AI agent

In its guidance on AI agents (in German), Germany's Federal Office for Information Security (BSI) describes them as systems that are meant not only to give answers but to act independently in the digital world. Unlike a chatbot, an agent works through the tasks users set in several steps, largely unaided, without a person confirming each step. One risk is loss of control: the more independently an agent acts, the harder it is to follow what it is doing. The advice is to give an agent only the access its task really needs.

The German business encyclopaedia Gabler Wirtschaftslexikon calls AI AI Artificial intelligence (AI) covers methods that let a computer perform tasks which would require intelligence if a person did them. Read in the glossary → agents autonomous systems that can perceive, decide and act towards a goal. Modern agents are integrated into technical and organisational processes through interfaces Interface An interface is the point at which two systems, programs or components meet and exchange data; the word also covers the user interface between a person and software. Read in the glossary → , for example to handle orders, support sales or write code.

Each agent we build fits the client's use case and works through defined interfaces such as a REST API API An API (application programming interface) is a defined set of functions and rules that lets one piece of software use the features or data of another system or service. Read in the glossary → or MCP MCP MCP, the Model Context Protocol, is an open protocol that connects applications built on large language models with external data sources and tools. Read in the glossary → tools. For a pharmaceutical logistics Pharma logistics Pharma logistics is the term we use for storing, transporting and delivering medicinal products and other pharma-sector goods, such as promotional material and supplies for patients. Read in the glossary → company we built a knowledge platform in which an email agent turns incoming emails into tasks and knowledge-base entries. It cannot delete anything, and the system logs everything it writes. Agents produce drafts; nothing is published until a person approves it. A task with fixed logic and a predictable output does not need an agent.