TerminologytermStep 2: Models, labs & frontierAll

Term: Large Language Model (LLM)

~6 min read

Estimated time: ~6 min read — for the in-app brief plus opening the primary source.

What this is

An LLM is a generative model specialized in language: drafting, summarizing, answering, rewriting tone. It predicts useful text; it does not guarantee truth.

Everyday example

When you type into ChatGPT, Claude, or Grok and get a written answer or draft, you are using an LLM — a model specialized in language.

An LLM is a large language model — AI specialized in reading and writing language.

  • It predicts the next token, at scale, across a huge range of topics.
  • That is why one tool can draft a contract clause and a campaign line.
  • It is not a database and not a search engine, even when it sounds sure.
  • ChatGPT, Claude, Gemini, and Grok are LLM products you already know.

Next action: Treat LLM output as language, not as a system of record, until you ground it.

What changes in how you lead

How decision rights, process, and ownership should change.

  • Do not let an LLM be the only source of a number that will be decided on.
  • Pair LLMs with retrieval or a human check for high-stakes facts.

Compare related ideas

Large Language Model (LLM) vs Frontier model

An LLM is a type of model. 'Frontier' describes capability tier — the newest, most capable general models. You can also buy smaller, cheaper LLMs that are not frontier.

Open Frontier model

Deep dive

Strengths: first drafts, summarization, extraction into structured formats, translation of tone.

Weaknesses: hallucinations, limited private knowledge unless connected to your systems, uneven specialized expertise.

ChatGPT, Claude, Gemini, and Grok are product experiences built on LLMs from different labs. Enterprise vs consumer versions differ on retention and controls — verify.

Related terms

Related weekly lessons

terminologyllmgenerativechatgptclaudegrok