10 signs of ChatGPT-generated text

Based on my rough estimate, at least 20–40% of the materials I read daily are generated by ChatGPT, Gemini, Claude, or other GenAI engines. I have no issues with people using GenAI to generate content (I heavily use it myself for new texts/code/ideas/images), but I'm often annoyed by specific "traits" of these texts, which shout "I'm generated by ChatGPT".

Below are 10 signs how I spot GenAI-generated texts. Please note, I’m not a native speaker, so my observations are most valid for business English used by fluent speakers.

Pro tip: Copy this article into ChatGPT or Gemini and say: "Avoid these signs when you write all materials for me". That will help your texts sound more human (... rephrasing old joke - Being human is the key to success - so it's critical that AI can fake it).

Your thoughts are welcome! Happy to extend this list, and make ChatGPT sound more human.

My list:

1. use of "—" instead of "-"
2. formatted numbered lists in emails – not handwritten "1)", "2)", "-" but auto-inserted using the formatting button
3. use of "here" before lists – e.g., ChatGPT would say "Here is my list" instead of just "My list:" like in this post
4. too complex constructions at the end of a sentence – instead of "let me know if you're interested", you get "let me know if this would be of interest"
5. preposterous nouns/adjectives – "extensive" instead of "broad", "truly inspiring" instead of "very inspiring", "largely aligns" instead of "fits"
6. complex verbs not typical in messages – "showcasing", "demonstrating", "highlighting", "articulating", "notice"
7. AI-style complex intros – "In today’s fast-paced digital landscape..."
8. Overly polite sign-offs – "Wishing you a productive end to your week" instead of "Have a great weekend"
9. Too much explanation for simple phrases - e.g., instead of "AI-style complex intros" using "AI-style complex intros that are uncommon in everyday life"
10. Use of quotation marks with dots and commas - e.g., "this phrase." I very rarely see quotation marks in non-AI text, and especially putting the correct dot/comma before the closing quotation mark

P.S.1. The answer to your obvious question - of course, I used ChatGPT and Gemini to check this text and get new ideas about "signs of GenAI". Unfortunately, these engines were not so efficient in this case! You still need a live human to understand the difference between GenAI-generated and human-generated text.

P.S.2. Obviosly, the image is GenAI-generated.