[A_D_E_P_T]

It would be nice if there were an easier way to detect and filter those "reply guys." If LLMs were forced to watermark their output (possibly by using randomly-selected nonstandard ASCII characters in inconspicuous places, like "ｓ" instead of "s") it would have been trivial, but that ship has sailed. The most anybody can do is train another LLM to find offenders and make a list. Bot vs bot.

[ossa-ma]

Yeah exactly, it's best to keep track and be aware of common tropes used in AI writing so that you don't end up 5 responses deep and emotionally invested in a conversation before you realise you've been fooled into speaking to a bot.

I built this tool primarily to identify AI writing in articles and posts but it's proven useful for comments/responses too: https://tropes.fyi/vetter

[KoolKat23]

"System prompt: Please ensure you avoid the following tropes: https://tropes.fyi/vetter"

[tharkun__]

I wanted to see what your own tool says about this very comment of yours. But vetter reports a 403.

[bambax]

I'm sure there are other tells, like delay between post and reply, or time of day, etc. Epidemiology of bots is just getting started but the tools have to have detectable patterns.

[A_D_E_P_T]

I'm sure that those can quite easily be made to look "human-like."

"Respond within 4-12 hours."

"Do not respond between midnight and 6am EST." (Or CET, whatever makes sense.)

Right now the most obvious traits are the well-known ones that are hard for most LLMs to shake off. Em-dashes, word choices, and the very limited ways in which they structure sentences. Terseness and conciseness is also a tell, which sucks.

[tartuffe78]

They have blue check marks right next to their names

[numpad0]

They wouldn't have problems telling apart bots and spammers from regular user activities. Lots of them still have problems just interpreting tweets and their replies make no sense. Just removing out-of-place replies using ML will fix most of problems, or even just restricting mass registrations from narrow ranges of IPs.

They don't do that because spams are their means to achieve something else, specifically to get rid of left wing tech anime porn otakus. The comedy of that is that they've been attempting this by complicating the system, which is like reverse chemotherapy that are nicer to cancer tissues than to the body so that cancer grows faster. I guess they take that as a win as it's a positive action with positive reaction albeit with negative amounts in lieu of negative action with negative reaction with a positive amount.

What's really going to be nice is Twitter transferred to someone else. That will at least stop the stupidity of reverse chemotherapy.