AI glossary

Every artificial-intelligence term explained simply: LLM, RAG, agents, tokens, fine-tuning, hallucination… Clear, up-to-date definitions.

• AGI (artificial general intelligence)AGI refers to a hypothetical AI as versatile as a human, able to perform any intellectual task.
• AI agentAn AI agent combines a model, tools and a decision loop to accomplish a multi-step task autonomously.
• Chain-of-thoughtChain-of-thought has a model reason step by step before answering, which markedly improves complex tasks (math, logic).
• Constitutional AIConstitutional AI is Anthropic's method for aligning Claude: the model follows a set of principles (a 'constitution') guiding helpful, honest and harmless responses.
• Context windowThe context window is the amount of text (in tokens) a model can consider at once — its 'working memory'.
• Diffusion modelA diffusion model generates an image (or video) by starting from random noise it progressively 'denoises' into the result.
• DistillationDistillation trains a small model ('student') to imitate a large one ('teacher'), yielding a lighter, faster model close to the original.
• EmbeddingsAn embedding is a numerical representation (a vector) of text that captures meaning: texts close in meaning have close vectors.
• Fine-tuningFine-tuning re-trains an existing model on specific data to adapt it to a domain or style.
• HallucinationA hallucination is an AI answer that is false but stated confidently — the model 'makes up' plausible information.
• InferenceInference is when an already-trained model produces an answer from an input — running the model, as opposed to training it.
• LLM (large language model)An LLM (Large Language Model) is an AI model trained on vast text corpora to understand and generate language.
• Mixture of ExpertsA Mixture of Experts (MoE) model activates only part of its parameters ('experts') per request, for more power at lower compute cost.
• MultimodalA multimodal model handles several input types — text, image, audio, sometimes video — not just text.
• Open weightsAn 'open-weight' model has its parameters published: you can download, self-host, audit and fine-tune it.
• ParametersParameters are the internal values a model learns during training; their count (often in billions) hints at its size.
• Prompt engineeringPrompt engineering is the craft of writing clear, structured instructions (prompts) to get the best answers from a model.
• Prompt injectionPrompt injection is an attack where malicious content (in a page or document) hijacks a model's instructions to make it take unintended actions.
• QuantizationQuantization reduces a model's numerical precision (e.g. from 16 to 8 bits) to make it lighter and faster, with limited quality loss.
• RAG (retrieval-augmented generation)RAG (Retrieval-Augmented Generation) feeds the model relevant documents retrieved on the fly, so it answers with up-to-date facts and cites sources.
• RLHF (reinforcement learning from human feedback)RLHF trains a model from human preferences: humans rate answers, and the model learns to produce the preferred ones.
• TemperatureTemperature controls the randomness of a model's output: low = safe, deterministic; high = more creative but less reliable.
• TokenA token is the basic unit an LLM processes: a chunk of a word (often ~4 characters in English). Text is split into tokens before processing.
• TransformerThe Transformer is the neural-network architecture behind modern LLMs, based on the 'attention' mechanism.
• Vibe coding'Vibe coding' is the practice of programming by describing your intent in natural language to an AI that writes and edits the code.