I don't like talking to AI, I'm definitely not going to argue back and forth with it or read through AI slop, but there are also legitimate uses for it as some have explained. I suppose as long as AI isn't simply being used to spam and the posts don't otherwise violate the forum rules, I really don't see zos actually spending resources trying to determine which posts were or were not written using AI.

And some people actually do write in a similar style to AI; it's not always easy to tell. (In many cases it's quite obvious though).

But overall if people have a point to make and want to actually convince others, I encourage them to try and write in their own words and stick to relying on AI as a tool to create an outline that one fills in with their own thoughts, and/or to clean up one's writing after the fact if they struggle with it or if, for example, English is not their primary language.

As someone who essentially argues for a living and now has to deal with interns using AI, I will say that AI does an atrocious job: it frequently overlooks important details or favors random, trivial points over convincing ones; it has trouble staying consistent and will happily argue against its own points or change its arguments around to suit the prompt; and it hallucinates and needs to be fact checked.

(Or not the same as an outright hallucination, but it pulls from unreliable sources - particularly for something like ESO, where there are limited verified or "official" sources for AI to use - in playing around with it, I have found that it often takes bits and pieces from forum posts or reddit and treats them as fact).

Not to mention it excels at telling the user exactly what they *want* it to tell them, thus acting as an echo chamber instead of encouraging one to evaluate their own ideas or listen to constructive criticism.

Maitsukas wrote: »

It becomes a problem if the AI starts bringing up responses that were made by the AI itself earlier and having someone behind the keyboard copy-pasting them here.

AI does it if a real person allows.
It is up to a human, what part they pass to AI and what part they do on their own.

Paramedicus wrote: »

i guess it is just one of these murica things to think everyones first language is english

There’s other forum languages. And people who have posted using a translator or their best shot at English and it’s not like they’re impossible to understand.

Edit: Personally, I’m quite fond of talking to people who clearly aren’t English first language because it’s a wonderful experience of human connection across cultures. I’m not pleased how much human brainpower is being outsourced to AI.

I hate Generative AI more than most people here probably do, but the problem with this is, I've seen far too many innocent people accused of it just because their posts are written with proper formatting and such. Like if you're well-written, you'll often get accused of using AI. Only short, emotional, not-well-worded posts actually DON'T get accused.

AI learned from well-written people, authors and the like. So yes, these very same people are going to sound like the AI that stole from them.

Bguk wrote: »

whitecrow wrote: »

spartaxoxo wrote: »

whitecrow wrote: »

Make them prove they're human.

It hasn't been by bot accounts thus far. Regular humans with consistent and years long posting history.

I guess I don't understand. Why would a person use AI to make a post?

One poster I am aware of has some cognitive issues due to an injury if I remember correctly. Their AI posts are way clearer than their own typed posts.

There are countless reasons for using ai to format and formulated comments including the above, language barriers, grammar deficiencies, etc. These are all valid reasons to use ai to generate a statement that YOU then post.

whitecrow wrote: »

spartaxoxo wrote: »

whitecrow wrote: »

Make them prove they're human.

It hasn't been by bot accounts thus far. Regular humans with consistent and years long posting history.

I guess I don't understand. Why would a person use AI to make a post?

i was hit by 10 tons ( DRILL RIG BLOCKS ) in the head so i use it to format my thoughts as when typing them my thoughts and words get twisted and things comes out scrabbled , and after being told to use AI for help am able to make my post legible so that my views can be better understood

as one that uses AI i put that im using it and why , and i believe that should be the general rule for anyone that uses it ..

What next, no Google?
Better artificial intelligence than no intelligence.

whitecrow wrote: »

Bguk wrote: »

i was hit by 10 tons ( DRILL RIG BLOCKS ) in the head so i use it to format my thoughts as when typing them my thoughts and words get twisted and things comes out scrabbled , and after being told to use AI for help am able to make my post legible so that my views can be better understood

I'm glad it helps you. I did not know AI could be used that way.

some times i have to refined what it says but thats where reading what is being put out 1st matters more then the use of AI as even it gets crossed up on some things

It's comical how ingrained in the human psyche witch-hunting is.

DenverRalphy wrote: »

TheMajority wrote: »

I don't think AI should be allowed, ever, and that people should not rely on it. I do not care, you get better at speaking other languages by practicing it, not by using crutches. AI is bad for your mental health and mental strength, and it weakens the mind and thoughts. You should think for yourself, and use your own words.

Future generations will be mentally weak if we keep this up. Already we see the effects of it on the minds of adults.

The same was said about calculators and personal computers.

Yeah, except calculators aren't causing people to become mentally ill.

ETA: And actually, calculator use excessively discourages you from memorizing math skills, so yes, it creates mental weakness.

Four_Fingers wrote: »

It is a research tool.
Just don't copy and paste.
I can remember when people thought word processor programs was cheating. Oh and no spell checkers either.

you should actually do some research on what AI does and how insidious it really is before you call it a "just a research tool."

Four_Fingers wrote: »

Oh, how is it insidious?
It has no original thought yet, it relies on data it has learned from previously posted data, mostly from the Internet and pieces it together.
But I understand, harder for old people to accept change.

Strawman arguments and insults don't make you right.

Again, do your due diligence and actually research the subject. I'll wait.

Lekjih wrote: »

I'm generally of the opinion that if you can't be bothered to write it, I can't be bothered to read it.

I understand there are times people cannot tell. I've been accused of making ai crochet videos before because someone didn't know that soft sculpting exists. It's not an enforcable rule, but if it were enforcable I would disallow it.

A fellow crocheter! I understand how you feel, AI has become increasingly troublesome in the crochet community. I take it you also do Amigurumi, since you mentioned soft sculpture?

I got accused of using AI because my stitches were too "clean" looking on my last project. I was using the yarn under technique rather than yarn over, and I'm very meticulous about it, so the rows were "too neat" according to them. I don't touch AI at all when it comes to my hobbies and crafts, so it was rather irritating. I ended up taking new "Imperfect" photos using my phone rather than my lightbox and good camera to prove that the pattern I'd written was real >_>

AI scams are definitely a thing within the crochet community, so I don't blame people for being cautious, but people who have been doing the craft for years are getting accused of using AI, when they never would. And the reality is, generative AI is stealing from their product images and style.

Which raises a question in regards to the forums. How do people plan on proving that a post was written with AI? While I don't support AI and view it as a nuisance, especially within creative communities, all of the tools available to prove that something was written with AI are extremely flawed. They'll flag posts/stories written without AI assistance as "possible AI" simply because someone used a particular turn of phrase, has a decent vocabulary, or god forbid, an em dash (which is a perfectly valid piece of punctuation I've had to train myself out of using, simply to avoid false accusations of AI use >_> )

While I have ethical concerns regarding the use of AI, I can't rightly condone an AI "witch hunt" on the forums, simply because innocent users will ultimately end up in the line of fire, along side those who use it as an accessibility tool. Yes, in some cases, it's very easy to tell when AI language is being used. But there are those who will look at any lengthy, detailed post and immediately come to the conclusion that it -must- be AI, because they themselves lack the time and patience to write out their thoughts.

AI detection tools are getting better, but they still have a high false positive rate. Do we really want something else cutting off conversations on these forums, aside from the rules themselves, which are already difficult enough to navigate?