AI Fundamentals
AI Terminology Glossary
A simple guide to common AI terms you might encounter while reading our guides
This glossary explains AI terminology in simple, everyday language, organized by complexity level. We’ve focused on terms you’re most likely to encounter while reading our guides.
Essential Terms
Essential terms you’ll encounter in our guidesAI Assistant
AI Assistant
A specialized AI system designed to help with tasks through natural conversation, such as when you chat with Omnifact to get help with writing an email or analyzing a document.
Prompt
Prompt
The question, instruction, or request you give to the AI—like when you type “Please help me summarize this report.”
Response
Response
The AI’s answer or output based on your prompt, such as the summary, analysis, or content the AI creates for you.
Context
Context
Background information that helps AI understand your request better, for example, when you explain “This is for a team meeting next week” or attach relevant documents.
Omnifact-Specific Terms
Terms unique to the Omnifact platformPrivacy Filter
Privacy Filter
Omnifact’s automatic system that protects sensitive information by detecting and masking data like personal information, financial details, or company secrets. It also restricts images from being sent to non-EU hosted AI models for compliance. For example, if you paste an email containing employee names, they’re automatically masked before processing. Learn more →
Space
Space
A focused environment in Omnifact for specific topics or projects, including a dedicated knowledge base, customized AI behavior, and team access settings. For example, you might create a “Product Launch” Space with all relevant documents and team access. Learn more →
Knowledge Base
Knowledge Base
A collection of documents within a Space that the AI can reference, such as company documents, project materials, or team resources. The AI uses these to answer questions, like when it references your uploaded product documentation. Learn more →
Chat Instructions
Chat Instructions
Guidelines that tell the AI how to behave in a specific Space, controlling things like response style, specialized knowledge, and behavior patterns. For example, you might set instructions for the AI to use technical language in an engineering Space.
Common AI Concepts
Basic AI termsArtificial Intelligence (AI)
Artificial Intelligence (AI)
Technology that enables computers to perform tasks that typically require human intelligence, such as understanding text, answering questions, analyzing information, and generating content. For example, Omnifact understands your request and provides a helpful response.
Query
Query
Another word for your question or request to the AI. For example, asking “What are the sales figures for Q1?” is a query.
Agent
Agent
A task-focused automation within Omnifact that can perform specific actions, like sending data to another app or executing a workflow.
Training Data
Training Data
The collection of text, documents, and examples that the AI learns from in order to generate its responses.
Inference
Inference
The process where the AI generates an answer based on what it has learned without “retraining” each time you ask a question.
Fine-Tuning
Fine-Tuning
Teaching an AI to perform better on specific tasks by training it further on specialized data, like company-specific documents.
Retrieval-Augmented Generation (RAG)
Retrieval-Augmented Generation (RAG)
A method where AI combines searching external documents (retrieval) with generating new answers, improving accuracy and relevance.
Embedding
Embedding
A mathematical representation of text that captures its meaning, allowing AI to search and relate concepts even if different words are used.
Latency
Latency
The small delay between when you send a prompt and when the AI responds. Lower latency means faster replies.
Overfitting
Overfitting
When an AI model learns too much from training data, including noise and errors, making it less flexible or accurate with new data.
Machine Learning
Machine Learning
How AI systems improve their performance through experience by recognizing patterns, processing data, and getting better at tasks over time. For instance, AI gets better at understanding different ways people ask similar questions.
Natural Language
Natural Language
Everyday human language—like English, Spanish, or Mandarin—that people use to communicate. When you type or speak to Omnifact, you’re using natural language, as opposed to computer code or technical commands. AI systems are designed to understand and generate natural language so you can interact with them just like you would with another person.
Natural Language Processing (NLP)
Natural Language Processing (NLP)
The ability of AI to understand and generate human language, enabling it to process text, understand questions, generate responses, and maintain conversations. For example, you can ask questions in plain English and get coherent responses.
Advanced Concepts
Deeper technical terms you might encounter occasionallyLarge Language Model (LLM)
Large Language Model (LLM)
The type of AI system that powers Omnifact and similar assistants. These models are trained on vast amounts of text, understand and generate language, and process information contextually—enabling Omnifact to understand and respond to your requests.
Pattern Matching
Pattern Matching
How AI identifies similarities and connections in information to find relevant details, understand context, and generate responses. For example, AI recognizes that questions asked in different ways are asking for the same information.
Hallucination
Hallucination
When AI generates incorrect information that sounds plausible. This can produce misleading information, so it needs human verification and fact-checking. For example, AI might confidently state something that sounds reasonable but isn’t factually correct.
Technical Terms Made Simple
Complex terms explained simplyAlgorithm
Algorithm
A set of rules the AI follows to process information—think of it as a recipe the AI uses to complete tasks, like the steps it follows to understand your question and create a response.
Neural Network
Neural Network
A complex system that helps AI process information, similar to the AI’s “brain” for connecting different pieces of information and understanding context.
Token
Token
The small pieces of text that AI processes—these are the building blocks of language that AI understands, breaking down your message into processable pieces.
Remember
You don’t need to memorize all these terms to use Omnifact effectively. This glossary is here as a reference when you encounter unfamiliar terms.
Keep It Simple
Focus on using Omnifact to help with your work - you don’t need to be an AI expert
Ask Questions
If you encounter terms you don’t understand, feel free to ask Omnifact to explain them
Next Steps
- Return to How to Interact with AI to put these terms into practice, or explore Capabilities and Limitations to understand what AI can do
- Learn about Knowledge Base Retrieval and Chat Instructions to get the most out of Omnifact’s features
- Review Privacy & Security to understand how Omnifact protects sensitive information