You rewrite the same AI draft every single day.
Delete the em dashes. Cut the "it's not just X, it's Y." Add the joke you'd actually make. Fifteen minutes, every time, forever, because the AI never remembers what it got wrong yesterday.
Teach it once instead. Permanently.
A guide, not a subscription
Free to join. 30 day guarantee when it opens.
You paste your writing into ChatGPT, Claude, Gemini. Something comes back. Clean. Structured. Mediocre. Not you.
So you fix it. Line by line, cutting the slop and putting back the way you'd actually say it. Fifteen minutes later the draft is finally yours.
Tomorrow, you do it again. Same fix. AI doesn't remember what it got wrong yesterday. It's Groundhog Day: a new conversation, spelling out who you are and how you sound, from scratch, again.
Add up enough of those fifteen minute fixes and you've spent more time managing AI than it would've taken to just write the thing yourself. You're not imagining it.
That's backwards.
AI is supposed to save you time. Instead it's costing you time.
You're not afraid of AI. You're afraid of getting called out for using it.
If you're an employee, that's the fear of your boss reading what you wrote, or looking like you're phoning it in at exactly the wrong moment.
If you run your own business, it's a client noticing something's not quite right in what you wrote, and trusting you a little less.
Different judge, same fear
One person got questioned by an editor over swapping a single word, "combined" for "unified". Not kidding. That's the bar now. A word you changed without thinking, and somebody noticed. Not because it was wrong. Because it wasn't quite you.
The people ready to call your writing AI slop are running the same drafts through the same tools. Recent research on LinkedIn found more than half of long posts are now likely AI written. Over half of employees hide their own AI use at work. The ones who don't hide it pay for it: Duke University found people who disclose using AI face measurable judgment about their competence.
You don't want to give up what AI gives you either. The speed. The hours back. The version of your job that doesn't eat your whole life. You just don't want to be accused of writing AI slop. It's gross, and you know it.
So we're not here to stop you from using AI.
We're here to make sure you never write slop again.
We're here to make it sound human.
Long before AI existed, I trained under Alison Armstrong to become a PAX certified dating and relationship coach, not tech, relationships. She called the core skill a Great Ask: say what you need, show what it looks like, explain why it matters, ask what they need from you to make it happen.
Turns out AI needed the exact same four parts.
Stop rewriting what AI hands you. Go back to what actually bothered you about it, that's a great ask waiting to happen. Say what you need. Show what it looks like. Ask what it needs from you. Same method, aimed at a machine instead of a person.
You've probably already tried the obvious fixes.
A brand voice guide
have one
train AI
And most still call the results disappointing. It doesn't learn from your corrections or notice what it missed last time. You're still catching every mistake, by hand, forever.
An AI humanizer tool
for the same broken fix
Built to beat a detector, not a person. Detectors retrain constantly, so today's fix fails tomorrow, and it can't stop your boss or your client from noticing either.
Neither one fixes the actual problem: the AI itself never learns. You're the one doing all the remembering.
There's a different way to do this.
Fix the AI. Not the output.
The AI Slop Stopper doesn't touch the output. It fixes the AI that produced it.
Join the waitlistAsk AI for a draft.
Anything, an email, a post, a proposal.
Don't edit it. Rewrite it.
Write your own version from scratch, the way you'd actually say it.
Hand both back.
Ask the AI what changed.
Tell it what you need.
From it, to get more of what you want.
It answers with a fix.
Say "Lock it in" to confirm. The AI updates itself, not just this draft.
You're not managing a tool. You're training one, and it never needs the same feedback twice.
Huge improvements in 7 days. It sounds like you on the first try in about 3 weeks. Less editing every time you use it after that.
A humanizer subscription runs $120 to $480 a year and never stops charging. $47 once outlasts five years of theirs.
Today
one time · no subscription
Join the waitlistFree to join. Works inside the AI tools you already use. 30 day guarantee when it opens.
Business owners.
Employees.
Anyone tired of AI making them sound like everyone else, which means sounding like no one. If you've felt the fear of sounding like AI slop, this is for you.
Not technical. Five minutes gets you set up and through your first real correction. It gets sharper every time after that.
Not for you if you're happy with how AI already sounds. This is for people who keep rewriting.
Nobody told you AI writing has to stay generic.
You've just heard it from every AI bro selling a prompt pack, calling it a feature instead of a fix nobody's bothered to build. That belief is the product an entire industry sells you, one subscription at a time.
You don't need to understand how the model works. You need one method that sticks. That's the whole point of the AI Slop Stopper. You work too hard for this to be your first impression.
"It makes it so easy for me to just show up and do things that before felt overwhelming."
Pamela
"Carol made it beautifully simple."
Marcus
"I use this on every draft that leaves my desk."
Happy client
Same rigor, a different kind of fix: a funnel diagnosis that took checkout conversion up 5x.
Rhonda
It will be $47, one time. No subscription.
The AI Slop Stopper is not open yet. When it is, it will be forty seven dollars, paid once, with no subscription behind it. Joining the waitlist costs nothing and gets you the first email when it opens.
The It Just Works Guarantee
When it opens you will have 30 days. If it does not work for you, email me and I will refund you.
There is nothing to decide today. If you have questions before it opens, email me directly: carol@carolroderick.ca
Get on the list.
The AI Slop Stopper opens soon. Put your name down and you will get the first email when it does, before anywhere else.
Free to join. It will be $47 when it opens.
Is this complicated?+
No. Copy, paste, five minutes to get started. Huge improvements in 7 days. Actually sounds like you on the first try in 3 weeks. Less editing required every time you use it.
Does this work on my platform?+
Yes. ChatGPT, Claude, Gemini all support persistent instructions. The guide walks through each one.
I don't have time to learn a new system.+
You're not learning a system. You're doing what you already do (correcting AI drafts) and adding three words at the end.
What if I get stuck?+
Everything's in the setup guide, step by step, with the wording specific to each platform. If you're still stuck, email me.
Does this replace ChatGPT, Claude or Gemini?+
No. Works inside whichever one you already use.
Will this stop working when the AI updates?+
It's about how you interact with the tool, not a hack tied to one model version. Built to keep working across updates.
Can I use this for client work, not just my own voice?+
Yes. Same method works for any voice you need AI to match, including a client's.
Sources
Time spent editing AI content: MarTech. AI written posts on LinkedIn: Originality.AI and Originality.AI engagement study. Judgment for disclosing AI use: Duke University, PNAS. Search volume for "humanize AI" and "AI humanizer": OpenSEO keyword data, United States, August 2026.