Do any of your fish ever disappear and reappear?

[j.w]

Its about training the AI language models. They are not doing this for pond knowledge. It is to train these AI models how humans converse and how AI can be more human. So online forums are a great way for the models to learn, because there are discussions back and forth. Reddit is a big source for learning. This is an open forum that is free so it is perfect for these AI models. Its a big issue for online media companies like WSJ, NYT, WAPO because their content is being used without compensation.
A quote from one article:

This battle exists because of the way AI chatbots are made. The so-called large-language-model algorithms that power these bots have to be trained by taking in and processing oceans of existing language to try to mimic what humans say, and how they say it. This isn’t the kind of data we’re used to thinking about as a commodity on the internet, such as the behavioral and personal information used to target ads by companies like Facebook parent Meta Platforms.

This data is the creative output of the human users of various services, such as the hundreds of millions of posts by Reddit’s users. Only on the web can you find sufficiently sized repositories of such human-generated words. And without it, all of today’s chat-based AIs and related technologies wouldn’t work.

So what you are saying is that they aren't really going to hurt us here? And if that is the case why do we kick them out? Are they just a bother and take up space? Should we communicate here w/them? I was wondering about all of this also like @Lisak1

 

[Pablo]

So what you are saying is that they aren't really going to hurt us here? And if that is the case why do we kick them out? Are they just a bother and take up space? Should we communicate here w/them? I was wondering about all of this also like @Lisak1

I am definitely no expert on AI. I noticed last year that there were quite a few random posts for old threads. Around that time I read a couple of articles that detailed how AI training happens. They mentioned that online forums are used because they have so much data that is available. I dont think we really have to worry about anything here, as it is not like we post personal information. Media companies/publications are upset because the AI models are using information that those organizations created. Before when you did your google searches you would get links to relevant articles, now you may notice that at the top of the search results you get the AI answer and no link to articles, the AI answer is potentially using copyright information, hence the fight.
It is not to say every new post here is AI. What I notice is when it is an old thread that seems very random and they do not reply to any responses we make. They also generally make statements, they are not asking questions, usually if it is a person that is searching on the internet about ponds for a particular issue they find a thread (even an old one) and will post a question. This thread seemed very random as there are hundreds of threads about missing/dead fish, so why choose one from 15 years ago. But like @Lisak1 said it is nice to see old threads with members that used to post quite a bit.

 

[Lisak1]

AI has its place I'm sure, but it makes me question everything I see, read, or experience. Lately I've noticed on Facebook that multiple accounts will post the exact same picture and post - for example, I saw a post from "Ina Garten" about her favorite soup from her childhood. Not long after - "Pioneer Woman" posts not only the same recipe but the same picture and the same story to accompany the recipe. Very weird. Pretty much lost all interest in social media as a result. Nothing is real.