AI Literacy / Glossary

Large language model (LLM)

A large language model (LLM) is an AI model trained on huge amounts of text to predict the next small piece of text. Repeating that prediction again and again lets it write answers, summaries, code and conversations.

An LLM reads your message as a sequence of tokens (word pieces) and calculates which token is likely to come next. It picks one, adds it, and repeats until the answer is finished. Chatbots are built on top of LLMs.

Because it has seen so much text, an LLM can produce fluent, useful writing on almost any subject. But likely-sounding is not the same as true: it can invent facts, sources or quotes with complete confidence.

Treat an LLM like a fast, well-read assistant that needs checking. Give it clear context, ask it to show sources, verify anything important and keep private information out of your prompts.