A system that combines a foundation model with the ability to reason about a goal, plan steps, invoke external tools or APIs, and use the results to decide the next action, instead of just generating a single text response.
In AIF-C01 (Domain 2/3) this is the conceptual foundation of "agentic AI." It differs from a traditional chatbot in that the agent can autonomously execute actions (query an API, write to a database), not just converse.
AI agent is a concept that can show up on the exam. Check the full guide to see it in context.