We may or may not be able to detect whether LLMs wrote something, but I'd bet my life savings that when an administrator at the department of agricultural sciences and natural resources asks ChatGPT whether a particular essay was written by GPT, there will be significant numbers of false positives and false negatives.
Well, that notion is wrong, otherwise, what's the explanation for something like this:
https://www.theregister.com/2023/05/17/university_chatgpt_gr...