The 2026 AI landscape

Free
0/10 steps visited

Step 8 of 10

Try it — upgrade one prompt

Do it · ~5 min

Try it — upgrade one prompt

Reading about prompts is useless until you feel the difference. Five minutes — do it now, properly.

The task

Open any AI chat you already use — ChatGPT, Claude, Gemini, or Gertrude on this site.

Step 1 — Send the weak prompt

Copy this exactly:

Write an email about a late payment.

Read the reply. It's fine. It's also probably generic and a bit limp.

Step 2 — Send the upgraded prompt

Now copy this:

You are a small-business owner in the UK. Write a short email to a customer who is 2 weeks late paying an invoice for tattoo work. Tone: firm but polite, not threatening. Under 120 words. End with a clear next step (how to pay). Do not invent account numbers or legal threats.

Read the new reply.

You should see: a far more specific email — right length, right tone, fewer random inventions — because you handed it role, context, task, format and constraints.

That's the whole anatomy of a useful prompt. We practise it properly in later lessons; today you only need to clock one thing: you steered the pattern. Same app, better instructions.

This was my own first "oh — that's how it works" moment. Not a new model, not a fancier subscription tier. Same tool, sharper instructions. I still rewrite a prompt twice when something actually matters.

Optional stretch

Change one line of the upgraded prompt — the industry, the tone, or the word limit — and send it again. Watch exactly what moves.

Continue once you've tried both prompts (or you're happy you've got the idea).

Warning

Real power. Educational use only.

What we teach you to build is genuinely powerful — uncensored assistants, agents, and automations on your own hardware. In the wrong hands, that is as dangerous as malicious code in the wrong hands. We do not teach illegal, malicious, or harmful use. You are responsible for what you deploy.

See what we mean →