You are viewing a single comment's thread from:

RE: Reward Pool Rape?

in #hivewatcherslast year (edited)

Ok, i'll follow my own advice & take time out to come to centre & consider before replying again. Not that i feel anger to him or you, but there are some uncomfortable feelings.

Also note this is only the second time (i think) i've ever reported anyone to hivewatchers in 4 years (again, after first giving opportunity for the "offender" to respond).

i also realise that you may have an emotional charge around this matter as due to the "controversial" nature of what you post about, you have very likely received downvotes based on opinions, rather than on chain abuse.

Sat Nam

Sort:  

I actually received DVs only once, (writing about guy having sex with dolphin) and loooong time ago :)

Yes, I know who are the main police officers on Hive and that you are not part of that pack 😃

No hard feelings in the long run on my side. Resentment is a way nowhere. I just feel helpless and disappointed with each case of great people leaving in such cases.

Edit: I also engaged in a DV battle once while defending a person accused of plagiarism.

Can you oint me toward this body of evidence re. AI please.

Did you read this (including the comments of 3 yrs ago? https://inleo.io/@hivewatcher/s7eu0y

https://medium.com/@AmandeepPanesar/the-failure-of-ai-detection-false-positives-in-essay-writing-ff55e868540

This medium article might be a good start, also check references to actual research at the bottom.

Thanks. i did a bit of research this morning. i read a couple of articles:

From this one: https://writehuman.ai/blog/how-accurate-are-ai-detectors#!

Imagine the University of Kansas researchers created a tool that can spot AI-generated content in research papers with an amazing 99% accuracy.

In contrast, OpenAI's Classifier achieves only a 26% success rate and mislabels a human written work as AI-generated content 9% of the time. Another tool, AI Writing Check, predicts its accuracy to fall between 80% and 90%.

So it depends which tool is used, in part anyway.

This one https://www.makeuseof.com/accurate-ai-text-detectors/ has GPTzero (which i just happened to use) at the top of the list.

But still, in the case in question,
a/ from looking at past actions and attitude (from the aforementioned reply by hivewatcher),
b/ from my initial sense of something not being right possibly confirmed by GPT zero with 98% possibility on 1 post & 85% on the other, and
c/ from diggndeepers failure to appeal in HW discord either this time (or presumably last time or in discord (i'm in the room & posted a message) and his actions toward me, i'm inclined not spend take more time on it unless i hear from him.

i've spent several hours over the last weeks monitoring HW discord appeals channel (& other channels) & looking at the posts in question. Having approached it from a neutral stance (even though i'd built up a negative perspective over the years due to all the complaints about HW i've seen), my view now tends toward HW doing not a bad job (recently since i've been looking anyway) and getting lots of abuse from what appears to be clear cases of chain abuse.

incidentally, i removed my downvote from the post last night. Don't know why, but i did.