What is an AI agent? A plain-English guide
5 min read
The short answer
An AI agent is software that uses a language model to pursue a goal. Instead of answering one prompt, it decides which steps to take, calls tools such as search, email or a database, looks at the results, and repeats until the job is done or it needs help.
Chatbot, workflow or agent?
- A chatbot replies to a message and does nothing else.
- A workflow runs a fixed sequence of steps you designed. AI may be used inside one step, such as summarizing a text.
- An agent chooses its own steps and loops, so its path can change from run to run.
The four parts of an agent
- Model: the language model that reads the situation and decides what to do next.
- Instructions: the role, goal, rules and output format you give it.
- Tools: the actions it may take, such as searching the web, reading a spreadsheet or creating a calendar event.
- Memory and context: the information it can see, such as the conversation, your documents and earlier results.
When an agent is worth it
Use an agent when inputs vary and judgment is needed, for example sorting support emails, researching a sales lead, or preparing a meeting brief. If the steps are always the same, a plain workflow is cheaper, faster and more predictable. Start with the simplest option that works.
Next step
Read the guide on building your first agent without code, then check the safety checklist before you give an agent access to real accounts.