15 entries

Encyclopedia

Understand AI, one concept at a time.

All entries

AI agentAn AI agent is a system that uses a language model to pursue a goal through several steps: reasoning, using tools, observing the result and adjusting.
Artificial intelligenceArtificial intelligence refers to programs able to perform tasks that seemed to require human intelligence, most often by learning from examples rather than following hand-written rules.
Connectors and MCPA connector links an AI assistant (Claude, ChatGPT, Le Chat…) to a service you already use (GitHub, Vercel, Figma, Canva, Gamma, Notion…) so it can read information there and act on it. The MCP protocol has become the most widely adopted standard for this.
EmbeddingsAn embedding represents a text, an image or another object as a list of numbers (a vector), so that items with similar meanings have similar vectors.
Fine-tuningFine-tuning means continuing to train an already pre-trained model on a smaller dataset to adapt it to a specific task, style or domain.
Generative AIGenerative AI refers to models that can produce new content (text, images, sound, video, code) from a request, based on what they have learned from very large numbers of examples.
HallucinationA hallucination is a response from a generative AI model that sounds plausible and confident but is false, made up or not grounded in the sources.
LLM (large language model)An LLM is an AI model trained on vast amounts of text to predict how a sequence of words continues, which lets it write, summarise, translate or answer questions.
Multimodal AIA multimodal AI understands or produces several types of data at once (text, image, sound, video): for example describing a photo, answering a question about a chart or holding a spoken conversation.
Open and closed modelsA closed model is only available through its provider's service; an open model publishes its weights, which anyone can download and run. In between, licences and degrees of openness vary widely.
Privacy and personal dataWhat you type or send to an AI assistant may be stored, reviewed or used to improve the service. In Europe, the GDPR governs this processing; a few simple habits limit the risks.
PromptA prompt is the text (instructions, context, examples) given to an AI model to get a response; how it is worded strongly affects the quality of the result.
RAG (retrieval-augmented generation)RAG means retrieving relevant documents from a knowledge base, then giving them to a language model so that it bases its answer on them.
TokenA token is the unit of text (word, word fragment, punctuation) that a language model handles; it is also the unit used to measure the length and cost of a request.
TransformerThe Transformer is the neural network architecture, based on the attention mechanism, that underpins most modern language models.