AI Glossary: Plain-English definitions.
Every term you need to hold a conversation with any AI vendor, consultant, or technology article — without needing a translation.
AI (Artificial Intelligence)
Computer systems that can perform tasks that typically require human intelligence — like understanding language, recognising patterns, making decisions, and generating content. In business, AI refers to software tools that go beyond fixed rules to learn, adapt, and produce useful outputs.
Why it matters: AI is the umbrella category. Everything else in this glossary lives under it.
Algorithm
A set of instructions a computer follows to complete a task or solve a problem. In AI, algorithms learn from data to make predictions or decisions.
Why it matters: When someone says "the algorithm changed," they mean the rules the AI uses to decide what to show or do have been updated.
Automation
Using technology to perform tasks with minimal or no human involvement. AI automation goes beyond simple rule-based automation by handling tasks that require some level of judgement or pattern recognition.
Why it matters: Automation is the most immediate and measurable return on AI investment for most businesses.
Chatbot
A software program that simulates conversation with users — typically through text, but increasingly through voice. AI-powered chatbots can understand natural language and handle complex queries, not just scripted responses.
Why it matters: A well-built chatbot can handle your most common customer interactions 24/7 without staff involvement.
Claude
An AI assistant developed by Anthropic, known for its strong reasoning, long-context processing, and focus on safety. One of the leading large language models alongside ChatGPT and Gemini.
Why it matters: Claude is one of the tools you may use or encounter when building AI-powered business solutions.
Computer Vision
AI that can interpret and understand visual information — photos, video, documents. It can identify objects, read text, detect defects, and analyse scenes.
Why it matters: Used in construction site safety monitoring, retail inventory management, document processing, and quality control.
CRM (Customer Relationship Management)
Software that manages a company's interactions with current and potential customers. AI-enhanced CRMs can score leads, predict churn, automate follow-ups, and surface insights from customer data.
Why it matters: AI makes your CRM proactive rather than just a database — it tells you what to do next, not just what happened.
Data
Recorded information — numbers, text, images, transactions, interactions. AI learns from data. The quality and volume of your business data directly determines how useful AI can be for your specific needs.
Why it matters: Businesses with clean, well-organised data get dramatically more value from AI than those without it.
Deep Learning
A type of machine learning that uses multi-layered neural networks to learn from large amounts of data. The technology behind image recognition, voice assistants, and large language models.
Why it matters: Deep learning is what makes modern AI capable of tasks that used to require human perception.
Fine-tuning
The process of taking a pre-trained AI model and training it further on your specific data or use case. This makes the model more accurate and relevant for your particular business needs.
Why it matters: Fine-tuning is how you get an AI that knows your products, your industry language, and your customer patterns.
Generative AI
AI that creates new content — text, images, video, audio, code — rather than just analysing or classifying existing content. ChatGPT, Claude, Midjourney, and DALL-E are all generative AI tools.
Why it matters: Generative AI is the category most businesses start with because the outputs are immediately visible and useful.
Hallucination
When an AI confidently produces information that is inaccurate or entirely fabricated. A known limitation of large language models.
Why it matters: Always verify AI-generated content before publishing or acting on it — especially factual claims, statistics, or legal/financial information.
Integration
Connecting two or more software systems so they share data and work together. AI tools are most powerful when integrated with your existing business software.
Why it matters: An AI tool that sits in isolation provides limited value. Integration is what turns a tool into a workflow.
Large Language Model (LLM)
The type of AI that powers ChatGPT, Claude, Gemini, and similar tools. Trained on vast amounts of text, LLMs can understand, generate, and reason about language at a sophisticated level.
Why it matters: LLMs are the engine behind most of the AI tools businesses use for communication, content, and research.
Machine Learning (ML)
A subset of AI where systems learn from data to improve their performance over time without being explicitly programmed for each situation.
Why it matters: Machine learning is what allows AI tools to get better the more they're used — and why your historical business data has increasing value.
Model
The AI "brain" that has been trained on data to perform a specific task. Different models have different strengths — some are better at language, some at images, some at prediction.
Why it matters: When choosing an AI tool, you're often choosing which underlying model powers it — and that affects quality and capability.
Natural Language Processing (NLP)
AI's ability to understand and work with human language — reading, writing, and speaking in the way people naturally communicate.
Why it matters: NLP is why you can type a question in plain English and get a useful answer, rather than needing to write code.
Prompt
The instruction or question you give to an AI tool to get a response. The quality of your prompt directly affects the quality of the output.
Why it matters: Learning to write clear, specific prompts is the single most valuable AI skill for non-technical business users.
Prompt Engineering
The practice of designing and refining prompts to get the most useful outputs from AI tools. Not as technical as it sounds — more about clarity and specificity than coding.
Why it matters: The difference between a mediocre AI output and a great one is often just how well the prompt was written.
RAG (Retrieval-Augmented Generation)
A technique where an AI model is connected to a specific knowledge base — your documents, website, product catalogue — so it answers questions based on your information, not just its general training.
Why it matters: RAG is what allows you to build an AI assistant that knows about your specific business, products, and policies.
SaaS (Software as a Service)
Software delivered via the internet on a subscription basis rather than installed locally. Most AI tools are SaaS products.
Why it matters: SaaS pricing models (monthly subscriptions) make enterprise-grade AI affordable for SMBs.
Token
The unit AI models use to process text — roughly 3/4 of a word. AI models have a "context window" limit measured in tokens — how much text they can consider at once.
Why it matters: Relevant when choosing AI tools for processing long documents or complex analyses.
Workflow Automation
The use of software to automatically move tasks, data, and approvals through a defined process — replacing manual steps. AI-enhanced workflow automation can handle decision points, not just routing.
Why it matters: Workflow automation typically delivers the fastest measurable ROI of any AI application for service businesses.