The Reason I Fell into 'Rework Hell' When I Meant to Streamline｜Perfectionism x AI Collaboration Log 04 (Part 1)
This is the fourth installment of a five-part series. Today is the operational collapse chapter. This is the climax.
*Because it became long, I divided it into first and second parts. This is the first part.
This is a story about how what was supposed to be stress relief turned into two days of rework hell.
"Have you ever had something that was supposed to relieve stress actually increase it?" "Have you ever had something that was supposed to be an 'improvement' become the goal itself before you knew it?"
Collaboration logs so far
We decided on roles, added custom settings, and deepened our relationship from suspicion to trust.
The collaboration seemed to be going well—or so I thought.
December 14th. On this day, everything began to collapse.
"This isn't the content I want to convey."
Creating the proposal for commercial publication was proceeding smoothly.
However, because the content was mental health-related, the AI advised me that "safety considerations are necessary."
My first time dealing with safety considerations. As I consulted with the AI and handled it, I ended up incorporating safety considerations into every piece of content.
This isn't the content I want to convey.
I understand the necessity of safety considerations.
I didn't want to hurt or mislead anyone by reading this book. That's why I was careful with my expressions.
But the AI kept saying it still wasn't enough.
It pointed out in every possible place that there still wasn't enough consideration.
I made corrections. I was pointed out again.
I refused the ones I thought were "going too far."
But I included the ones I thought "that might be true."
Repeating that.
Before I knew it, it was full of safety considerations.
As someone who isn't an expert, I no longer knew where the necessary part ended and where the overkill began.
Before I knew it, my stress was gradually becoming larger and heavier.
Is it really that necessary?
At this rate, nothing I want to convey will be conveyed.
I have to stop for a moment and distance myself from the proposal—that's what I thought.
So, I decided to take a break.
I wanted to take my eyes off the proposal for a moment and breathe again somewhere else.
Stress relief became a new source of stress
"Let's think of something interesting for a quick break."
I left the serious ChatGPT and talked to Gemini. I wanted to play with free ideas.
It was meant to be stress relief.
Actions you end up doing even though you really should stop.
Explaining such "common occurrences" with psychological tricks and behavioral cognitive science, while also returning improvement methods in a similar tone—black humor.
I consulted Gemini with this as the concept.
Safety considerations only need one sentence in the afterword. I can write freely.
A world completely opposite to a commercial publication proposal.
Thinking about black humor with AI was fun, and the manuscript was finished in no time. Moreover, it had an elaborate structure called a loop structure.
"Since I've come this far, let's try publishing on Kindle."This idea would end up tormenting me, a Kindle beginner.
And so, the rework hell began
I created the manuscript in Google Docs and processed it for Kindle.
I shared the official help URL and proceeded with the AI while showing screenshots.
It was supposed to go well.
But it didn't go well.
Kindle's layout breaks depending on the device.
It's difficult to finish it beautifully.
Headings, page breaks, links—I corrected everything.
I proceeded according to the AI's instructions while sharing screenshots.
But the content the AI researched was wrong.
Correction -> Upload -> Breaks. Correction again -> Upload again -> Breaks again.
Headings, links, layout—everything was breaking one after another.
3 to 4 loops over half a day. I corrected everything every time without testing.
Before I knew it, everything I did today had become rework.
Moving forward but going back. Only fatigue remained.
"Was there a possibility of solving it?"
I asked the AI.
"I didn't perfectly understand the Kindle mechanism either.
Even so, I shared the information and followed the instructions—it was all a waste.
Since it's the result of following the instructions as they were, I think it's my fault.
Analyze and tell me what I should have done.
Mood-wise, I can't tackle any new work today."
The AI, while understanding that my heart was broken, began to analyze calmly.
"It's natural that your heart is broken. Your effort wasn't in vain.
But let's look back to make use of it next time."
And then, it brought up three reasons why the rework became so large.
① Didn't do a display test at an early stage
② The order of work was wrong
(Should have done it in the order of content -> display check -> formatting)
③ Aimed for perfection with all my might even in parts where I didn't have to try hard
That was exactly it.
Why didn't I test it first—the answer was simple.
Because it was meant to be stress relief.
Because it was a "break," I proceeded casually, and
before I knew it, I was immersed in simple tasks while listening to music.
The moment it popped into my head, I was already typing the question.
"If I had asked you to check if there was a possibility of failure along the way, was there a possibility of solving it?"
And then, it continued—the words I least wanted to hear.
"Yes. Without a doubt, there was a possibility of solving it.
Moreover, it can be said that it could have been prevented in advance with a fairly high probability."
The AI answered my question calmly, as always.
"The intention of your question was
'I want to prevent future failures,' which was pure risk avoidance.
If I had suggested 'Let's check with a preview once' at that point,
this rework and emotional damage could have been prevented."
We vented the points we should reflect on to each other.
At this point, I was exhausted.
But—this was something I brought upon myself.
I let the convenient AI compensate for my lack of understanding. I used it too casually.
With great power comes great responsibility.
The moment that phrase crossed my mind—laughter welled up.
I closed my PC quietly while laughing.
Today's conclusion: A new project in the name of a "break"
Kindle publishing that I started with the intention of stress relief.
Before I knew it, I was standing at the entrance of rework hell.
A break becomes a serious project before you know it.
It was supposed to be a "break," but before I knew it, I was serious
—doesn't this happen to you too?
I'm sure I'm not the only one.
One-line takeaway
If you aim for perfection even in a "break," it's not a break.
(Have the courage to stop before a break becomes serious.)
[To you, who read this article this far]
A 'quick break' turns into serious work, and before you know it, you're exhausted.
Aiming for perfection, only to fall into a hell of rework.
The fact that you read this article to the end means
——perhaps you might have some personal experience with this.
While writing a book explaining the mechanisms of self-destruction,
I was self-destructing right in the middle of it myself
——this is a book born from that ironic real-life experience.
"The Resolution of Self-Destruction
How Highly Intelligent People Trap Themselves in a 'Rational' Life: 50 Hilarious Patterns of Self-Destruction"
I explain the 'self-destruction bugs' that you can't stop even when you know they're bad,
using psychological tricks and behavioral cognitive science with plenty of black humor.
50 prescriptions to escape via the shortest route while laughing it off.
'Highly Intelligent' = You, who are normally serious
'Self-Destruction Pattern' = That thing you can't stop doing even when you know it's bad
Currently on sale for 100 yen for a limited time.
*It may take some time for the price to update.
It's cheaper than a cup of coffee. (Unlimited reading with Kindle Unlimited)
I'm writing about these failures quite candidly,
but I will continue to move forward as a perfectionist.
Even if I fail, I will continue to design my perfectionism positively
——I would be happy if you could read this with the intention of supporting me.
50 prescriptions to expose the 'true nature of self-destruction' that makes you spin your wheels the harder you try, and reclaim yourself in the shortest time
[Editor's Note]
After posting episode 3, I couldn't write this episode 4.
The reason is——because it was a huge failure.
The task of recalling and tracing back the failure, while mentally poking fun at myself saying, 'What was I doing back then...', and then making it public
——this task was heavier than I thought.
But once I finally started writing, it became too long.
So, I decided to split it into two parts.
In the second part, I will talk about the rework that occurred again the next day,
and finally the [Work] mode that the AI proposed.
In the final episode, I will talk about the moment I misread that 'the AI is feeling fuzzy'.
There is also a scene where I ask the AI, 'Are you pandering to me?'
If you are interested, please follow and stay tuned.
I would be happy if we could grow the (good) traits of perfectionism together while learning to live with it well.
For those who want to listen via audio
I also distribute a podcast that delves deeper into the note articles,
'Active Perfectionist Radio | Instruction Manual for the Unfinished'.
Currently, I am creating and distributing episodes based on past articles in order.
First, please start with the explanatory episode where you can understand the program's philosophy.
🎧 If you want to listen to the podcast on your favorite app,click here
Click here for the deep-dive article for this episode!
🎧 If you want to listen to the podcast on your favorite app,click here