The AI glossary
Plain-English explainers for the concepts and techniques shaping AI — each one tied to the latest news as it happens.
Retrieval-Augmented Generation
A technique that grounds an LLM's answers in documents retrieved at query time, reducing hall…
Embeddings
Numeric vector representations of text or data that let machines compare meaning by distance.
Vector Database
A database built to store and search embeddings by similarity, the backbone of RAG systems.
Fine-Tuning
Further training a pretrained model on a narrower dataset to specialise its behaviour.
AI Agents
LLM systems that plan, call tools, and act over multiple steps toward a goal.
Model Context Protocol
An open standard for connecting AI assistants to external tools and data sources.
Quantization
Compressing a model's weights to lower precision so it runs faster and on less memory.
RLHF
Reinforcement Learning from Human Feedback — tuning models to match human preferences.
Context Window
The maximum amount of text a model can consider at once when generating a response.
Diffusion Models
Generative models that create images or audio by iteratively denoising random noise.
Transformers
The neural-network architecture, built on attention, behind nearly all modern LLMs.
Inference
Running a trained model to produce outputs — the compute cost users actually pay at runtime.
Prompt Engineering
The practice of crafting model inputs to reliably get the outputs you want.
Mixture of Experts
A model design that routes each input to a subset of specialised sub-networks for efficiency.
Chain of Thought
Prompting a model to reason step by step, which improves accuracy on complex tasks.