Definition
A chatbot is a program that interacts with people through a conversational interface, mainly in the form of text Q&A, evolving from early rule-based scripts to large-language-model-driven systems. Its defining trait is conversational input and output rather than graphical interface operation.
How it works
The typical flow is receiving a user message, generating a reply, and optionally combining retrieval or tools. Chatbots of the rules era relied on fixed scripts and state machines; chatbots of the LLM era have the model generate replies directly, with capability bounded by the model and the context it is given.
Usage
This usage appears in Next Token’s discussions: in Weekly #004, it was noted in the Chinese transcript that ordinary consumers “open a Chatbot or an Agent and half of them don’t know what to do”, using the chatbot as shorthand for consumer-facing conversational AI products. Customer support, Q&A assistants, and companion apps are its typical settings.
Related and confusable terms
The common dividing line from an Agent: a chatbot’s main output is the conversation itself, usually responding within a single turn or session, while an agent aims to complete tasks — planning, calling tools, executing multi-step work in the background. Chatbots today keep absorbing agent capabilities, and at the product level the boundary grows blurrier.