Knowledge
A new AI hype every week. What stays?
Hypes come and go. What stays are people who build something for their own work. And you turn a fun experiment into something that lasts together.
Yourself as an action figure in a box. A baby presenting a podcast. A shark in sneakers.
At work it moves just as fast. Copilot helps write your e-mail. An AI note-taker writes the minutes of your meeting. And with “vibe coding” you describe the app you need, and AI builds a first version.
I love everything that is possible. But from business owners, in care and at municipalities I keep hearing the same question: how do we really get something out of it, without chasing every hype?
I believe the answer is not with the big tech companies, but with your own people. They do the work every day. They know where it can be better. And with AI they can more and more often do something about it themselves.
Look back one year
How many of those hypes do you still use? Most are already forgotten. Fun, but they changed little about your work.
Ask AI to make the same photo ten times and you get ten different photos. Fine for a post. But for a quote, a care plan or a decision you want it to be right, every time.
What stayed is your organisation: your people, your customers, your software. And one thing from all those hypes stays too: people start to make software themselves.
What stays: people who build something themselves
A dashboard of their own, a small app for the team, a tool for a quote. With AI it gets easier and easier to turn an idea into something you can try.
A customer e-mails: “Can you deliver 400 units before 1 December?” Your colleague in sales opens the stock system, calls purchasing and searches an Excel list from last week. Half an hour later she says “yes”, and hopes it is right.
So she builds something herself: one screen with the stock, what is on order and when it arrives. Now she sees in a minute: 300 in stock, 100 arrive on 24 November. She says “yes”, and knows why.
A tool like that is not a problem. It is a signal: your software is missing something here.
A week later
Her colleagues want it too. And then the questions come. Is the stock still right when that Excel list is not updated? Who may see the purchase prices? And what if she is on holiday?
This is where most home-made tools get stuck. Or they keep running on one laptop, and nobody knows exactly how.
From a first version to something that lasts
Take the next step together. Sales points out what is missing. Purchasing knows why a delivery sometimes moves. IT connects it safely to what is already there.
And set up the foundation well, once: who may see what, where the data comes from and who looks after it. You try a hype once. You use a foundation again with every next idea. The next idea then does not start from zero.
Three questions for tomorrow
Does someone where you work already have their own list, dashboard or app? Sit down next to them and ask:
- What does it solve? What is missing in the software we have?
- Who else has this problem? A colleague, another team, the customer?
- What if you are away for a week? Can someone else carry on with it?
The answers show where it chafes. Then you see what stays when the next hype is over: your people, and what they build together.
Which hype did you let go? And which tool would you like to make?


