AI Glossary
16 AI terms in plain English — one analogy, three reasons it matters, one concrete example. No jargon explaining jargon.
What is an LLM?
An LLM — large language model — is an AI system trained on enormous amounts of text to predict the next word, and through that to write, summarize, translate and reason.
What is a token in AI?
A token is the small chunk of text — roughly three-quarters of an English word — that an AI model reads and writes, one at a time.
What is a context window?
The context window is the maximum amount of text — measured in tokens — that an AI model can "see" at once in a conversation.
What is an AI hallucination?
A hallucination is when an AI states something false with full confidence — invented facts, fake citations, wrong numbers, non-existent court cases.
What is prompt engineering?
Prompt engineering is the craft of writing instructions that get reliably good output from an AI model — by specifying role, context, format and constraints.
What is an AI agent?
An AI agent is a model that doesn’t just answer — it takes actions: browsing the web, running code, calling tools and working toward a goal across multiple steps.
What is RAG?
RAG — retrieval-augmented generation — lets an AI look up relevant documents first and answer from them, instead of relying only on what it memorized during training.
What is fine-tuning?
Fine-tuning takes an already-trained model and trains it further on your own examples, so it adopts your task, tone or domain by default.
What is an embedding?
An embedding turns text into a long list of numbers that captures its meaning, so computers can measure how similar two pieces of text are.
What is a transformer in AI?
The transformer is the neural-network architecture behind modern AI. Its "attention" mechanism lets a model weigh every word against every other word at once.
What is inference in AI?
Inference is when a trained AI model is actually used — every reply it generates. Training builds the model; inference runs it.
What are parameters in an AI model?
Parameters are the internal numbers a model adjusts during training — its "knowledge dials". More parameters generally means a more capable, more expensive model.
What is temperature in AI?
Temperature controls how predictable or adventurous a model’s word choices are — low means consistent and safe, high means varied and creative.
What is chain-of-thought reasoning?
Chain-of-thought is when a model works through a problem step by step instead of jumping straight to an answer — usually improving accuracy on hard problems.
What does multimodal mean in AI?
A multimodal model understands more than text — images, audio, sometimes video — in one system, and can often produce them too.
What is AGI?
AGI — artificial general intelligence — is a hypothetical AI that matches or exceeds humans across most intellectual work, not just one narrow skill. It does not exist yet.
No terms match — try another word.