Entries are in alphabetical order.
In brief
The artificial intelligence concepts that marketing people need, explained without the jargon.
- AI agent
- A model given a role, goals and tools — searching, reading, calculating — to carry out a multi-step task. Read more →
- Hallucination
- False information produced by an AI model in a confident tone. It’s reduced by anchoring the model to sources and asking it to state the gaps. Read more →
- Embedding
- A numerical representation of a text’s meaning, which makes it possible to search by meaning instead of by words. Read more →
- Fine-tuning
- Further training of a model on specific data, to adapt it to a task or a style.
- GEO
- Generative Engine Optimization: getting found, cited and described correctly in AI assistants’ answers. Read more →
- Graph RAG
- A RAG that also uses a graph of entities and relationships extracted from documents, to answer when the answer lies in the connections. Read more →
- Knowledge graph
- A network of entities — brands, competitors, channels, audiences — and the relationships between them, which people and models can query.
- LLM
- Large Language Model: a language model trained on large amounts of text, capable of understanding and producing language.
- MCP
- Model Context Protocol: an open protocol through which AI assistants connect to external data sources and tools.
- Reasoning model
- A model that works through intermediate steps before answering, useful for problems with many variables. Read more →
- Ontology
- The vocabulary of entity types and relationship types used to organize a knowledge graph.
- Prompt
- The instructions and context you give an AI model to get an answer.
- RAG
- Retrieval-Augmented Generation: the model retrieves the relevant passages from your documents and bases its answer on them. Read more →
- Semantic search
- Search by meaning, based on embeddings: it finds relevant texts even when they don’t use the same words. Read more →
- Multi-agent system
- Several AI agents with different roles working on the same problem: experts, moderator, judge. Read more →
- Token
- The unit of text models read and write in: on average, part of a word. Model costs and limits are measured in tokens.
- Synthetic user
- A person simulated by a model, built on a market’s data, used to prepare and test ideas. Read more →
Written by the MAPS editorial team with the help of AI, checked and reviewed by the team.
Try these tools in MAPS
Go from theory to practice: generate analyses, strategies and insight with AI in a matter of minutes.
Related articles
AI & Marketing·9 min di lettura
RAG and Graph RAG in marketing: what they are and why they matter
How an AI answers from your documents instead of “from memory”, why classic RAG falls short when the answer lies in the connections, and what a graph adds.
Read →
AI & Marketing·8 min di lettura
Vector embeddings: search by meaning, explained simply
What embeddings are, why they let you find information even without the right words, and where they reach their limits.
Read →
GEO & AI Search·10 min di lettura
What GEO (Generative Engine Optimization) is and how to do it
Getting found, cited and accurately described by AI assistants: how it differs from SEO, what the research says, what to do in practice and how to measure it.
Read →