alt Hacker NewsAnything where an LLM asks you a question without disclosing that it's an LLM is an abomination.
I get these on Twitter all the time, and they're starting to show up on Bluesky as well.
Most of my tweets have 20+ garbage spam replies now. I don't read them any more, which sucks because I used to get to talk to actual people there.
The ones that ask a question are particularly offensive because they risk you falling for them and wasting several minutes of your time thinking about and then responding to them... when no human being ever asked or cares about the answer.
Replies
In my experience Pangram has been, if not perfect, then at least good enough for my purposes, and I'm glad that at least the big labs seem incentivised to have text outputs to be clearly identified as written by AI.
Probably doesn't help with Twitter though because individual engagement-baiting messages will probably be too short, but... makes me wonder if we're going to move to something more like an answering-machine internet where we're using LLMs as intermediaries in any and all lower-trust environments (which seems like it would be anywhere strangers can congregate?) or if it's going to be more like proof-of-humanity or just proof-of-attention.
The point of answering questions in public spaces is the public, not the person (or LLM) posing one?
While faking human interaction possibly should be prohibited outright, one cannot ever be sure that is being adhered to? And humans can be disingenuous as well to begin with.
It's how you engage in a discussion and conduct yourself that matters, because other people are there witnessing it.
Now, when that LLM fakes human sentiment and you or others fall for it, that is a problem beyond the use of LLMs? The same would be true with sock puppets operated by real people, only the scale differs.
When you solve that problem, by putting high value on the factual and logical truth of statements for instance, you solve multiple problems at once.
> Most of my tweets have 20+ garbage spam replies now. I don't read them any more, which sucks because I used to get to talk to actual people there.
Genuinely curious: Is it the same when you post on Mastodon?
(I've occasionally interacted with you there).
What annoys me is people calling others LLMs or bots without proof. I've seen literal friends who I know aren't botting or using LLMs to post on social media being accused of it. It's just gotten out of control. Meanwhile captchas get more ridiculous, soon you'll need to hire a bot to get in.
Not that this will help, but I saw one guy on Twitter who was actively selling those engagement question replies as a service. I know Levelsio was working on a shared blocklist to try and stop them, and you'd think Elon could just block this MagnetX service directly on the X side.
https://x.com/aksenHQ/status/2118994144945389678