politics
Google has introduced Gemini 2.5 Computer Use, an advanced AI model that can navigate the internet like a human being. Built on Gemini 2.5 Pro, the model is designed to interact directly with websites — it can scroll, click, type, fill out forms, and even use keyboard shortcuts to perform complex actions online.According to Google’s blog post, the AI can understand user requests and execute tasks such as submitting forms or organizing data on a website, using a virtual browser that mimics real human behavior. Developers can now access it via Google AI Studio and Vertex AI.
The company also released demo videos showing Gemini 2.5 performing real-time actions such as dragging sticky notes on a webpage, organizing tasks, and interacting with buttons — all sped up for demonstration.
While Gemini 2.5 currently supports 13 types of actions, Google says it already outperforms other models in accuracy and speed. However, it’s currently limited to browser-based operations and is not yet optimized for desktop-level control.
Google teams are already using the model for UI testing and automation, helping developers speed up software testing and interface checks. The technology also powers some intelligent automation features in Search AI Mode, Firebase Testing Agent, and Project Mariner — where AI agents assist users in tasks like research, planning, and data entry.This launch marks another step in making AI tools behave more like real human assistants, capable of performing hands-on digital work.
Comments (1)
Good To Hear