Large language model (LLM)
Definition
A large language model (LLM) is an AI system trained on huge amounts of text to predict which words should come next. At scale, that one skill lets it answer questions, summarize, translate and write in a way that reads like a person wrote it.
By Ronan Liedmeier, fractional head of growth and course author · Published 17 October 2026 · Last updated 17 October 2026
Also called: LLM, language model
ChatGPT and Claude are both products built on LLMs. It is the engine under most AI writing, summarizing and chat tools used at work, and it can state a wrong fact as confidently as a right one, which is called a hallucination.
How it works
During training the model reads text and keeps adjusting itself to guess the next token (a small chunk of a word) more accurately. The transformer design most LLMs are built on was introduced in a 2017 paper by Vaswani and colleagues. The model itself does not look anything up when you ask a question. The model writes its reply one token at a time, each choice based on your prompt and everything it has written so far.
So the reply depends on what you put in: a request that includes your own figures and audience gets a far more specific answer than a vague one.
Example
An operations manager pastes last month's support tickets into Claude and asks for the five most common complaints, with ticket numbers for each. The summary comes back in seconds. Because it works from text she supplied, she only spot-checks the ticket numbers before sending it to her team.
Common mistake
Treating the model like a search engine with a database of facts behind it. Ask for a competitor's current price or a statistic with no source attached and an LLM will produce a plausible number from memory. Verify specific numbers, dates and citations before they go into anything you publish or sign.
Learn it in AI Academy
- How AI Actually Works (The 20% You Need)Beginner · 12 minFree
Related terms
- Generative AI
- Generative AI is software that creates new content, such as text, images, audio, video or code, from a plain-language request. It learns patterns from large amounts of existing examples and produces fresh output that follows those patterns.
- Token
- A token is the small chunk of text an AI language model reads and writes, usually a short word or a piece of a longer one. In English one token averages about three-quarters of a word, and context windows and API prices are counted in tokens.
- Hallucination
- A hallucination is a false statement an AI model presents as fact, such as a made-up statistic, quote or source. It happens because language models are trained to produce likely-sounding text, and likely-sounding text is sometimes wrong.
- Reasoning model
- A reasoning model is an AI language model trained to work through a problem step by step before it gives a final answer. The extra thinking makes it slower and more expensive per answer, and noticeably better at math, coding and other multi-step tasks.