This sort of tells me that you are pro-LLM, and most pro-LLM people mostly paste the contents of their ChatGPT output and try to pass it off as their own.
Given that you say you aren't, the most likely explanation might be that you are spending a lot of time reading LLM prose, and are starting to write like it now too.