Sometimes I ask AI to make something sound more human.
Then it gives me polite wallpaper.
Everything is smooth.
Everything is reasonable.
Everything is just a little too pleased with itself.
You know the voice.
“I hope this message finds you well.”
“It is important to note…”
“In today’s fast-paced world…”
“I wanted to kindly follow up…”
“Please do not hesitate to reach out…”
Nothing is exactly wrong.
That is the problem.
It sounds fine, but it does not sound like a person who actually had to send the thing.
So this week’s Friday Spark move is slightly backwards.
Instead of asking AI:
“Make this sound more human.”
Try asking it to go the other way first.
Ask it to write the most obvious AI-sounding version possible.
Let it show the costume.
Then ask:
“What patterns did you use that made this sound AI-generated?”
Once the pattern is visible, you can reverse it.
Less fluff.
Less corporate fog.
Less fake warmth.
More like you.
This week’s 0.1 move
Use AI to reveal the AI habits first.
Then ask it to avoid them.
Copy this:
The Reverse-AI Prompt
I want to rewrite this so it sounds less AI-generated.
Step 0:
ask me to input the text to rewrite
Step 1:
Rewrite it in the most obvious AI-sounding style possible.
Make it overly polished, overly polite, generic, safe, and a little too smooth.
Step 2:
List the patterns you used that made it sound AI-generated.
Look for things like:
overly formal openings
vague phrases
corporate tone
extra enthusiasm
generic advice
unnecessary explanation
phrases that sound polished but not personal
Step 3:
Now rewrite my original text again.
Avoid all those patterns.
Keep my meaning.
Use plain, natural language.
Do not make it fancy.
Do not make it corporate.
Do not add fake enthusiasm.
Make it sound like something I would actually say.
That is the practice.
Not “AI bad.”
Not “never use AI.”
Just this:
Let the robot show its costume first.
Then decide what to keep.
Good. That is 0.1.