AI slop is everywhere now, and it’s polluting the internet.
You see it in blog posts, LinkedIn posts, thumbnails, product descriptions, stock-looking AI images, YouTube scripts, and more.
However, AI is not the problem.
At Run the Prompts, we use AI all the time, which I’m sure is not shocking. We use it for writing, brainstorming, image generation, editing, testing, analyzing data, and so much more.
The problem is copy-paste AI that provides little to no value to anyone. That is what I consider AI slop.
This guide will help you spot AI slop when you see it, avoid consuming it, and avoid making it when you use AI yourself.
Because yes, people are watching for this stuff now. Once they think your content smells like AI slop, good luck getting them to trust it ever again.
Let’s party.
A Quick Warning Before We Begin
None of these signs automatically prove something was made by AI, so take this guide with a grain of salt.
Think of it as a guide instead of a slop bible.
Also, if you want to take things in the complete opposite direction, be sure to check out our funny guide on how to create the sloppiest AI slop imaginable.
Seven Ways to Spot AI Slop
Text-based AI slop is usually easy to spot once you know what to look for.
A lot of you can spot it just by using ChatGPT or Claude all day because you get used to their writing patterns.
Here are all of the things I’ve noticed so that you can, too.
1. Words and Phrases AI is Madly in Love With
AI tools like ChatGPT and Claude tend to overuse certain words and phrases.
Some of these are not bad on their own. The problem is when they show up again and again, especially in content that otherwise sounds oddly generic.
Common offenders include:
- “clean”
- “cleanly”
- “pair”
- “pairing”
- “no fluff”
- “core”
- “wild ride”
- “dive in”
- “dive into”
- “delve in”
- “delve into”
- “here’s the kicker”
- “making waves”
- “buzzing”
- “abuzz”
- “brimming”
- “let’s be honest”
- “at your fingertips”
- “hand-wavy”
- “the truth?”
- “Enter: the”
- “it’s like having a”
- “think of it as”
- “pipeline”
- “polished”
Again, these words do not automatically mean AI wrote the article.
But if you see five of them in the first few paragraphs, then get your mouse ready. It’s likely time to hit the “Back” button on your browser and run away screaming.
How to Avoid This
After using AI to draft something, search for these phrases, or even have AI search for them.
Then replace them with normal language.
Instead of: “Let’s dive into the core benefits of this game-changing tool.”
Write: “Here is what this tool does well.”
That was easy and fun, right?
2. Em Dashes Killed the Robot Star
This one is annoying because em dashes are technically correct. In case you’ve never cracked a book in your life, they existed long before ChatGPT, Claude, Gemini, or whatever AI tool is currently trying to write your breakup text.
But now, em dashes have become one of those things people associate with AI writing.
It may not be fair, but it is what it is.
If readers see a wall of em dashes and think, “This was probably written by AI,” you have a problem.
How to Avoid This
Use commas, periods, or parentheses instead.
We actually created an article with a special prompt in it that describes how to stop ChatGPT from using em dashes.
3. The “It’s Not Just X, It’s Y!” Sentence Structure
You have seen this one everywhere.
“It’s not just a tool, it’s a productivity system.”
“It’s not just a camera, it’s a creative partner.”
“It’s not just a diet, it’s the future you that isn’t fat.”
This structure existed before AI. Plenty of humans used it. But AI loves it now, and readers have noticed it.
How to Avoid This
Say things directly.
Instead of: “It’s not just an AI image generator; it’s a full creative workspace.”
Write: “This AI image generator also gives you tools for editing, organizing, and finishing your work.”
Alternatively, you can literally tell ChatGPT this: “Do not use this sentence structure: It’s not just X, it’s Y.” However, as with anything AI, it may not follow those instructions 100% of the time.
Isn’t avoiding sounding like a robot fun?
4. Short Sentences Everywhere
AI sometimes tries to sound human by using a bunch of tiny dramatic sentences.
Like this.
One word.
Then another.
For emphasis.
Because drama.
And honestly? No.
This style can work sometimes. But when it is overdone, it feels fake. It reads as if Tony Robbins got trapped inside a fortune cookie factory.
How to Avoid This
Use short sentences when they help the rhythm.
Do not use them as a costume.
Good writing has a variety of sentences: short, long, plain, weird, etc. Think about it. That is how humans write.
5. And Honestly? I Get It. It’s Everywhere.
“And honestly? I get it.”
This phrase needs to be taken out back and shot several times.
AI uses this line when it wants to sound warm, relatable, and emotionally available.
How to Avoid This
Just say what you mean.
Instead of:
“And honestly? I get it. Writing with AI can feel overwhelming.”
Write:
“Writing with AI can feel overwhelming at first.”
Or, and I don’t mean to sound like a broken record, you can also instruct the AI not to use this phrase, and it will usually listen.
6. Cut the Filler and Repetition That Pretends to Say Something
This is one of the biggest signs of AI slop.
AI writing often looks useful at first glance. It has headings, bullets, and real paragraphs.
Then you read it and realize it says almost nothing.
It is content-shaped air.
Example:
“In today’s fast-paced digital landscape, creators need powerful solutions that help them produce better content more efficiently while staying ahead of the curve.”
That sentence sounds like it should mean something, but it doesn’t.
How to Avoid This
Ask this after every paragraph:
“What did the reader actually learn?”
If the answer is “nothing,” delete it or rewrite it.
Good AI-assisted writing should explain something, show or teach something interesting, or make the reader laugh.
If it only fills space, it goes in the garbage, where it belongs.
7. High Scores on ZeroGPT.com
The higher the score on ZeroGPT.com, the higher the likelihood of slop. There are a lot of tools like this now, but ZeroGPT is considered the gold standard.
Copy and paste an article or social media post into it, and you’ll see what I mean.
It’s not 100% accurate and never will be, but it will at least give you a good idea of what’s going on.
Here’s the rating for this article you’re reading right now. Anything under 15% is considered good:
This post got 8%… nice.
The Real Rule: USE AI, But Do Not Let AI Think for You
Using AI does not make you lazy. Using it badly makes you lazy.
There is a huge difference between these two workflows:
Bad workflow:
- Ask AI for an article.
- Copy it.
- Paste it.
- Publish it.
- Wonder why nobody cares.
Better workflow:
- Bring your own opinion.
- Use AI to draft, organize, brainstorm, or edit.
- Cut the generic parts.
- Add examples.
- Add personal experience.
- Rewrite anything that sounds robotic or stupid.
- Publish something useful.
That is the difference.
AI should help you move faster. It should not replace your taste, judgment, experience, humor, or ability to notice when something sucks.
Quick Checklist Before Publishing AI-Assisted Content
For writing:
- Remove phrases AI overuses.
- Cut generic intros.
- Replace filler with actual information.
- Add personal opinions.
- Add examples from your own experience.
- Vary your sentence length.
- Remove fake enthusiasm.
- Avoid corporate-sounding, LinkedIn-esque language
- Read it out loud.
- Ask, “Would I actually say this?” and “Would I actually enjoy reading this?”
AI is a tool. Use it the right way or get out of my way because it will use you.
AI Slop FAQ
AI slop is AI-generated content that was produced with little or no human thought or oversight. It inherently has little to no value to anyone, and it’s polluting the internet.
“AI slop” came from internet slang around 2022, was popularized by tech writers like Simon Willison in 2024, and spread because people needed a simple word for cheap, low-quality, mass-produced AI content.
You don’t. If you want to make money with AI slop, you either won’t entirely, or you will for a couple of months until the algorithms catch up to you. Making any amount of money with AI slop is extremely difficult because it relies on making money from content that pretty much nobody wants to consume.
The main reason people make AI slop is that they are lazy and lack moral fiber. They want to make money or get attention online by doing nothing.
AI detectors like ZeroGPT do not prove authorship; they make educated guesses from statistical writing patterns, so human writing can still get flagged as AI when it is refined, formal, predictable, template-based, short, heavily grammar-edited, or written in careful/non-native English. A high score is only a signal, not evidence, and the best defense is version history, drafts, notes, outlines, timestamps, and revision records showing how the writing was created.
Wrapping It Up
If you really think the AI slop situation is going to get any better anytime soon, you’re dreaming.
It’s only going to get worse.
That’s why it’s on you to always follow the golden RTP formula: AI + your brain.
It’s how I’ve built a successful AI-powered solopreneurship business.
Successful businesses are not built with slop.
AI slop is not content made with AI. AI slop is content made with AI and no human brain cells. It’s worthless hot dog content that nobody wants.
If you have noticed other signs of AI slop in writing, add a comment in the comments section below. If it’s good enough, I’ll add it to this post.
Until next time, remember to run the prompts and prompt the planet.
Tired of AI filters and data-harvesting in tools like ChatGPT? Try Venice today, built for more creative freedom and privacy. Get 20% off any paid Venice subscription for a limited time with promo code RUNTHE20. Disclosure: This is an affiliate link, and I may earn a commission if you purchase.
