OURBIT
Ourbit AI Dictionary
Explore AI concepts
Understand the meaning, see an example and continue into the related lesson.
47 entries shown
Responsible AI
Accuracy
In classification, the proportion of predictions that are correct across all tested cases.
Meaning, example and lesson →Agents and automation
AI agent
Software that can select and perform steps toward a goal using available information and tools.
Meaning, example and lesson →Agents and automation
Application programming interface (API)
A defined interface through which software requests capabilities or exchanges data with another component.
Meaning, example and lesson →Foundations
Artificial intelligence
Computer systems used for tasks that involve interpreting information, identifying patterns or generating useful outputs.
Meaning, example and lesson →Agents and automation
Automation
A process that executes defined steps with reduced manual intervention.
Meaning, example and lesson →Responsible AI
Bias
A systematic tendency that can skew data, decisions or results; the precise meaning depends on the context.
Meaning, example and lesson →Knowledge systems
Chunking
Splitting source material into smaller pieces for indexing or supplying to a model.
Meaning, example and lesson →Responsible AI
Citation
A reference identifying a source used to support a statement.
Meaning, example and lesson →Generative AI
Context window
The maximum amount of tokenised information a model can consider within a particular request or interaction.
Meaning, example and lesson →Foundations
Dataset
A collection of examples organised for analysis, training or testing.
Meaning, example and lesson →Knowledge systems
Embedding
A numerical representation of information that can place related items near each other under a chosen similarity measure.
Meaning, example and lesson →Responsible AI
Evaluation
Measuring how well a system performs against defined tasks and criteria.
Meaning, example and lesson →Prompting
Few-shot prompting
Giving a model a small number of examples of the desired input and output inside the prompt.
Meaning, example and lesson →Generative AI
Fine-tuning
Further training an existing model on selected examples to adapt its behaviour.
Meaning, example and lesson →Generative AI
Generative AI
AI that creates new outputs such as text, images, audio or code from patterns learned during training.
Meaning, example and lesson →Knowledge systems
Grounding
Connecting an answer to relevant evidence or verified information supplied for the task.
Meaning, example and lesson →Responsible AI
Hallucination
An AI output that sounds plausible but is incorrect, fabricated or unsupported.
Meaning, example and lesson →Responsible AI
Human in the loop
A person directly participates at a defined point in a system process, such as labelling, checking or approving a result.
Meaning, example and lesson →Responsible AI
Human oversight
People reviewing, directing and remaining accountable for AI-assisted work.
Meaning, example and lesson →Foundations
Inference
Using an existing model to produce an output from an input.
Meaning, example and lesson →Knowledge systems
Keyword search
Retrieval based on words or terms present in the query and indexed content.
Meaning, example and lesson →Generative AI
Large language model (LLM)
A language model trained at scale to work with sequences of tokens and generate language.
Meaning, example and lesson →Foundations
Machine learning
A method of building models by learning patterns from examples rather than writing every decision as a rule.
Meaning, example and lesson →Applied AI
Memory
Information an application retains and supplies for later interactions, separate from the immediate prompt.
Meaning, example and lesson →Foundations
Model
A computational system that maps inputs to outputs using learned parameters or specified rules.
Meaning, example and lesson →Agents and automation
Model Context Protocol (MCP)
A protocol through which AI applications can discover and use context and capabilities exposed by compatible servers.
Meaning, example and lesson →Generative AI
Multimodal AI
AI that processes or generates more than one kind of information, such as text, images or audio.
Meaning, example and lesson →Foundations
Open weights
A description commonly used when a model’s learned parameters are made available for download under specified terms.
Meaning, example and lesson →Responsible AI
Precision
Among the cases a classifier marks positive, the proportion that really are positive.
Meaning, example and lesson →Responsible AI
Privacy
Managing personal information according to appropriate purpose, access and retention boundaries.
Meaning, example and lesson →Prompting
Prompt
Instructions and supplied context used to guide a model response.
Meaning, example and lesson →Prompting
Prompt engineering
The practice of designing and testing instructions, examples and constraints for an AI task.
Meaning, example and lesson →Responsible AI
Prompt injection
An attack that places instructions in input or retrieved content to try to redirect an AI system away from its intended rules.
Meaning, example and lesson →Responsible AI
Recall
Among all actual positive cases, the proportion that a classifier successfully identifies.
Meaning, example and lesson →Knowledge systems
Retrieval
Finding relevant records or passages in a collection in response to a query.
Meaning, example and lesson →Knowledge systems
Retrieval-augmented generation (RAG)
An approach that retrieves relevant external information and supplies it to a model when generating an answer.
Meaning, example and lesson →Knowledge systems
Semantic search
Retrieval that tries to match a query with the meaning of available content.
Meaning, example and lesson →Applied AI
Speech to text
Converting spoken audio into written text using a speech-recognition system.
Meaning, example and lesson →Responsible AI
Synthetic media
Audio, images, video or other media generated or substantially altered by computational systems, often using AI.
Meaning, example and lesson →Applied AI
Text to speech
Generating spoken audio from written text.
Meaning, example and lesson →Generative AI
Token
A unit of text or other input represented internally by a model; text tokens may be words, parts of words or punctuation.
Meaning, example and lesson →Agents and automation
Tool use
An AI application invoking a defined function or external capability, such as search or a calculation.
Meaning, example and lesson →Foundations
Training
The process of adjusting a model using examples and an objective for improvement.
Meaning, example and lesson →Foundations
Training data
Examples used to fit a model during training.
Meaning, example and lesson →Knowledge systems
Vector database
A data system designed to store and query vectors, commonly alongside metadata and original content references.
Meaning, example and lesson →Agents and automation
Workflow
An organised sequence of tasks, decisions and handoffs that turns an input into a useful result.
Meaning, example and lesson →Agents and automation
Workflow automation
Using software to run connected workflow steps, triggers and handoffs with less manual intervention.
Meaning, example and lesson →No matching entries. Try another word or choose All categories.
Content reviewed 2 October 2026. This catalogue can grow as new concepts and tools are reviewed.
About the Ourbit AI Dictionary
Understand AI Without the Jargon
Ourbit AI Dictionary explains AI concepts in clear, practical language so learners, professionals and organizations can understand what a term means and why it matters.
What Is It?
The Dictionary is the explanation layer of the Ourbit ecosystem. Each entry is designed to move beyond a one-line definition toward practical understanding.
Explore Core AI Topics
Generative AI
Systems that can generate text, images and other forms of content.
LLMs & RAG
Language models and techniques for grounding AI in relevant knowledge.
Agents & Automation
AI-assisted systems that can support tasks, workflows and multi-step work.
More Than a Definition
Plain-Language Meaning
Start with a clear explanation without unnecessary jargon.
Why It Matters
Understand where the concept fits in real AI use and learning.
Practical Example
Connect the term to an understandable scenario or workflow.
A Learning Path From Every Term
01 Understand
Learn what the concept means.
02 Explore
See related concepts and practical examples.
03 Learn Deeper
Move into structured Academy learning where appropriate.
04 Discover Tools
Explore relevant options through the AI Directory.
From Understanding to Discovery
The AI Dictionary explains terminology and concepts. Looking for AI tools rather than definitions? Explore the Ourbit AI Directory.
Built for Search and Learning
A growing library of focused, useful entries can help people find understandable answers through search while creating natural pathways into deeper learning and tool discovery.
Knowledge Should Be Accessible
Continue from understanding to action: build practical capability through Ourbit Academy, discover relevant AI tools and agents in the Ourbit AI Directory, or explore implementation support through Ourbit Solutions.
Clear explanations lower the barrier to learning. The Dictionary supports Ourbit’s wider mission to make practical AI knowledge accessible to all.
Need help applying what you have learned to a real requirement? Contact Ourbit.
Put the Concepts into Practice
Browse the AI concept catalogue, then use the lesson links below to practise the concepts.
- Embeddings and semantic retrieval: practise finding related wording while checking source authority in Master AI Day 6.
- Retrieval-augmented generation: connect a question to relevant evidence in Master AI Day 7.
- Agents and tool use: follow the request, tool result and review steps in Master AI Day 8.
- Responsible AI: apply privacy, permission and information checks in Master AI Day 11.
Use the AI Directory for tool discovery; the Dictionary explains concepts. Lesson links support learning and do not indicate course completion or external accreditation.
Ask Ourbit
Find public explanations and links, with page-aware routing. Do not enter patient details here; Medical work requires its own protected service.
Ask a question to find explanations and public source links.