It's a great day to look at a spammy network that's been tweeting repetitive political tweets for the last year. Plot twist: many of the spammy repeated tweets are anti-Biden, but the network mostly retweets pro-Biden accounts. #ThursdayAstroturf
This spam network consists of (at least) 1167 similarly-named accounts created between January 2022 and January 2023. Thus far, all of their tweets were allegedly tweeted via the Twitter Web App.
For the first few months of 2022, the accounts in this network only retweeted, producing no tweets of their own. This changed in early May 2022; from that point forward, the network posted roughly equal numbers of tweets and retweets on any given day.
The network's tweet content is highly repetitive, with many tweets duplicated verbatim by dozens or hundreds of accounts. Negative tweets about Joe Biden and China are a recurring theme, although other topics turn up as well.
Despite the frequent anti-Biden stances seen in the network's own (repetitive) tweets, the network retweets a lot of pro-Biden accounts (including @JoeBiden himself). There are a few exceptions, such as @Jim_Jordan, @dbongino, and @elonmusk.
Tweets and retweets make up 99.9% of this network's content. The remainder is a mix of empty quote tweets (possibly a glitch in automation or a human spammer accidentally hitting quote tweet instead of retweet) and repetitive pro-Serbia replies from a handful of accounts.
As is often the case with spammy astroturf networks, many of the accounts in this network use plagiarized profile photos. TinEye and Google were roughly equal in their ability to track down other occurrences of the images used by the accounts in the network.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Just for fun, I decided to search Amazon for books about cryptocurrency a couple days ago. The first result that popped up was a sponsored listing for a book series by an "author" with a GAN-generated face, "Scott Jenkins".
cc: @ZellaQuixote
Alleged author "Scott Jenkins" is allegedly published by publishing company Tigress Publishing, which also publishes two other authors with GAN-generated faces, "Morgan Reid" and "Susan Jeffries". (A fourth author uses a photo of unknown origin.)
As is the case with all unmodified StyleGAN-generated faces, the facial feature positioning is extremely consistent between the three alleged author images. This becomes obvious when the images are blended together.
The people in these Facebook posts have been carving intricate wooden sculptures and baking massive loaves of bread shaped like bunnies, but nobody appreciates their work. That's not surprising, since both the "people" and their "work" are AI-generated images.
cc: @ZellaQuixote
In the last several days, Facebook's algorithm has served me posts of this sort from 18 different accounts that recycle many of the same AI-generated images. Six of these accounts have been renamed at least once.
The AI-generated images posted by these accounts include the aforementioned sculptures, sad birthdays, soldiers holding up cardboard signs with spelling errors, and farm scenes.
The common element: some sort of emotional appeal to real humans viewing the content.
As Bluesky approaches 30 million users, people who run spam-for-hire operations are taking note. Here's a look at a network of fake Bluesky accounts associated with a spam operation that provides fake followers for multiple platforms.
cc: @ZellaQuixote
This fake follower network consists of 8070 Bluesky accounts created between Nov 30 and Dec 30, 2024. None has posted, although some have reposted here and there. Almost all of their biographies are in Portuguese, with the exception of a few whose biographies only contain emoji.
The accounts in this fake follower network use a variety of repeated or otherwise formulaic biographies, some of which are repeated dozens or hundred of times. Some of the biographies begin with unnecessary leading commas, and a few consist entirely of punctuation.
It's presently unclear why, but over the past year someone has created a network of fake Facebook accounts pretending to be employees of the Los Angeles Dodgers. Many of the accounts in this network have GAN-generated faces.
cc: @ZellaQuixote
This network consists of (at least) 80 Facebook accounts, 48 of which use StyleGAN-generated faces as profile images. The remaining 32 all use the same image, a real photograph of a random person sitting in an office.
As is the case with all unmodified StyleGAN-generated faces, the main facial features (especially the eyes) are in the same position on all 48 AI-generated faces used by the network. This anomaly becomes obvious when the faces are blended together.
None of these chefs exist, as they're all AI-generated images. This hasn't stopped them from racking up lots of engagement on Facebook by posting AI-generated images of food (and occasional thoughts and prayers), however.
cc: @ZellaQuixote
These "chefs" are part of a network of 18 Facebook pages with names like "Cook Fastly" and "Emily Recipes" that continually post AI-generated images of food. While many of these pages claim to be US-based, they are have admins in Morocco per Facebook's Page Transparency feature.
Between them, these 18 Facebook "chef" pages have posted AI-generated images of food at least 36,000 times in the last five months. Not all of the images are unique; many have been posted repeatedly, sometimes by more than one of the alleged chefs.
Can simple text generation bots keep sophisticated LLM chatbots like ChatGPT engaged indefinitely? The answer is yes, which has some potentially interesting implications for distinguishing between conversational chatbots and humans.
For this experiment, four simple chatbots were created:
• a bot that asks the same question over and over
• a bot that replies with random fragments of a work of fiction
• a bot that asks randomly generated questions
• a bot that repeatedly asks "what do you mean by <X>?"
The output of these chatbots was used as input to an LLM chatbot based on the 8B version of the Llama 3.1 model. Three of the four bots were successful at engaging the LLM chatbot in a 1000-message exchange; the only one that failed was the repetitive question bot.