Local group on Facebook for finding houses and apartments, and hundreds upon hundreds of comments on every single post for the past couple months are completely AI generated scam spam links. They are probably using the API version of GPT4o, and break even off of scamming enough people.
You must log in or register to comment.
Not sure if this is AI or some dumb and aggressive script.
Imo, this is most likely AI. Facebook and Instagram both have incredibly sophisticated spam preventative measures. These companies are likely using artificial intelligence to generate brand new and highly intelligent scripts that continuously change to evade spam detection tools. Same thing happened on Twitter (You can look at my post history for an example). There was a bot spamming on Twitter and it had run out of credits for GPT resulting in the AI instructions to start being spit out instead of the fake messages it was supposed to produce.
Well the name checks out, at least. You’re ascribing far too much capability to spammers. And that Twitter thing was an obvious fake.
No, actually, I did my own research on it
https://www.npr.org/2024/05/14/1251072726/ai-spam-images-facebook-linkedin-threads-meta
When researchers at Georgetown and Stanford universities investigated more than 100 Facebook pages that routinely post AI content — sometimes dozens of times a day — they found that many are engaging in scams and spam.
Other people have found the same conclusion that I have
Yeah I doubt those particular comments have anything to do with “AI”. It just seems fashionable to blame AI for absolutely everything nowadays
This looks more like classic spambots than AI.
Text is identical, this is not AI, more like a cheap autoposting script. If AI was actually used, it could generate wildly different text every time, it could even use what user said in some post as a context to suggest something to him. That would actually be the only reason to use AI here.
Fucking hell. The word “AI” really lost all fucking meaning huh? Not just is it not applicable to LLMs, it’s especially not the correct term to use for ordinary spam bots, which exist for decades by now.
