How much content is AI generated on Sanctioned Suicide and do we observe a trend of more AI content on here?

[noname223]

I could not say whether we reached a new peak of AI content on here. I think there were crackdowns of threads only created by AI with no addition of the OP.

How do you think should this forum handle AI content? I like this forum because people on here are real.
Personally, I am asking AI sometimes to give me feedback for improving my language skills when creating threads. I am not a native speaker and my English got so much worse since I left school.

It is strange to think that faking an identity is now way easier with the help of AI. You don't have to think for yourself. You can let AI chatbots write you a sad story and that's it. Though, I am not sure whether people who would do something like that would't prefer to think for themselves. But I think I have issues to understand the motives of such people.

I like to get feedback from AI in my daily life. I think sometimes the feeback is pretty good but you always have to be very critical and don't take it without a grain of salt.

I think though it can be pretty annoying if people online use AI for debating other people. Personally, I like politicial discussions. I had one with a friend today. I would find it boring and stupid to use AI for political discussions. Whereas I think using google is valid for example in order to search for a quote or data.

I think I am not good in spotting AI content on SaSu. You get the feeling over 50% of the content on X is already AI generated.
I have the feeling there are only a few threads created with AI on here. And the ones I mean make it explicit. The format can make a post suspicious. My guess would be 2-3% of the text on here is generated by AI. I could completely be wrong. It could be more. But it is up for a debate how to define generated by/with the help of AI.

 

[thatonegirl]

I think your constant use of AI is making you unnecessarily critical of real people. What was AI trained on? Real writing. Every single thing that an AI can produce can also be produced by a real person. You literally cannot look at what someone has written and determine if it was AI just by looking at it. Humans are fully capable of producing exactly what AI produces.

 

[sanctionedusage]

I haven't seen an original post here that looks AI, excluding isolated comments responding to a difficult question or argument with gen AI... and even that's pretty rare ime

AI and AI-modified writing is really easy to spot in a big enough sample size if you're familiar with llm's yourself. cadence and wordslop mostly; its very different from human wordslop: people just trying to sound intelligent, overly formal, poetic, etc.

it gets harder and harder to pass ai as human writing the longer the posts are, as they tend to be on sasu, compared to one liners on x, reddit comments, youtube..