Anthropic just gave its Claude AI the ability to take control of your computer and do work for you. The new feature can open files, browse the web, and run programs automatically – even when you’re not there.
This is a big leap from chatbots that just answer questions. Now Claude can actually click buttons, type messages, and complete tasks on your screen like a digital assistant that never takes breaks.
Your AI Employee Never Calls In Sick
The update works through Claude’s Code and Cowork tools, which can now see your screen and interact with any app or website. Anthropic says there’s no complicated setup needed – Claude just starts working with whatever programs you already have installed.
This means Claude could theoretically handle boring tasks like organizing files, sending routine emails, or updating spreadsheets while you sleep. For businesses, it’s like having an employee who works 24/7 without getting tired or making mistakes from fatigue.
But letting AI control your computer raises obvious security questions. Claude will have access to everything on your screen, including passwords, personal files, and sensitive business data. Anthropic hasn’t detailed exactly how they’ll protect this information.
What Happens Next
Expect other AI companies to rush out similar features. Microsoft and Google are probably already working on their own computer-controlling AI tools. The race is on to create AI assistants that don’t just chat, but actually do the work for you.