Has the AI detector actually changed anything?

So it’s been a little while since that new AI detector went into beta.

Just curious what everyone is actually seeing on their end. I feel like I’ve noticed slightly fewer of those massive copy-paste robot essays, but maybe people are just getting better at hiding it?

Do you guys feel like the spam is actually down, or is there still a lot of work to do?

Overall, is reading the threads better now?

Honestly there's a lot of spam. Usually from accounts that joined in the last few days. They use proper grammar threads, almost nothing looks wrong with them. Yet they come across as people with some expertise or smart newbies, which makes them kind of stand out

i joined the forum a few days, i noticed multiple threads replies are easily identified as AI generated. You'll find that the main thread question or problem get's replicated in the user response a sort of technique that LLMs use to respond to users. the question here is why ? i don't get it

We're obviously keen to hear feedback on how it's working so far. But please do remember that it's very much in its beta stages, we're continuing to refine its detection parameters and it's a long way from perfect and will be for a while yet.