Glossary
AI terms in plain English
Every Word of the Day becomes a permanent page here. No jargon defined using more jargon — each one explains the term the way you would explain it to a colleague, with an example of how it actually shows up in your work.
Get it free- Prompt EngineeringPrompt engineering is the practice of structuring instructions so a language model relia…
- Few-Shot PromptingFew-shot prompting gives the model a few worked examples of the task so it copies the pa…
- Zero-Shot PromptingZero-shot prompting asks the model to do a task with only an instruction and no examples…
- One-Shot PromptingOne-shot prompting includes exactly one worked example to anchor the model's format and …
- Chain-of-Thought PromptingChain-of-thought prompting asks the model to reason step by step before answering, impro…
- System PromptA system prompt is a persistent instruction that sets the model's role, rules, and behav…
- Role PromptingRole prompting tells the model to adopt a specific persona or expertise ('act as a tax a…
- Temperature (LLM)Temperature is a setting that controls randomness in AI output: low is focused and consi…
- Top-p (Nucleus Sampling)Top-p limits the model to the smallest set of words whose combined probability reaches p…
- TokenA token is the chunk of text an AI model reads and generates — roughly a word or word-pi…
- Context WindowThe context window is the maximum amount of text (in tokens) a model can consider at onc…
- AI HallucinationA hallucination is when an AI states something false or fabricated — a fake citation, st…
- Retrieval-Augmented Generation (RAG)RAG feeds the model relevant documents at query time so its answer is grounded in your s…
- EmbeddingsEmbeddings turn text into numeric vectors so similar meanings sit close together, poweri…
- Fine-TuningFine-tuning further trains a model on your examples to permanently shift its style or be…
- Prompt InjectionPrompt injection is an attack where hidden or malicious text overrides an AI's instructi…
- Output FormattingOutput formatting is instructing the model to return a specific structure — a table, JSO…
- Negative Prompting (Constraints)Negative prompting tells the model what NOT to do — no jargon, no preamble, don't invent…
- Self-ConsistencySelf-consistency runs a reasoning prompt several times and takes the most common answer,…
- Prompt ChainingPrompt chaining breaks a big task into a sequence of prompts, feeding each step's output…
- Self-Critique (Reflection)Self-critique asks the model to review and improve its own answer, catching errors and w…
Get one in your inbox each weekday
Free. One word, one prompt, one skill, three minutes.
Knowing the word is half of it. The other half is the mistakes to stop making.