AI agent
A system in which an LLM decides which actions to take, observes the results, and repeats until a goal is met.
An AI agent is a program that uses a language model as its control loop: the model receives a goal and context, chooses an action (usually a tool call), observes the outcome, and decides the next step. What distinguishes an agent from a chatbot is that the model, not a fixed program, determines the sequence of steps. Anthropic's 2024 taxonomy separates workflows (LLMs orchestrated through predefined code paths) from agents (LLMs that dynamically direct their own process and tool usage). In surveys, 'agent in production' usually means a multi-step LLM system with tool access that runs without a human approving every step.