Does My Post Sound Like AI? How to Tell — and What to Do About It.

The fear is legitimate

You used ChatGPT or Claude or some tool to help draft a LinkedIn post. Now it's sitting in front of you and something feels off. It's perfectly competent. Grammatically flawless. And it sounds like it was written by a very polite robot.

You're not imagining it. AI-generated text has a recognizable feel, and people are getting better at spotting it. Not because they're running detector tools — but because they've read enough AI-sounding content to develop an instinct.

The concern isn't just aesthetic. LinkedIn's 360Brew algorithm evaluates content substance and authenticity. And your network — the real humans reading your posts — can tell when something doesn't sound like a person wrote it.

The telltale signs

AI writing has patterns. Once you see them, you can't unsee them:

It uses "landscape" and "leverage" and "navigate" and "delve" at rates no human ever would.

Every paragraph is exactly the same length. Every sentence follows the same rhythm.

It hedges everything. "It's important to note that..." "While there are many factors to consider..." AI never just says a thing — it qualifies it into oblivion.

It lists three things. Always three. "This improves engagement, builds trust, and drives results." Real people don't naturally structure their thoughts in triads.

It has no rough edges. No half-finished thoughts. No weird phrasing that's uniquely yours. It's smooth in a way that real writing isn't.

And the biggest tell: it has no specific details. AI can write about "leadership challenges" all day. It can't write about the specific meeting last Tuesday where you realized your team doesn't trust you.

Why it matters on LinkedIn specifically

LinkedIn isn't a blog. It's a feed full of people who know each other — or are one degree away from knowing each other. When your colleague from two jobs ago reads your post, they know how you talk. They know your vocabulary. They know your perspective.

If the post doesn't sound like you, the best case is they scroll past it. The worst case is they think less of you — not for using AI, but for posting something that has your name on it and no personality in it.

There's a growing body of research suggesting AI content is becoming easier to detect, with some studies showing people can identify AI text at rates above 60-70%. But the real detection system isn't an algorithm. It's the people who know you.

How to fix it

The goal isn't to write without AI. It's to make sure the final post sounds like you, not like a language model.

Start with your own words. Talk through the idea out loud. Record a voice memo. Write an ugly first draft in your own voice. Then use AI to clean it up — not to create it from scratch.

Add a specific detail that only you would know. The name of the project. The city you were in. The number of people in the room. AI can't invent these details, which is exactly why they make your writing sound human.

Cut the qualifiers. If the post says "It's worth noting that effective communication can significantly impact team dynamics," replace it with what you actually mean: "I started being more direct in 1:1s and my team started being honest with me."

Read it out loud. If you wouldn't say it to a friend over coffee, rewrite it until you would.

The goal is you, just clearer

The best use of AI for writing isn't replacement. It's refinement. You have the idea, the experience, the perspective. AI can help you organize it, tighten it, fix the grammar.

But the voice has to be yours. The details have to be real. The opinion has to be one you'd defend in person.

That's the difference between using AI as a ghostwriter and using it as an editor. A ghostwriter speaks for you. An editor helps you speak better.