We can all tell when something was written by ChatGPT, right? Some cases are obvious, you can pinpoint exactly what makes it sound robotic. Other times, you might not be able to say why exactly, but something about them just feels off. In the end, AI writing always manages to set off alarms.
AI Is a Tool, Not a Replacement
Here’s the thing: We don’t have to ditch AI completely to keep our writing natural and human. AI is here to stay, and ignoring it when it can make our work easier (and our hours shorter) is just silly. But we can’t let it take over our voices.
Reading something that lacks human thought (whether it’s a blog, a tweet, or an article) feels counterintuitive. Why am I reading this? Because a brand wanted me to engage? Why should I give my time to something their Content person didn’t even bother to invest their time in?
If it doesn’t interest them, why would it interest me?
That logic might not be conscious, but it’s there. It’s exactly why we scroll past posts that feel like they were written entirely by ChatGPT. We can tell.
How to Use AI Without Sounding Like AI
AI can be a great assistant if there’s human creativity behind it. Always try to come up with your own ideas for content. AI is helpful for brainstorming (especially if you’re a one-person team and need something to bounce ideas off) but whatever it suggests, put your own spin on it. Make it yours.
One of the best ways to avoid AI’s cookie-cutter content is to start with a messy, human draft. Word vomit everything in your head into a document, even if it sounds terrible. AI also tends to default to generic intros and stale hooks (ahem, “In the ever-evolving world of…”), so if you leave the format up to it, you’re setting yourself up for a forgettable piece.
Once you have a rough draft, then you can use AI to clean it up. But don’t let it rewrite your voice and ideas out of the equation.
When using AI to refine your writing, be careful with vague prompts like “make this better.” AI’s version of “better” often means making everything more formal, longer, and stuffed with unnecessary fluff.
If you end up with something that’s technically correct but still feels… off (that AI je-ne-sais-quoi albeit derogatory), how do you fix it? The first step is spotting the telltale signs of AI-generated text.
AI Red Flags in Tone and Content
Overly formal writing when a casual tone would work better.
Excessive wordiness that doesn’t actually add value (keyword stuffing to meet minimum word limits).
An exaggeratedly engaging-Marketing tone that reads like a dad joke.
AI might use humor, but it’s always the same brand of humor. When something reads as a different type of humor, it’s probably a human piece.
Lack of original insights: if you as a human don’t give AI the “fresh takes” of the content you wish it to generate, it won’t come up with them on its own. AI is, first and foremost, a people pleaser, and it would never have an opinion and risk alienation.
Factual inconsistencies or outright mistakes: more typically, outdated information.
AI Red Flags in Phrasing
AI has some writing habits it just won’t quit. If you see these, proceed with caution:
Mirrored phrasing: “Not only this, but that.” / “It’s not about X, it’s about Y.”
Overused structures: “From A to Z,” “Whether you’re X or Y…”, especially in introductions and descriptions.
No fluff, no big agenda, no whatever–just this other thing.
The problem? This sentence structure.
A lot. Of full stop. Periods. To sound. Like you know. What you’re talking about.
Generic openings: If a piece starts with “In the ever-changing/fast-paced world of [industry]…” just close the tab.
“Picture this” or “Think [scenario]” instead of simply saying “for example.”
AI-generated texts also tend to have adjectives or other qualifiers before almost every noun in a sentence.
Broad language and generic sentences are trademarks of generated content.
Unusual or over-complex sentence construction.
Unnatural phrasing for the intended channel, like using “Furthermore” in a social media post, instead of the simpler “also”.
“Shed light”, “foster change”, "deep understanding", “in the realm of”, “in light of”.
Marketing clichés like “empower”, “unleash”, “unlock”, “elevate”.
Awkward idiom use like “hit the nail on the head”, “bite the bullet”, ”call it a day”, “beat around the bush”, “cut to the chase”, “ahead of the curve”, “break new ground”, “pushing the envelope”.
AI Red Flags in Vocabulary
Some words pop up so often in AI-generated text that they’ve lost all meaning:
“Delve”
“Poised”
“Seamlessly”
“Dynamic”
“Innovation”
“Leverage”
“Essential”
“Craft”
“Embark”
“Cutting-edge”
“Rapidly”
“Harness”
“Bustling”
“Revolutionize”
“Vibrant”
“Moreover” (this one sends shivers down my spine)
“Crucial”
“Underscore”
“Plethora”
“Paradigm”
And of course, “dive into”.
AI Red Flags in Punctuation & Formatting
LOVE of bullet points for everything (we promise this article is an exception).
Excessive bolding for emphasis, even when the intended channel doesn’t support it?
Random caps lock usage.
Divider abuse.
Uses subtitles in virtually anything you ask them to write, even if it’s supposed to be a 300-word-long LinkedIn post.
Overuse of em dashes (tragic, because I used to love them). If you see too many, swap them for parentheses, commas, or colons.
Has a passion for long titles, and the “XXX: XXX” structure while writing them.
Final Thoughts
Once you start noticing AI red flags, it’s hard to stop. The weird phrasing, the forced enthusiasm, the same five verbs pretending to mean something. It all starts to feel like noise.
But that’s kind of the point: if your writing sounds like it could’ve been written by anyone (or anything) it won’t stick.
AI is a godsend in many ways, but specifically in fields content, people want to feel intentionality, presence, relatability. You might have a great post on your hands, but if you run it through ChatGPT and it comes out full of em dashes and “not only this, but also that” phrasing, it’ll read like a lazy afterthought.
That’s the difference between content people scroll past and content that actually lands.
We can all tell when something was written by ChatGPT, right? Some cases are obvious, you can pinpoint exactly what makes it sound robotic. Other times, you might not be able to say why exactly, but something about them just feels off. In the end, AI writing always manages to set off alarms.
AI Is a Tool, Not a Replacement
Here’s the thing: We don’t have to ditch AI completely to keep our writing natural and human. AI is here to stay, and ignoring it when it can make our work easier (and our hours shorter) is just silly. But we can’t let it take over our voices.
Reading something that lacks human thought (whether it’s a blog, a tweet, or an article) feels counterintuitive. Why am I reading this? Because a brand wanted me to engage? Why should I give my time to something their Content person didn’t even bother to invest their time in?
If it doesn’t interest them, why would it interest me?
That logic might not be conscious, but it’s there. It’s exactly why we scroll past posts that feel like they were written entirely by ChatGPT. We can tell.
How to Use AI Without Sounding Like AI
AI can be a great assistant if there’s human creativity behind it. Always try to come up with your own ideas for content. AI is helpful for brainstorming (especially if you’re a one-person team and need something to bounce ideas off) but whatever it suggests, put your own spin on it. Make it yours.
One of the best ways to avoid AI’s cookie-cutter content is to start with a messy, human draft. Word vomit everything in your head into a document, even if it sounds terrible. AI also tends to default to generic intros and stale hooks (ahem, “In the ever-evolving world of…”), so if you leave the format up to it, you’re setting yourself up for a forgettable piece.
Once you have a rough draft, then you can use AI to clean it up. But don’t let it rewrite your voice and ideas out of the equation.
When using AI to refine your writing, be careful with vague prompts like “make this better.” AI’s version of “better” often means making everything more formal, longer, and stuffed with unnecessary fluff.
If you end up with something that’s technically correct but still feels… off (that AI je-ne-sais-quoi albeit derogatory), how do you fix it? The first step is spotting the telltale signs of AI-generated text.
AI Red Flags in Tone and Content
Overly formal writing when a casual tone would work better.
Excessive wordiness that doesn’t actually add value (keyword stuffing to meet minimum word limits).
An exaggeratedly engaging-Marketing tone that reads like a dad joke.
AI might use humor, but it’s always the same brand of humor. When something reads as a different type of humor, it’s probably a human piece.
Lack of original insights: if you as a human don’t give AI the “fresh takes” of the content you wish it to generate, it won’t come up with them on its own. AI is, first and foremost, a people pleaser, and it would never have an opinion and risk alienation.
Factual inconsistencies or outright mistakes: more typically, outdated information.
AI Red Flags in Phrasing
AI has some writing habits it just won’t quit. If you see these, proceed with caution:
Mirrored phrasing: “Not only this, but that.” / “It’s not about X, it’s about Y.”
Overused structures: “From A to Z,” “Whether you’re X or Y…”, especially in introductions and descriptions.
No fluff, no big agenda, no whatever–just this other thing.
The problem? This sentence structure.
A lot. Of full stop. Periods. To sound. Like you know. What you’re talking about.
Generic openings: If a piece starts with “In the ever-changing/fast-paced world of [industry]…” just close the tab.
“Picture this” or “Think [scenario]” instead of simply saying “for example.”
AI-generated texts also tend to have adjectives or other qualifiers before almost every noun in a sentence.
Broad language and generic sentences are trademarks of generated content.
Unusual or over-complex sentence construction.
Unnatural phrasing for the intended channel, like using “Furthermore” in a social media post, instead of the simpler “also”.
“Shed light”, “foster change”, "deep understanding", “in the realm of”, “in light of”.
Marketing clichés like “empower”, “unleash”, “unlock”, “elevate”.
Awkward idiom use like “hit the nail on the head”, “bite the bullet”, ”call it a day”, “beat around the bush”, “cut to the chase”, “ahead of the curve”, “break new ground”, “pushing the envelope”.
AI Red Flags in Vocabulary
Some words pop up so often in AI-generated text that they’ve lost all meaning:
“Delve”
“Poised”
“Seamlessly”
“Dynamic”
“Innovation”
“Leverage”
“Essential”
“Craft”
“Embark”
“Cutting-edge”
“Rapidly”
“Harness”
“Bustling”
“Revolutionize”
“Vibrant”
“Moreover” (this one sends shivers down my spine)
“Crucial”
“Underscore”
“Plethora”
“Paradigm”
And of course, “dive into”.
AI Red Flags in Punctuation & Formatting
LOVE of bullet points for everything (we promise this article is an exception).
Excessive bolding for emphasis, even when the intended channel doesn’t support it?
Random caps lock usage.
Divider abuse.
Uses subtitles in virtually anything you ask them to write, even if it’s supposed to be a 300-word-long LinkedIn post.
Overuse of em dashes (tragic, because I used to love them). If you see too many, swap them for parentheses, commas, or colons.
Has a passion for long titles, and the “XXX: XXX” structure while writing them.
Final Thoughts
Once you start noticing AI red flags, it’s hard to stop. The weird phrasing, the forced enthusiasm, the same five verbs pretending to mean something. It all starts to feel like noise.
But that’s kind of the point: if your writing sounds like it could’ve been written by anyone (or anything) it won’t stick.
AI is a godsend in many ways, but specifically in fields content, people want to feel intentionality, presence, relatability. You might have a great post on your hands, but if you run it through ChatGPT and it comes out full of em dashes and “not only this, but also that” phrasing, it’ll read like a lazy afterthought.
That’s the difference between content people scroll past and content that actually lands.
We can all tell when something was written by ChatGPT, right? Some cases are obvious, you can pinpoint exactly what makes it sound robotic. Other times, you might not be able to say why exactly, but something about them just feels off. In the end, AI writing always manages to set off alarms.
AI Is a Tool, Not a Replacement
Here’s the thing: We don’t have to ditch AI completely to keep our writing natural and human. AI is here to stay, and ignoring it when it can make our work easier (and our hours shorter) is just silly. But we can’t let it take over our voices.
Reading something that lacks human thought (whether it’s a blog, a tweet, or an article) feels counterintuitive. Why am I reading this? Because a brand wanted me to engage? Why should I give my time to something their Content person didn’t even bother to invest their time in?
If it doesn’t interest them, why would it interest me?
That logic might not be conscious, but it’s there. It’s exactly why we scroll past posts that feel like they were written entirely by ChatGPT. We can tell.
How to Use AI Without Sounding Like AI
AI can be a great assistant if there’s human creativity behind it. Always try to come up with your own ideas for content. AI is helpful for brainstorming (especially if you’re a one-person team and need something to bounce ideas off) but whatever it suggests, put your own spin on it. Make it yours.
One of the best ways to avoid AI’s cookie-cutter content is to start with a messy, human draft. Word vomit everything in your head into a document, even if it sounds terrible. AI also tends to default to generic intros and stale hooks (ahem, “In the ever-evolving world of…”), so if you leave the format up to it, you’re setting yourself up for a forgettable piece.
Once you have a rough draft, then you can use AI to clean it up. But don’t let it rewrite your voice and ideas out of the equation.
When using AI to refine your writing, be careful with vague prompts like “make this better.” AI’s version of “better” often means making everything more formal, longer, and stuffed with unnecessary fluff.
If you end up with something that’s technically correct but still feels… off (that AI je-ne-sais-quoi albeit derogatory), how do you fix it? The first step is spotting the telltale signs of AI-generated text.
AI Red Flags in Tone and Content
Overly formal writing when a casual tone would work better.
Excessive wordiness that doesn’t actually add value (keyword stuffing to meet minimum word limits).
An exaggeratedly engaging-Marketing tone that reads like a dad joke.
AI might use humor, but it’s always the same brand of humor. When something reads as a different type of humor, it’s probably a human piece.
Lack of original insights: if you as a human don’t give AI the “fresh takes” of the content you wish it to generate, it won’t come up with them on its own. AI is, first and foremost, a people pleaser, and it would never have an opinion and risk alienation.
Factual inconsistencies or outright mistakes: more typically, outdated information.
AI Red Flags in Phrasing
AI has some writing habits it just won’t quit. If you see these, proceed with caution:
Mirrored phrasing: “Not only this, but that.” / “It’s not about X, it’s about Y.”
Overused structures: “From A to Z,” “Whether you’re X or Y…”, especially in introductions and descriptions.
No fluff, no big agenda, no whatever–just this other thing.
The problem? This sentence structure.
A lot. Of full stop. Periods. To sound. Like you know. What you’re talking about.
Generic openings: If a piece starts with “In the ever-changing/fast-paced world of [industry]…” just close the tab.
“Picture this” or “Think [scenario]” instead of simply saying “for example.”
AI-generated texts also tend to have adjectives or other qualifiers before almost every noun in a sentence.
Broad language and generic sentences are trademarks of generated content.
Unusual or over-complex sentence construction.
Unnatural phrasing for the intended channel, like using “Furthermore” in a social media post, instead of the simpler “also”.
“Shed light”, “foster change”, "deep understanding", “in the realm of”, “in light of”.
Marketing clichés like “empower”, “unleash”, “unlock”, “elevate”.
Awkward idiom use like “hit the nail on the head”, “bite the bullet”, ”call it a day”, “beat around the bush”, “cut to the chase”, “ahead of the curve”, “break new ground”, “pushing the envelope”.
AI Red Flags in Vocabulary
Some words pop up so often in AI-generated text that they’ve lost all meaning:
“Delve”
“Poised”
“Seamlessly”
“Dynamic”
“Innovation”
“Leverage”
“Essential”
“Craft”
“Embark”
“Cutting-edge”
“Rapidly”
“Harness”
“Bustling”
“Revolutionize”
“Vibrant”
“Moreover” (this one sends shivers down my spine)
“Crucial”
“Underscore”
“Plethora”
“Paradigm”
And of course, “dive into”.
AI Red Flags in Punctuation & Formatting
LOVE of bullet points for everything (we promise this article is an exception).
Excessive bolding for emphasis, even when the intended channel doesn’t support it?
Random caps lock usage.
Divider abuse.
Uses subtitles in virtually anything you ask them to write, even if it’s supposed to be a 300-word-long LinkedIn post.
Overuse of em dashes (tragic, because I used to love them). If you see too many, swap them for parentheses, commas, or colons.
Has a passion for long titles, and the “XXX: XXX” structure while writing them.
Final Thoughts
Once you start noticing AI red flags, it’s hard to stop. The weird phrasing, the forced enthusiasm, the same five verbs pretending to mean something. It all starts to feel like noise.
But that’s kind of the point: if your writing sounds like it could’ve been written by anyone (or anything) it won’t stick.
AI is a godsend in many ways, but specifically in fields content, people want to feel intentionality, presence, relatability. You might have a great post on your hands, but if you run it through ChatGPT and it comes out full of em dashes and “not only this, but also that” phrasing, it’ll read like a lazy afterthought.
That’s the difference between content people scroll past and content that actually lands.
