We have all seen it. That specific style of writing that screams "I copied this from ChatGPT."
It’s polite. It’s grammatically perfect. And it is completely soulless. It uses words like "delve," "landscape," "tapestry," and "unleash." When was the last time you heard a human being use the word "tapestry" in a casual conversation at a coffee shop?
Probably never.
In psychology, there is a concept called the Fluency Heuristic. It states that if information is easy to process (simple words, clear structure), we judge it as being more true and trustworthy. If information is difficult to process (fancy jargon, complex sentences), we subconsciously trust it less.
When you let AI write your LinkedIn posts without constraints, it uses complex, flowery language. It violates the Fluency Heuristic. You sound smart, but you don't sound real. You land in the "Uncanny Valley" of text—close to human, but just weird enough to be creepy.
The 'No More Robot Words' prompt
The solution isn't to stop using AI. It's to give the AI "Negative Constraints." You have to tell it what not to do.
Here is the prompt I use to force AI to write like a founder, not a professor.
The Prompt:
"I want you to rewrite the text below to make it sound like a human conversation.
Constraints:
- Use B1 level English (simple, clear, accessible).
- Strictly avoid these 'AI words': Delve, Tapestry, Unleash, Elevate, Realm, Landscape, Game-changer, Synergy.
- Keep sentences short.
- Write as if you are emailing a friend, not writing a university essay.
Here is the draft to rewrite: [Paste Text]"
Why this works
By forcing the AI to drop its favorite vocabulary, you force it to focus on the message. You strip away the fluff.
At Stix, we believe that your personal brand should sound like... you. That is why our built-in AI rewriter is tuned to avoid these "robot words" by default. But until you are using Stix, use this prompt to keep your LinkedIn feed human.
