Lesson 10 — The comparison lab.
You've got the map (Lesson 7), the privacy rules (Lesson 8), and you've clicked the menus (Lesson 9).
Now you run the same jobs on multiple apps and write down your winner. Not mine. Not a YouTube benchmark. Yours.
By the end (~60 minutes)
- Four tasks on fictional Example Ink (Leeds tattoo studio) — three chat, one optional image
- The same prompts pasted into ChatGPT, Claude, and Gemini (Grok optional)
- A winner log: which reply you'd actually use, and one line why
- Honest notes when free tier caps or missing features get in the way
Before you start
- Accounts: At least two of chatgpt.com, claude.ai, gemini.google.com. Grok optional.
- Cheat sheet: Your notes from Lesson 9 — you know where the buttons are.
- Lab rules: Fictional data only. Same scenario as Lessons 8–9. No real client names, files, or invoices.
- This is not: picking "the best AI in the world," installing local models, or choosing a membership tier.
Winner log preview
You'll fill this as you go (copy to Notes or paper):
| Task | ChatGPT | Claude | Gemini | Grok | Winner | One-line why |
|---|
The comparison lessons are where rented AI finally stopped being abstract for me. Same daft fake shop every time — boring for the model, safe for me — and I finally saw my own preferences instead of someone else's hype.
Continue — how to compare without turning into a fanboy.