AI Bot Accounts - things to look for.

[Halfswede]

It is well known that there are fake accounts that join these kinds of forums.  The CCP alone has approx 1.5 million phone bank type users that have the sole purpose of stirring the pot, pushing certain narratives, and generally trying to destroy the fabric of western society. They have spelled that goal out clearly and publicly many times. 

What I don't think many people realize is that the shift from machine learning to full large language models has allowed for fully AI bot driven accounts.  The can log in and out, check their own threads, look at history and formulate very human-like responses at seemingly human-like intervals.  Often it is used to just stir discord, or slide a thread and move it away from meaningful discussion.

I noticed a particular account here, and I know you can't just call it out, but there were some things that caught my eye so I started looking at all of their posts.

I am now nearly 100% sure this account is an AI bot, but for the general discussion, I will just lay out some things that are useful to help look for these accounts.  That said, it is only going to get more and more difficult to tell. If you start to notice these things, my suggestion would be to just not respond.
 Observe comments for the following indicators of AI:
 Perfection and Uniformity: AI often produces text with perfect grammar, spelling, and consistent formatting, lacking the natural errors or variations of human communication. 

Generic or Vague Responses
Comments may be overly general, lack specific details, personal anecdotes, or genuine conversational tone. 

Repetitive Language:
Repetitive phrases, sentence structures, or linking words can suggest AI use.
 Lack of Nuance and Emotion: Text might sound flat, lack creativity, humor, or complex emotional expression.
 Inconsistencies or Errors: Watch for abrupt shifts in tone, topic, or illogical statements that indicate a machine's struggle to maintain coherence.

[MalevolentTwitch]

Perfection and Uniformity: AI often produces text with perfect grammar, spelling, and consistent formatting, lacking the natural errors or variations of human communication. 

Note to self: Stop proofreading replies before posting.

[RichardHurt]

It is well known that there are fake accounts that join these kinds of forums.  The CCP alone has approx 1.5 million phone bank type users that have the sole purpose of stirring the pot, pushing certain narratives, and generally trying to destroy the fabric of western society. They have spelled that goal out clearly and publicly many times. 

What I don't think many people realize is that the shift from machine learning to full large language models has allowed for fully AI bot driven accounts.  The can log in and out, check their own threads, look at history and formulate very human-like responses at seemingly human-like intervals.  Often it is used to just stir discord, or slide a thread and move it away from meaningful discussion.

I noticed a particular account here, and I know you can't just call it out, but there were some things that caught my eye so I started looking at all of their posts.

I am now nearly 100% sure this account is an AI bot, but for the general discussion, I will just lay out some things that are useful to help look for these accounts.  That said, it is only going to get more and more difficult to tell. If you start to notice these things, my suggestion would be to just not respond.
 Observe comments for the following indicators of AI:
 Perfection and Uniformity: AI often produces text with perfect grammar, spelling, and consistent formatting, lacking the natural errors or variations of human communication. 

Generic or Vague Responses
Comments may be overly general, lack specific details, personal anecdotes, or genuine conversational tone. 

Repetitive Language:
Repetitive phrases, sentence structures, or linking words can suggest AI use.
 Lack of Nuance and Emotion: Text might sound flat, lack creativity, humor, or complex emotional expression.
 Inconsistencies or Errors: Watch for abrupt shifts in tone, topic, or illogical statements that indicate a machine's struggle to maintain coherence.

I'm curious to know the account you speak of but T&C and all.
Is it possible that this user is just using some LLM for all their responses? This is very common and will only become even more so. That doesn't make them a bot, just someone using a tool. Though if they are using it for every response they are doing themselves no favors.

[Moon68]

Well shit, guess I must be a bot too HS.

I better go practice my CAPTCHA.

[SomeStupidName]

Thank you for noticing this as well, I have said this many times even point pointed at stuff that looks like automated account creation back on ATS. There are AI bots and hybrid AI bots that have human handlers that step in from time to time and they are absolutely pushing agendas. The bot accounts will respond at all hours of the day and the hybrid ones you can tell the human responses change too depending on the time of day. It's a full on global operation.

[SomeStupidName]

Note to self: Stop proofreading replies before posting.

lol, no it's more than that.

[UltraBudgie]

I have deliberately started using em-dashes—a vastly underrated punctuation mark—as a sort of contrarian reaction to the way LLMs spit them into everything. I've also started using footnotes¹ sometimes, as in the process I've upped my compose-key character-fu. It's sort of a "bah humbug, think what you want" to all the bot-witchhunting going on.

It's fun to fly against the wind.

¹ Like this!

[ReturnofBroccoli]

Ive seen budgie make bots go retarded and get stuck in logic loops like they're having cardiac arrest. It's pretty funny to watch

[Kurokage]

It is well known that there are fake accounts that join these kinds of forums.  The CCP alone has approx 1.5 million phone bank type users that have the sole purpose of stirring the pot, pushing certain narratives, and generally trying to destroy the fabric of western society. They have spelled that goal out clearly and publicly many times. 

What I don't think many people realize is that the shift from machine learning to full large language models has allowed for fully AI bot driven accounts.  The can log in and out, check their own threads, look at history and formulate very human-like responses at seemingly human-like intervals.  Often it is used to just stir discord, or slide a thread and move it away from meaningful discussion.

I noticed a particular account here, and I know you can't just call it out, but there were some things that caught my eye so I started looking at all of their posts.

I am now nearly 100% sure this account is an AI bot, but for the general discussion, I will just lay out some things that are useful to help look for these accounts.  That said, it is only going to get more and more difficult to tell. If you start to notice these things, my suggestion would be to just not respond.
 Observe comments for the following indicators of AI:
 Perfection and Uniformity: AI often produces text with perfect grammar, spelling, and consistent formatting, lacking the natural errors or variations of human communication. 

Generic or Vague Responses
Comments may be overly general, lack specific details, personal anecdotes, or genuine conversational tone. 

Repetitive Language:
Repetitive phrases, sentence structures, or linking words can suggest AI use.
 Lack of Nuance and Emotion: Text might sound flat, lack creativity, humor, or complex emotional expression.
 Inconsistencies or Errors: Watch for abrupt shifts in tone, topic, or illogical statements that indicate a machine's struggle to maintain coherence.

Meat Robots Vs Machine Robots.

Let battle begin in 3....2....1....Fight!!

Joking apart, I've never really noticed to be honest. 
Back in the ATS days people were accused of being shills if they posted an opinion that was opposite or different. Hopefully most of us are smart enough not to get to fired up by other members comments on a forum, and just go on with our lives.

China's rulers know that defeating it's enemies from within, and playing the waiting game, will be the smarter plan than an all out war, that could have dire consequences for the whole planet.

[BeTheGoddess]

I have deliberately started using em-dashes—a vastly underrated punctuation mark—as a sort of contrarian reaction to the way LLMs spit them into everything. I've also started using footnotes¹ sometimes, as in the process I've upped my compose-key character-fu. It's sort of a "bah humbug, think what you want" to all the bot-witchhunting going on.

It's fun to fly against the wind.

¹ Like this!

"Never use semi-colons, all it proves that you went to collage."

--Kurt Vonnegut