Can we replace human moderators with AI?

Status
Not open for further replies.
Ascribing fear to people who don't agree with you is a bad look. Algorithms and models are good for lots of things. Moderating human interactions isn't one of them.
We said the same thing about many things before. And it got proven wrong time and time again.


If this forum is one of the first, they might even be able to sell the data from their AI mods to other forums.
 

log in or register to remove this ad



God, no. No thanks. AI is not impartial. It's hard to pin down what its prejudices are, but it's illuminating to look at, for example, how it has judged beauty- lightness of skin seemingly being a big factor for it. It carries the baggage of the information that has trained it, and that's definitely not always a good thing.
 

The purpose of mods is not to be popular, but rather to help the owners/admins of a forum realize their vision for the forums, IMHO.
Well the people on forums and what they write is the content and the customers.


So keeping them happy and posting as much useful stuff as possible (to sell the data to AI companies) and visiting the site as often as possible (to looking at advertisement to generate money) should be the prime goal.


So in that sense ideal moderators should be slaves to the normal posters and trying anything to keep them happy. They are what keeps the forum alive.
 

God, no. No thanks. AI is not impartial. It's hard to pin down what its prejudices are, but it's illuminating to look at, for example, how it has judged beauty- lightness of skin seemingly being a big factor for it. It carries the baggage of the information that has trained it, and that's definitely not always a good thing.
Right, like, if we're training it off of our posts, that's not good. Like, I post here, and all my posts are objectively garbage. Before we'll know it it's going to start red texting any thread not about watching Knightrider in their underpants or whatever.
 

God, no. No thanks. AI is not impartial. It's hard to pin down what its prejudices are, but it's illuminating to look at, for example, how it has judged beauty- lightness of skin seemingly being a big factor for it. It carries the baggage of the information that has trained it, and that's definitely not always a good thing.
Well if the beaty standard is that skintone X is more beautiful, then the AI is not wrong.


I mean diferent countries have different preferences. For example in Europe not being in the sun was really popular in the past, until Sissy changed that. Nowadays being "sunbathed" is considered more beautiful..


Of course AI can also have biases, based on training data etc. that will always be the case. But thats what we as users could stear by voting on moderators and their decisions.
 


We said the same thing about many things before. And it got proven wrong time and time again.


If this forum is one of the first, they might even be able to sell the data from their AI mods to other forums.
You seem to be under the impression that LLMs have thoughts and make judgements. They don't. They are advanced predictive machines. They aren't what you think and can't do what you want.
 

Right, like, if we're training it off of our posts, that's not good. Like, I post here, and all my posts are objectively garbage. Before we'll know it it's going to start red texting any thread not about watching Knightrider in their underpants or whatever.
post someone GIF
 

Status
Not open for further replies.
Remove ads

Top