Regardless of whether AI detectors work or not, I worry we're reaching a new unfortunate era. While alarmists were worrying about alignment or evil AIs, the tru

Regardless of whether AI detectors work or not, I worry we're reaching a new unfortunate era. While alarmists were worrying about alignment or evil AIs, the true downside of LLMs is turning out to be... AI spam. I've started seeing it everywhere: Q&A sites, forums, etc. In some places it's banned, but how do you reliably detect it? And what makes people post AI spam when there's no reward or money involved? What are they trying to achieve? I've seen it in Q&A sites (sometimes tool specific, I won't name names) where supposed "experts" are simply pasting LLM-generated crap. All the tell-tale signs are there, often including the famed "I apologize for making this mistake. You're right the solution doesn't work because [reasons]". Note it's often not an official AI-generated answer (like, say, the Quora AI bot), but someone with a random human-sounding username posting it as the answer. There are no rep points or upvotes involved, so it boggles the mind... why do they do it? I don't know if HN has some sort of AI filter, but I bet we'll start seeing it here too. Instead of talking to other humans, you'll discuss things with a bot. I predict the arms race between AI spam and AI detectors will only get worse, and as a result it'll make the internet worse for everyone.