AI Terms

Agentic AI describes AI systems that pursue a defined goal by planning steps, using tools and data, and carrying out tasks with a degree of autonomy – rather than only answering questions.

An AI agent is a software system that uses an AI model to pursue a goal by deciding on steps, using tools, and acting on the results – within limits set by its designers and users.

Generative AI refers to AI systems that create new content – such as text, code, images, or structured data – based on patterns learned from training data and the input they receive.

A hallucination is output from an AI model that sounds plausible but is factually wrong, unsupported by the provided sources, or invented – for example, a non-existent rule, figure, or reference.

Human-governed automation is an operating model in which AI agents carry out defined, repeatable tasks, while qualified people set the rules, review the results, and remain accountable for decisions.

A large language model (LLM) is an AI model trained on very large amounts of text to understand and generate language. It predicts likely continuations of text and can follow instructions, summarize, translate, reason through problems, and write code.

The Model Context Protocol (MCP) is an open standard for connecting AI applications to external tools, systems, and data sources in a consistent way. It was introduced by Anthropic in November 2024.

A multi-agent system is a setup in which several AI agents, each with a specific role, work together on a larger task – coordinating, sharing results, and checking each other's work.

Prompt engineering is the practice of designing the instructions and context given to an AI model so that it produces accurate, relevant, and consistent results.

Retrieval-augmented generation (RAG) is a technique in which an AI model first retrieves relevant information from a defined knowledge source and then uses that information to generate its answer.