Computer use in Gemini 3.5 Flash

Google has integrated computer use as a native tool in Gemini 3.5 Flash, enabling developers to build AI agents that can see, reason, and take action across browsers, mobile, and desktop environments. The feature, previously only available as a standalone model, now supports enterprise automation tasks like software testing and knowledge work across professional applications. Google has also implemented security measures including adversarial training and optional safeguard systems to mitigate prompt injection risks for agents operating in live environments.