Glossary

AI Glossary

The key terms around AI agents and automation – briefly explained and clearly put into context.

Fundamentals

AI systems that plan and execute multi-step tasks autonomously, rather than producing single answers.

A software system that plans and carries out tasks on its own, using tools and data along the way.

A numeric representation of text that makes semantic similarity measurable.

Further training a model on your own data to adapt it to a specific task.

An AI model trained on large amounts of text that understands and generates language.

Coordinating multiple steps, tools and agents into one reliable workflow.

The instruction or question that directs a language model toward a task.

A method where a model retrieves relevant documents before answering and grounds its response in them.

The smallest unit of text a language model breaks input into – usually word fragments.

A database that stores embeddings and searches them for semantic similarity at high speed.

A model's ability to deliberately call external functions and systems.

The technical connection of an AI agent to existing systems such as ERP, CRM or email.

Having software perform recurring tasks on its own instead of people doing them by hand.

Technology that turns printed or scanned text into machine-readable data.

A central collection of company knowledge that AI agents can draw on.

A plausible-sounding but incorrect or fabricated answer from a language model.

A principle where people review and approve a critical decision made by an AI system.

In a short conversation we find the use case with the biggest leverage in your business – with a clear goal and a measurable result.