AI Glossary

An AI agent is a system built on a language model that can plan, use tools, and take a sequence of actions on its own to complete a task, rather than just responding to a single prompt.

Artificial general intelligence (AGI) is a hypothetical AI system that can understand, learn, and perform any intellectual task a human can, rather than excelling only at narrow, specific tasks.

The attention mechanism is the part of a transformer model that lets it weigh how relevant every other word in a sequence is to the word it's currently processing, capturing context and long-range relationships.

Chain-of-thought prompting is a technique that asks an AI model to reason through a problem step by step before giving its final answer, which often improves accuracy on complex tasks.

A context window is the maximum amount of text, measured in tokens, that an AI model can consider at once when generating a response, including both the input and its own output.

A diffusion model is an AI system that generates images, audio, or video by starting from random noise and gradually refining it, step by step, into a coherent output.

An embedding is a list of numbers that represents the meaning of a piece of text, image, or other data, positioned so that similar meanings end up close together in that numerical space.

Fine-tuning is the process of further training an already-trained AI model on a smaller, specific dataset so it becomes better at a particular task or adopts a particular style or behavior.

A foundation model is a large AI model trained on broad, general-purpose data that can be adapted, through fine-tuning or prompting, to many different downstream tasks rather than being built for just one.

Generative AI refers to AI systems that create new content, such as text, images, audio, or code, rather than just analyzing or classifying existing data.

An AI hallucination is a confident, fluent-sounding response from a language model that is factually wrong or entirely made up, rather than an honest admission that it doesn't know.

Inference is the process of running a trained AI model on new input to produce an output, such as generating a response to a prompt, as opposed to training the model in the first place.

A large language model (LLM) is an AI system trained on massive amounts of text that generates and understands natural language by predicting the most likely next word in a sequence.

Multimodal AI refers to models that can understand and generate more than one type of data, such as text, images, audio, and video, within a single system.

A neural network is an AI model made of layered, interconnected nodes loosely inspired by neurons in the brain, which learns patterns from data by adjusting the strength of connections between those nodes.

Prompt engineering is the practice of carefully wording instructions given to an AI model to get more accurate, useful, or consistent responses without changing the model itself.

Quantization is a technique that reduces the numerical precision of an AI model's parameters, making it smaller and faster to run with only a small, often negligible, drop in accuracy.

Reinforcement learning is a training method where an AI system learns by trial and error, taking actions and receiving rewards or penalties that guide it toward better behavior over time.

RLHF is a training technique that uses human ratings of AI-generated responses as the reward signal, teaching a model to produce answers people actually prefer.

Retrieval-augmented generation (RAG) is a technique that lets an AI model look up relevant information from an external source before answering, instead of relying only on what it memorized during training.

A token is the small chunk of text, roughly a word or part of a word, that a language model reads and generates one piece at a time.

Training data is the collection of text, images, or other examples an AI model learns from during training, and it directly shapes what the model knows and how it behaves.

A transformer is a neural network architecture that processes entire sequences of text at once using a mechanism called attention, and it underlies nearly every modern large language model.

A vector database is a database built to store and search embeddings efficiently, finding the entries most similar in meaning to a query rather than matching exact keywords.

Zero-shot and few-shot learning describe an AI model's ability to perform a new task with no examples at all, or with just a handful of examples given in the prompt, instead of requiring dedicated retraining.