Wonder, together.

Bring any question about AI, even the one that feels too simple to ask. You can handle the answer.

I’m Dennis. I explain how AI works in plain words, test what it can do, and show you where it stops.

AI can make your thinking sharper, more playful and broader than it would have been without it.

That only happens when it adds to your life instead of standing in for it.

I use it every day, and it still answers in the same confident voice whether it knows or not. So I test things before I pass them on, and I say so when I got it wrong.

We were never taught how to question an answer that sounds that sure. That is the part we can practice together.

You already have the part that matters: your own judgment. Everything here is built to give you more chances to use it.

What you’ll find here

Four questions I keep coming back to. Everything I make fits one of them.

  • How it works

    What actually happens when you ask something: it predicts the next word, again and again. Once you see that, most of the mystery goes.

  • How to use it

    The habits that change the result: what you give it, what you ask for, and what you keep doing yourself. Small things you can try today.

  • Where it breaks

    Wrong answers in a confident voice, viral claims put to the test, and the limits that rarely make it into the demo.

  • Where it fits

    What it changes in a normal week, at work and at home. And the parts of your life worth keeping for yourself.

Three small experiments

No model runs on this page. Everything here is prepared by hand, so the page stays quick and nothing you type ever leaves your browser.

Guess the next word

This is the whole trick behind a language model: it picks a likely next word, then does it again. Try it once and it stops being magic.

She poured the coffee and looked out the blank

It never checks whether the sentence is true. That part stays with you.

The list appears here as soon as you guess.

Sharpen a prompt

It cannot see your situation. Whatever it needs to know, you have to put in. Switch the parts on and watch the ask get clearer.

Parts you can add to the prompt

Write me a cover letter.

None of this is a trick. It is telling it what you already know.

Let it, or keep it?

There is no score here. The only question is what you want to hand over.

  • Turning your messy notes into a summary

  • The message that ends an argument with someone you love

  • Understanding the letter your doctor sent you

Where the line sits is your call, and it moves with practice.

How I think about AI.

Placeholder: the film is not shot yet. The frame holds its place so the page keeps its rhythm.

What I am building

Both are unfinished. The waitlist is how I find out whether they are worth finishing.

Coming soon

Trust yourself more

The AI guide. Working title.

For everyone who has to use AI and would rather understand it than be impressed by it. Plain words, small steps, nothing you have to take on faith.

Inside
[Platzhalter: drei bis fünf Inhalte, die feststehen]
Format
[Platzhalter: PDF, E-Mail-Kurs oder Video]
Planned
End of 2026

People on the waitlist get it first, at the launch price.

Join the waitlist

Coming soon

The creator guide

Working title. Comes after the AI guide.

Built from the analysis of 79 posts, tested on my own start in public. How attention actually works on a feed, taken apart piece by piece, and whether the same playbook holds when I run it myself.

Format
[Platzhalter]
Planned
[Platzhalter: Zeitfenster]
Join the waitlist

Join the waitlist

One email when a guide is ready, and the launch price for the people who waited. Nothing else, and you can leave with one click.

You will get one confirmation email first. Until you click it, you are not on the list.

Placeholder: the list is not connected yet. [Platzhalter: MailerLite]

Dennis Herold
Placeholder portrait, a proper one is coming.

Hey, I’m Dennis.

I am building this in public, from zero. No audience to point at yet, no numbers to wave around. What I do have is the question that got me started: why do so many capable people feel small next to a machine that guesses?

What I am up against is not the technology, and it is not the people who are sceptical about it. It is the fear of asking the obvious question, the hype that sets everyone up for disappointment, and the quiet habit of handing over our own thinking. None of that gets better by being impressed.

So what I want for you is simple and hard to fake: that you walk away a little more capable than you arrived. Not because I told you what you wanted to hear, but because you tried something and it worked in your hands.

I am ambitious about this. I want to reach a lot of people, and I want to sell the guides I am making. I would rather say that plainly than dress it up, and it changes nothing about the free part: every post carries a whole thought, not half of one.

If you end up asking better questions than I do, this worked.