Knowledge
What I learned and decided while running an AI transformation, sorted by job: topic pages with a position and sources, notes since 2023 and an AI glossary.
No match. Try another spelling or a German term.
Start here
Three routes through the site, depending on your job. Five stops each.
You run a company
Owners, executives, boards: where AI starts, what it costs, who owns it.
You run the operation
Process and IT owners: how AI runs day to day, gets checked and gets switched off.
You set the rules
Governance, law, knowledge: what gets checked before approval and what the law requires.
Topic pages
One page per topic, with a position, the reasoning and sources.
- Chief AI Officer: the role doesn't belong in ITA Chief AI Officer at a mid-market company needs the authority to change roles and workflows, or the job is software admin. Scope, limits and four checks.
- Mid-market AI: workflows before toolsMid-market AI starts with workflows. A subscription for everyone barely changes the work. Three questions to start, one shared foundation, cost per call.
- AI roadmap: foundation first, then the first use caseAn AI roadmap for business starts with a shared foundation, then use cases. Three questions set the order, five fields per project, one date to end it.
- AI automation: only as much leeway as you can checkAI automation should get only as much leeway as you can check. Five rules and six questions show whether a workflow can run without a human sign-off.
- EU AI Act for mid-sized firms: training and approval firstThe EU AI Act for mid-sized firms: AI literacy and transparency already apply, high-risk duties from December 2027. What deployers do now. As of November 2026.
- Measuring AI visibility: the source pool decidesMeasuring AI visibility starts with the source pool: for CRM, German content got no citations, for insurance it did. A twenty-minute GEO test and its limits.
- AI governance: the policy is the smallest partAI governance for mid-sized companies: the policy fits on one page. What works is access, permissions, cost per call and sign-off built into the workflow.
- AI knowledge management: AI can't search someone's headAI knowledge management starts with what's in people's heads and spreadsheets. Three questions show where to start; permissions and upkeep keep answers right.
- The transformation: four layers and one runHow the change is built: ways of working, roles, services, rules, plus a sample run through every checkpoint.
Recently revised
- AI answers by language: the market in the query decidesIn global categories, naming the market pulls in local sources for AI search, the language doesn't. What 15 Perplexity queries showed, plus a 20-minute check.
- Shutting down an AI system: keeping it is a decision tooShut down an AI system when usage, quality, upkeep or a better alternative say so. Four checks for every live system, and a shutdown order that strands nothing.
- Running AI: catch failures, respond under controlYou only catch AI failures if you check availability, output format and business quality separately. Responding under control means putting a limit on retries.
- AI cost management: know the cost per call or fly blindYour AI provider's invoice won't tell you which workflow is spending the money. How to track AI costs per call, budget each app and weigh cost against value.
- AI quality assurance: a glance is not a reviewCheck AI output in three separate passes: format, source, expert judgment. The sender stays responsible. With a review table and a rule for model switches.
- llms.txt: an overview with clear limitsAn llms.txt links a site's key pages with short descriptions for AI systems. It helps them find their way around, but it does not guarantee any citations.