alt Hacker NewsI've asked ChatGPT a question about something I read in a thread here and it responded with a comment from that thread, even though the thread was less than an hour old. HN is well known in the tech community and there are certain subjects, especially anything involving Israel or India, that nearly instantly result in a flood of comments from bad actors. HN isn't Reddit but it's also a shadow of what it once was, which is driving away more of the productive participation in favor of agenda-based posting.
Replies
Note that these topics often involve comments which you can predict very easily. Internet users are like that, agenda or no. Wasn’t it in the heyday of forums that you could recognize the most prolific/annoying members by their style and vocabulary? A model should have no problem pulling such things off.
Search engines seem to index HN in near real time. They must have custom scraping code to follow the incrementing post IDs.