How we research and verify use cases
"Real" is a strong claim, so here is exactly what it means on this site.
What counts as a real use case
A use case makes it onto this site only when we find working professionals describing it themselves — in industry surveys, professional forums and subreddits, trade press, or published case studies. We don't invent plausible-sounding uses of AI and we don't publish a use case we couldn't source.
Every use case page links its sources at the bottom, under "Where this comes from."
How the prompts are built
The prompts are written by us, informed by what professionals report actually works. Each one states a role, gives context, defines the task and the output format, and uses fill-in variables for the details that change every time. Sample outputs shown on use case pages are illustrative excerpts, and are labeled as such.
What we tell you that prompt libraries don't
Every profession has ways AI can get you in trouble — FERPA for teachers, HIPAA for dentists, fair housing rules for realtors. Each use case documents the pitfalls specific to that field, because a prompt without its guardrails is a liability, not a shortcut.
Corrections
If you work in one of these professions and something on your page is wrong, outdated, or missing, we want to know. A correction from a practitioner outranks anything we can research.