini API and the Gemini Enterprise Agent Platform. Practical applications already include auditing documentation for accessibility and analyzing software features. These tools enable agents to navigate professional applications, effectively bridging the gap between AI reasoning and real-world digital execution.

To address security concerns, Google has implemented targeted adversarial training to mitigate risks like prompt injection. Enterprises can also utilize optional safeguard systems that require human confirmation for sensitive actions or automatically halt tasks if suspicious activity is detected. These features are intended to be used alongside secure sandboxing and strict access controls to ensure safe operation in live environments.