Retrieval & Agents
Computer Use
Giving an AI agent control of a computer, letting it click, type, and navigate apps and websites like a person.
Computer use lets an agent operate a graphical interface directly, taking screenshots and issuing mouse and keyboard actions to carry out tasks across arbitrary software, not just through a dedicated API. It opens up automation of almost any on-screen workflow, from filling forms to booking travel. As of 2026 the capability is early and still unreliable for complex, high-stakes tasks, so it is usually run with guardrails and human oversight.
Related terms
AI Agent
An AI system that can take actions and use tools to pursue a goal, not just answer a single question.
Agentic AI
A general term for AI that acts with autonomy over multiple steps, rather than responding one prompt at a time.
Human in the Loop
Keeping a person in the decision path to review, approve, or correct an AI action, especially for high-stakes steps.
Multimodal AI
A model that can work with more than one type of input or output, such as text, images, audio, and video together.