# Artificial intelligence · Glossary · Zorraquino > Concepts for understanding and applying AI in products and processes. Includes language models, agents, information retrieval, evaluation and governance. URL: https://www.zorraquino.com/en/dictionary/category/inteligencia-artificial # Artificial intelligence 35 terms, explained one by one Concepts for understanding and applying AI in products and processes. Includes language models, agents, information retrieval, evaluation and governance. [The whole glossary](https://www.zorraquino.com/en/dictionary) ## A - [Agent memoryAgent memory brings together mechanisms for storing and retrieving information between the steps or interactions of an AI system.](https://www.zorraquino.com/en/dictionary/agent-memory) - [Agent orchestrationAgent orchestration coordinates tasks, tools, states and communications within a system with one or more agents.](https://www.zorraquino.com/en/dictionary/agent-orchestration) - [AI adoptionAI adoption is the process of incorporating artificial intelligence capabilities into organisational work in a useful, controlled and sustainable way. It includes selecting tasks, adapting processes, preparing data and people, and evaluating outcomes.](https://www.zorraquino.com/en/dictionary/ai-adoption) - [AI agentAn AI agent is a system that uses information and tools to decide on and carry out steps towards a goal, with a defined degree of autonomy.](https://www.zorraquino.com/en/dictionary/ai-agent) - [AI governanceAI governance establishes responsibilities, policies and controls for developing and using artificial intelligence systems.](https://www.zorraquino.com/en/dictionary/ai-governance) - [AI guardrailsAI guardrails are controls that limit or verify a system's inputs, responses and actions.](https://www.zorraquino.com/en/dictionary/ai-guardrails) - [AI hallucinationAn AI hallucination is generated output that presents false, invented or unsupported information as if it were valid.](https://www.zorraquino.com/en/dictionary/ai-hallucination) - [AI inferenceInference is the use of a trained model to produce an output from an input.](https://www.zorraquino.com/en/dictionary/ai-inference) - [AI model evaluationAI model evaluation checks whether a system meets quality and behaviour criteria on defined tasks.](https://www.zorraquino.com/en/dictionary/ai-model-evaluation) - [Algorithmic biasAlgorithmic bias is a systematic deviation in data, methods or results that can produce inappropriate effects or inequalities.](https://www.zorraquino.com/en/dictionary/algorithmic-bias) - [Artificial intelligenceArtificial intelligence brings together methods and systems that can carry out tasks such as recognising patterns, generating content or supporting decisions based on information.](https://www.zorraquino.com/en/dictionary/internet/artificial-intelligence.html) ## C - [ChatbotA chatbot is a system that holds a conversational interaction through text, voice or other means.](https://www.zorraquino.com/en/dictionary/digital-marketing/chatbot.html) - [Context engineeringContext engineering designs what information, instructions and resources an AI system receives at each moment.](https://www.zorraquino.com/en/dictionary/context-engineering) - [Context windowThe context window is the amount of information a model can take into account within an interaction, according to its processing limits.](https://www.zorraquino.com/en/dictionary/context-window) ## E - [EmbeddingAn embedding is a numerical representation of data that makes it possible to compare learned relationships between texts, images or other items.](https://www.zorraquino.com/en/dictionary/embedding) ## F - [Fine-tuningFine-tuning adjusts the parameters of an already trained model using additional data to adapt its behaviour to a task or domain.](https://www.zorraquino.com/en/dictionary/fine-tuning) - [Foundation modelA foundation model is a model trained on broad data that can be adapted or used for a range of tasks.](https://www.zorraquino.com/en/dictionary/foundation-model) ## G - [Generative AIGenerative AI produces content, such as text, images, audio or code, from learnt patterns and the instructions it receives.](https://www.zorraquino.com/en/dictionary/generative-ai) - [GroundingGrounding is the linking of an AI response or action to reference information available in the context.](https://www.zorraquino.com/en/dictionary/grounding) ## H - [Human-in-the-loopHuman-in-the-loop builds human intervention into certain decisions or stages of an automated system.](https://www.zorraquino.com/en/dictionary/human-in-the-loop) ## L - [LLMAn LLM, or large language model, is a machine learning model trained to process and generate language at scale.](https://www.zorraquino.com/en/dictionary/llm) ## M - [Machine learningMachine learning develops models that learn patterns from data in order to carry out tasks.](https://www.zorraquino.com/en/dictionary/digital-marketing/machine-learning.html) - [MCPMCP, or Model Context Protocol, is a protocol for connecting AI applications to tools and information sources through common interfaces.](https://www.zorraquino.com/en/dictionary/mcp) - [Multi-agent systemA multi-agent system combines several agents that interact to tackle a task or a shared environment.](https://www.zorraquino.com/en/dictionary/internet/multi-agent-system.html) - [Multimodal modelA multimodal model processes or generates more than one type of information, such as text, images, audio or video.](https://www.zorraquino.com/en/dictionary/multimodal-model) ## P - [PromptA prompt is the input that guides the response or action of a generative AI system.](https://www.zorraquino.com/en/dictionary/prompt) - [Prompt engineeringPrompt engineering is the practice of designing, testing and refining instructions to obtain suitable results from a generative model.](https://www.zorraquino.com/en/dictionary/prompt-engineering) - [Prompt injectionPrompt injection is an attempt to make an AI system follow instructions inserted into content that it should treat as data.](https://www.zorraquino.com/en/dictionary/prompt-injection) ## R - [RAGRAG combines the retrieval of external information with a generative model to produce answers based on relevant content.](https://www.zorraquino.com/en/dictionary/rag) ## S - [Semantic searchSemantic search retrieves information based on relationships of meaning, as well as literal word matches.](https://www.zorraquino.com/en/dictionary/semantic-search) - [Small language modelA small language model, or SLM, is relatively small in scale compared with the other models it is being compared against.](https://www.zorraquino.com/en/dictionary/small-language-model) - [Synthetic dataSynthetic data is generated artificially to represent the characteristics or cases of a data set.](https://www.zorraquino.com/en/dictionary/synthetic-data) ## T - [Token (AI)A token is a unit of representation that a model uses to process information, such as a word, part of a word or a symbol.](https://www.zorraquino.com/en/dictionary/ai-token) - [Tool use in AITool use allows a model to request external operations, such as looking up information or running an authorised function.](https://www.zorraquino.com/en/dictionary/ai-tool-use) ## V - [Vector databaseA vector database stores numerical representations and makes it possible to retrieve items by similarity.](https://www.zorraquino.com/en/dictionary/vector-database)