Back to Glossary
techniques

Computer Use

AI capability to control computers by viewing screens and performing mouse/keyboard actions.

Share:

Definition

Computer use is the ability of AI models to interact with computer interfaces like a human would - viewing screens, clicking, typing, and navigating applications.

How It Works: 1. AI receives a screenshot 2. Identifies UI elements 3. Decides on action (click, type, scroll) 4. Executes action 5. Receives new screenshot 6. Repeats until task complete

Capabilities: - Click on buttons and links - Fill in forms - Navigate websites - Use desktop applications - Multi-step workflows

Current State: - Still experimental - Can be slow - May make mistakes - Best for repetitive tasks

Providers: - Anthropic Claude (Computer Use API) - Google Project Mariner - Various agent frameworks

Examples

Claude filling out a web form by taking screenshots and generating mouse clicks and keystrokes.

Want more AI knowledge?

Get bite-sized AI concepts delivered to your inbox.

Free intelligence briefs. No spam, unsubscribe anytime.

Discussion