AI agent
Definition
An AI agent is a program that pursues a goal on its own: it decides which steps to take, calls tools or APIs to take them, checks the result and repeats until the goal is reached or it hands over. Unlike a chatbot it acts, rather than only answering.
An agent combines a language model for judgement with tools for action — a database query, an API call, a document generator. The model chooses the next step, the tools carry it out, and the outcome is fed back so the agent can correct course.
Reliability comes from constraints, not from the model alone: a narrow scope, validated inputs, a fixed toolset, logging of every action and a rule for when to stop and ask a human.
In practice
- Plans, acts and verifies in a loop instead of producing a single answer
- Needs explicit tool permissions and an audit trail for every action
- Performs best on one clearly bounded domain per agent
Related terms
- Agentic AIAgentic AI is the design approach in which AI systems plan and execute multi-step work autonomously, using tools and feedback, instead of generating a single response. It shifts the unit of automation from an answer to a completed task.
- OrchestrationOrchestration is the coordination layer that decides which agent, model or service handles which step, in what order, with which data, and what happens on failure. It is what turns individual agents into a dependable operational system.
- GuardrailsGuardrails are the technical limits around an AI system: allowed tools, value ranges, spending and volume caps, forbidden actions and mandatory approvals. They turn an unpredictable model into a system that fails safely and visibly.