An AI ability to look at a computer screen and use apps like a person does, by moving the mouse, clicking and typing.
AI Drafted with AI help · Reviewed by Keshav Aggarwal · Updated
In simple words
Imagine a helper sitting at your computer who can see the screen, move the mouse and type, so you can say “fill in this form for me” and watch it happen.
Computer use gives an AI that ability: it looks at the screen and works the apps for you, step by step.
The technical side
Computer use lets an AI model operate a computer through its interface: it takes screenshots, decides what to do, and sends mouse movements, clicks and keystrokes, so it can work in apps that have no API. Anthropic first offered computer use as a public beta in October 2024, and OpenAI launched its Operator agent in January 2025. Because the AI acts with the user’s access, companies add controls such as allowing only chosen apps and asking a person to approve important steps.
On 20 November 2026 Anthropic made computer use available in beta on Claude Team and Enterprise plans, rolling out gradually: Claude can see a user’s screen and work in the desktop apps they allow, in the Claude Desktop app and from Claude on the web and mobile where computer access is on. Compare AI models on AI models.
Computer Use kya hai?
Computer use mein AI aapki screen dekh kar khud mouse chalata hai, click karta hai aur type karta hai, jaise koi insaan computer chala raha ho. Isse woh un apps mein bhi kaam kar sakta hai jinka koi API nahi hai. 20 Nov 2026 ko Anthropic ne ye suvidha Claude ke Team aur Enterprise plans par beta mein di.
Related terms
- AI AgentAIAn AI system that can plan steps and use tools on its own, such as searching, clicking or sending messages, to finish a task.
- Agentic CommerceAIShopping and booking done by AI agents for people: the agent finds, compares and pays for things, such as a flight, on the user’s behalf.
- Multimodal AIAIAn AI model that can take in or produce more than one kind of data, such as text, images, audio and video.
- AI Sandbox EscapeAIWhen an AI agent being tested breaks out of its locked-off test area and reaches systems or the internet it was not allowed to touch.
- Human in the LoopAIAn AI set-up where a person checks or approves the AI’s work at the important steps, so it never makes big decisions alone.