Reference
Business-friendly definitions of the AI terms you'll hear in vendor pitches, board meetings, and bad LinkedIn posts.
28 terms
4 terms
Technology that lets computers do things that usually require human intelligence.
Why it matters · Helps businesses automate work, make predictions, and improve decisions.
A subset of AI where computers learn from data instead of being explicitly programmed.
Why it matters · Used for things like customer behaviour prediction, fraud detection, and process optimization.
A type of machine learning that uses many-layered neural networks to find complex patterns.
Why it matters · Powers image recognition, natural language understanding, and most modern AI systems.
A computing system inspired by biological brains — interconnected nodes that process information.
Why it matters · The underlying architecture for most AI models you'll encounter.
4 terms
Datasets large enough that traditional tools struggle to handle them.
Why it matters · Enables data-driven decisions and insights you can't see in spreadsheets.
Discovering patterns and relationships in large datasets.
Why it matters · Surfaces customer preferences, market trends, and operational inefficiencies.
Using historical data to predict future events or outcomes.
Why it matters · Used for sales forecasting, customer churn prediction, and demand planning.
A series of processes that collect, transform, and deliver data to its destination.
Why it matters · Ensures data flows reliably from your source systems into the AI applications that use it.
4 terms
AI that understands, interprets, or generates human language.
Why it matters · Powers chatbots, sentiment analysis, automated document processing, and more.
An AI-powered conversational agent that interacts via text or voice.
Why it matters · Provides 24/7 support, handles common questions, and improves response time.
Determining whether a piece of text expresses positive, negative, or neutral sentiment.
Why it matters · Useful for understanding customer satisfaction and brand perception from reviews or social posts.
Extracting meaningful information and insights from unstructured text.
Why it matters · Used to analyse customer feedback, support tickets, and any free-form text at scale.
4 terms
AI that interprets visual information from images or video.
Why it matters · Used for quality control, visual inspection, and any task that involves "looking" at something.
Identifying and classifying objects, people, or scenes in images.
Why it matters · Applied in manufacturing QC, retail analytics, and security systems.
Locating and identifying multiple objects within an image or video.
Why it matters · Used in autonomous vehicles, inventory management, and security monitoring.
Identifying or verifying individuals based on facial features.
Why it matters · Used for access control, customer identification, and personalised experiences.
4 terms
Using technology to perform tasks without human intervention.
Why it matters · Reduces costs, improves consistency, and frees up people for higher-value work.
Improving business processes to increase efficiency and reduce cost.
Why it matters · AI analyses workflows, identifies bottlenecks, and suggests changes.
Tailoring products, services, or experiences to individual preferences.
Why it matters · Drives engagement, satisfaction, and conversion.
AI that suggests products, content, or services based on past behaviour.
Why it matters · Increases sales, lifetime value, and the perceived helpfulness of your product.
4 terms
A set of rules or instructions an AI system follows to solve a problem.
Why it matters · The "recipe" that tells the AI how to process data and produce output.
Teaching an AI model to recognise patterns by feeding it data.
Why it matters · Requires quality data and meaningful time — minutes to weeks depending on the model.
A standard way for different software systems to communicate.
Why it matters · Lets you integrate third-party AI services into your existing tools.
Delivering computing services — including AI — over the internet.
Why it matters · Makes AI accessible without buying or maintaining your own infrastructure.
4 terms
Principles and guidelines for responsible AI development and deployment.
Why it matters · Ensures AI systems are fair, transparent, and beneficial — not just functional.
Systematic prejudice in AI outputs that can lead to unfair or discriminatory results.
Why it matters · Must be actively identified and mitigated — it doesn't go away on its own.
The degree to which AI decisions can be understood and explained.
Why it matters · Builds trust with users and is increasingly required by regulation.
Protecting personal information and ensuring it's used appropriately.
Why it matters · Critical for compliance and for maintaining customer trust at all.