I just watched a video about students getting their papers falsely flagged for using AI, even when they didn’t, and the advice was things like, “Leave in incorrect grammar,” “If you’re quoting something, don’t copy and paste it, type it out manually because it leaves a metadata trail that you used the copy/paste function and that's a flag,” “Write in the cloud so there’s a version history,” and the one that really got me, “if you find you write in a manner that can sounds too robotic or professional and it gets flagged, go to the writing center so a writing tutor can help you sound more humanly flawed,” and like what the actual fuck.
Like I get that is practical advice, but people should not have to fucking do that. They should not have to train themselves around not sounding like AI, when AI only sounds like that BECAUSE it was trained on them.
I spent so much of my life learning how to write, I shouldn't have to unlearn that because some computer algorithm learned from me.
Not to mention, encouraging actual human beings to dumb themselves down 'so they don't sound like AI' only makes AI seem like the superior option, because "human writers are making too many grammar mistakes and sounding stupid".
In this example, students are encouraged to intentionally make mistakes, to intentionally make their writing and papers look unpolished, because turning in something that's well written and revised will trigger the ai checker.
This is yet another example proving that it is virtually impossible to actually check for ai writing. If you're feeding writing into one of these "checkers", you are actually feeding ai and training it to be better.
The program was trained on our writing, and now we're told to write worse so we don't get confused for AI.
This is honestly NOT practical advice, it's horrible advice, likely from people who use or program AI to try and push their 'superior' option.
AI cannot die fast enough.
