User:CanonNi/Essays/We don't unblock AI
 | This essay on appealing a block is in development. It contains the advice or opinions of one or more Wikipedia contributors. Essays may represent widespread norms or minority viewpoints. Consider these views with discretion, especially since this page is still under construction. |
 | This page in a nutshell: We don't unblock users who use AI to make appeals. |
The Guide to appealing blocks says that a user needs to Understand in full the reasons of your block before requesting an unblock
. Some users choose to use generative AI, such as ChatGPT, to write their appeals in order to seem more professional. This is a terrible idea, for several reasons.
Admins can tell
[edit]With the rise of AI in recent years, tools to detect AI, such as GPTZero, have also been created. With a simple copy/paste, admins can tell whether or not you wrote the appeal. Back when I was active in recent changes patrol, I saw dozens of AI-generated requests, and I remember one user who didn't even try, and copy/pasted the "Sure, here is your convincing unblock request..." into the template.
I'm not an admin and don't wish to be one in the near future, but I'm certain that most reasonable admins will not unblock these requests, and, in extreme scenarios, will revoke your talk page access.
AI are not Wikipedians
[edit]The appeals process here on the English Wikipedia is detailed yet easy to understand for the average human. AI, on the other hand, cannot fully comprehend all of the information and therefore makes mistakes.