Well, Actually, Claude Can Now Use Your Computer Like an Undergrad Who Finally Showed Up to Lab
Anthropic has released a public beta of computer use for Claude 3.5 Sonnet. The system can move your cursor, click buttons, and type text by interpreting screenshots of your desktop. Small teams can now automate repetitive workflows without writing code, as Claude navigates interfaces visually rather than through APIs.
This teaches the principle of visual GUI automation through multimodal reasoning. You should rethink which tasks require human attention; if a process involves clicking through consistent sequences on screen, an AI agent may now handle it. The workflow shifts from scripting to demonstration: show, don't code.
Anthropic developed this capability for Claude 3.5 Sonnet, released as a public beta. The company emphasizes that small teams are the intended beneficiaries, though they note the system requires supervision as it can be error-prone.
Step 1: Visit claude.ai and start a new conversation with Claude 3.5 Sonnet, then describe a repetitive task you do in a browser (for example, filling out a form with data from a spreadsheet). Step 2: Give Claude explicit permission when it requests computer use access, then watch as it suggests a step-by-step plan before acting; approve or refine this plan. Step 3: Observe Claude taking screenshots and deciding where to click, and intervene if it misidentifies an element; the expected outcome is completing a simple multi-step task without you touching the mouse.