AI agent glossary
AI agent
A system where a language model works toward a goal by deciding which actions to take, using tools, and checking results in a loop.
An AI agent combines three things: a language model that does the reasoning, a set of tools it can use to act, such as reading files, searching the web, or calling an API, and a loop that feeds each result back to the model so it can decide the next step.
The difference from a chatbot is who drives. With a chatbot, you ask and it answers, one turn at a time. With an agent, you hand over a goal and the model chooses the steps, often dozens of them, until it is done or needs your input.
Agents are good at tasks that are open-ended but checkable, like fixing a bug until tests pass or researching a question across many sources. They are less suited to tasks where every step is already known, where a fixed workflow is cheaper and more predictable.
You ask an agent to find the three cheapest flights to Lisbon next month and put them in a spreadsheet. It searches, compares options, writes the file, and reports back.
Learn more in the course
Related terms
Go from definitions to doing
Learn to use and build AI agents in 65 practical lessons.