Guides · AI

AI

Model selection, local AI, RAG, governance and responsible operations.

Specialist articles

56 in-depth answers in this topic hub

Every article starts with a concise answer, identifies its author and review date, and links to sources and practical next steps.

The topic hub grows around concrete decisions. Each article follows a consistent structure covering context, approach, risks, measurements, a checklist and related paths.

TopicSort

Guides

56 in-depth specialist articles

9 of 56 visible

Räumliche Visualisierung eines neuronalen Netzes

AI · 15 minEvaluation of generative AI: operate reliably over the long termEvaluation of generative AI leads to comprehensible model decisions based on real specialist cases if monitoring, maintenance, responsibility and restart are clarified before tool selection. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

AI · 13 minRAG assistants with company knowledge: develop a viable conceptRAG assistants with company knowledge leads to proven answers from approved internal sources when the target image, roles, boundaries and verifiable assumptions are clarified before choosing the tool. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Nico Freitag · reviewed 11/08/2026

AI · 12 minRAG assistants with company knowledge: plan costs and procurement realisticallyRAG assistants with company knowledge leads to proven answers from approved internal sources when total costs, internal performance, dependencies and exit are clarified before choosing a tool. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

AI · 14 minEvaluation of generative AI: implement it safely and graduallyEvaluation of generative AI leads to comprehensible model decisions based on real technical cases if the pilot, releases, fallback path and gradual introduction are clarified before the tool is chosen. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

AI · 12 minAI-supported process automation: reliably capture the current stateAI-powered process automation leads to faster processes with controlled human decisions when inventory, dependencies and reliable starting values are clarified before tool selection. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

AI · 11 minEvaluation of generative AI: make informed decisionsEvaluation of generative AI leads to comprehensible model decisions based on real technical cases if requirements, exclusion criteria and decision-making scope are clarified before choosing the tool. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

AI · 12 minEvaluation of generative AI: reliably capture the current stateEvaluation of generative AI leads to comprehensible model decisions based on real specialist cases if inventory, dependencies and reliable initial values are clarified before tool selection. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

AI · 15 minAI-supported process automation: the complete practical checklistAI-powered process automation leads to faster processes with controlled human decisions when responsibility, technology, evidence and next steps are clarified before tool selection. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Nico Freitag · reviewed 11/08/2026

AI · 12 minLocal language models: Plan costs and procurement realisticallyLocal language models lead to controlled AI processing in your own operating room if total costs, internal performance, dependencies and exit are clarified before choosing the tool. The practical start consists of a limited scope of application, designated responsible parties, measurable baseline values and a fallback path. What is important is not the amount of technology used, but whether benefits, risks and operation can be proven together.Kevin Kröger · reviewed 11/08/2026

From knowledge to implementation

Assess the answers and apply them directly

The service applies the topic to your situation. The checklist helps prepare the first meeting or internal workshop.

View service Open checklist

Is a specialist question missing from this hub?

Send a question