AI agent

What is it?

An AI agent is an autonomous software or system component that perceives its environment, makes decisions, and takes actions to achieve goals. An agent typically combines a model or learning mechanism with sensors (inputs/observations) and actuators (output/actions) — the underlying AI model(s) are only one part of the whole. AI agents can be reactive, deliberative, learning-based, or part of multi-agent systems, and appear as chatbots, virtual assistants, game NPCs, or autonomous robots/drones.

Practical example

Imagine a mobile app that, based on your calendar, location and preferences, automatically suggests a daily schedule and sends reminders; the background service that reads data, weighs options and initiates actions (booking a ride, sending a notification) acts as an AI agent. In an XR game an AI agent can control NPC behavior: it observes player positions, decides using a behavior tree or policy, and executes movements in the 3D world. In audio/video production an agent might analyze raw footage, tag scenes and propose edits, automating repetitive tasks for human editors.

Test your knowledge

Which of the following characteristics most distinguishes an AI agent from a standalone AI model?

Ask Lex
Lex knows the context of this term and can give targeted explanations, examples, and extra context.
Tip: Lex replies briefly in the widget. For more detail, go to full screen mode.

Learn our language

Learn these terms from real professionals and take your skills further at KdG MCT.

Study at KdG