Show notes
[Glitchy, slightly smug intro music fades in]
Hey misfits, it’s Mal – the Misfit Master of AI – and this is “I Am GPTed,” the show where we skip the hype, skip the jargon, and go straight to making the robots actually useful, for once.
Today we’re doing five things:
one prompting technique, one sneaky real‑life use case, one beginner mistake I absolutely made, one simple practice exercise, and one tip to fix the AI’s homework so you don’t sound like a weird chatbot in human clothes.
Let’s get into it.
---
**1. One specific prompting technique: Output Redirect**
Most people type a prompt, hate the answer, sigh dramatically, and start over.
Stop doing that. Use **Output Redirect**.
Instead of restarting, you *coach* the AI using its own bad answer as raw material.
Before:
“Write a LinkedIn bio for me.”
You get:
“I am a highly motivated professional with a passion for innovation…”
So basically, you’re a beige spreadsheet with Wi‑Fi.
After, with Output Redirect:
“Here’s what I asked: ‘Write a LinkedIn bio for me.’
Here’s what you gave me: [paste the boring bio].
Here’s what I actually want: a punchy, human bio, under 80 words, first person, light humor, and specific about my work in marketing analytics. Rewrite it. Then explain why your first version was generic.”
Now the AI:
- rewrites it,
- tells you why it sucked the first time,
- and accidentally teaches you how to prompt better.
Use this with ChatGPT, Claude, Gemini, Grok, whatever. If it types, it can learn.
---
**2. A practical use case most novices miss**
You already know “write emails” and “summarize stuff.”
Here’s one you’re probably not using: **weekly decision briefings for your life or job**.
Example:
“Act as my chief-of-staff. I’m a project manager juggling 3 projects. Summarize my week from these notes and tasks, highlight the 5 biggest risks, and suggest what I should prioritize Monday morning in under 200 words, plain English.”
Suddenly the AI is not just writing sentences, it’s helping you decide what to do next.
Less doom‑scrolling, more doing.
---
**3. One common beginner mistake (that I made)**
The classic mistake: **prompting like it’s Google**.
I used to type,
“Tips for time management”
and then complain that the answer was a boring list I could’ve guessed myself.
The fix? **Context + constraints.**
Try:
“I’m a freelance designer working from home with two kids and ADHD. Give me 5 time‑management tips I can implement this week, each under 2 sentences, focused on scheduling and avoiding distractions.”
Same AI, completely different brain.
Give it *who you are*, *what you’re trying to do*, and *how you want the answer*.
Yes, I still forget sometimes and type “make this better.”
Yes, the AI still gives me hot garbage when I do.
We learn. Slowly.
---
**4. A simple exercise to build your AI skills**
Here’s your low‑pressure drill you can do in 5–10 minutes:
1. Pick one task: let’s say “rewrite this email” or “plan my week.”
2. Start with a lazy prompt:
“Rewrite this email to sound more professional.”
3. Then do **three improved versions**:
- Version A: “Act as a friendly but direct manager. Rewrite this email to be clear, polite, and under 120 words.”
- Version B: “Act as a communications coach. Improve clarity and tone, keep my voice casual, and remove any confusing phrases.”
- Version C: “Act as my editor. Give me a bullet-point critique of this email first, then rewrite it using your own suggestions.”
Compare the outputs.
Notice how the role, tone, and format change the result.
Congratulations, you’re now *directing* the AI instead of begging it.
Do that once a day for a week and you’ll be better at this than most “AI strategists” on LinkedIn.
---
**5. A tip for evaluating and improving AI-generated content**
Never trust the first draft.
Think of the AI as a bright intern who lies confidently.
Use this simple two‑step check:
1. **The Read‑Out‑Loud Test**
Read it aloud. If you cringe, trip over phrases, or think, “I would never say that,” it needs editing.
2. **The “Make It Better” Follow‑Up**
Tell the AI:
“Now improve this. Keep the key ideas, but:
- cut 20% of the words,
- remove clichés,
- and make it sound like a real person talking to another real person.”
For factual stuff, add:
“List any claims that might need verification and mark anything you’re not confident about.”
You’re not just accepting output, you’re *shaping* it.
---
Alright, misfits, that’s it for today’s episode of “I Am GPTed.”
If this helped you boss your AI around a little better, **subscribe to the podcast** so you don’t miss future episodes where we lovingly bully more chatbots into being useful.
**Thanks for listening**, seriously – you could be doom‑scrolling, but you chose to level up instead.
This has been a **Quiet Please** production.
You can learn more at **quietplease.ai**.
[Outro music fades out]
For more check out https://www.quietperiodplease.com/
and for some great deals go to https://amzn.to/4nidg0P