Is it just me or all these replays arriving in the last minutes simultaneously looking oddly similar in writing style look like some kind of AI attack testing the site by creating fake accounts??
It’s not just you.
I’ve taken some actions silencing some accounts that seem to have all been created around the same time and shared IP addresses. Keep reporting. I silenced them but I can delete and remove their posts
Moderators, don’t just silence them, but simply report as spam, delete the user and block their IP address. I just did that to a few such posts.
Going forward, we need to moderate all first posts.
Ok, I just did that in settings:
And if AI gets better that we genuinely can’t tell from the first post, we can moderate the first several posts.
Slightly more work for moderators, but existing users will not be spammed, which is much more important. We had to do this with spammers in mailinglists a long time ago, and that pretty much takes care of that.
We can’t have many AI accounts posting and pretending to be human, but skimming the AI posts, the content is not always bad, and it may be worth having an option to see responses from AI identified as such. Stack Overflow volume declined 99% because people prefer AI answers (which are distilled from Stack Overflow answers in some cases).


