My best AI workshop rule is never saying the word AI

My best AI workshop rule: we never say the word "AI".We only talk about real problems people feel today.

A session starts with sticky notes, not slides.

People write what hurts in daily work:• "We lose hours every day on manual checks."• "No one trusts the numbers from system X."• "Our process breaks each time one field is missing."

No models.No agents.No buzzwords.

We map the negative consequences, like:• Where time leaks away• Where mistakes sneak in• Where people no longer trust the system

Only after having prioritized the biggest inefficiencies and gains - and only then, we ask a different question:

If a smart assistant sat inside these workflows, under your control, what would it need to do so you use it every time?

People answer in very concrete terms:• "Flag missing data before I click send."• "Summarize these 20 pages for my task, not in general."• "Explain strange cases in clear language, with links to raw data."

Now the first step is completed and we can go deeper. Understanding the process end-to-end with stakeholders, inputs and outputs, quality rules and existing systems.

Up to now this is not an "AI project".It's a process improvement project that might or might not use AI inside the larger system.

It's a win-win.

Executives hear less noise about magic.Teams see a tool that respects their reality.

Trust grows not from the model, but from the way we tie it to problems people can already feel in their daily work.