Decoder. plain-English AI glossary

Multi-turn

● Core

A conversation where user and assistant exchange multiple messages, building context with each turn.

Think of it like

Tennis rally: user serves, assistant returns, user hits again, etc.

Example

User: "What's the capital of France?" Assistant: "Paris." User: "How many people live there?" Assistant: "About 2.1 million."

How it actually works

Multi-turn requires managing conversation state: remembering prior messages, understanding pronouns and references, and adapting based on feedback. Context windows limit how many turns fit. Solutions: summarization, sliding window, or external memory. Token cost scales with conversation length.

For product teams

Natural interaction. Users expect to refine questions and build on answers.

For engineers

Implement message history buffer. Watch token limits. Summarize old conversations if too long.

Chatbot — Core to chatbots and conversational AI.

Read anything AI without the jargon

Look up any term in plain English, or save terms as you read with the free Chrome extension.

Open Decoder Add to Chrome