60 of @Dilde97512368's followers have GAN-generated face pics, but that's not the only pattern. 15 of its followers use cat pics, 7 use anime pics, and some of the pics that are neither GAN faces, cats, nor anime pics are repeated across two or three of its followers.
We explored the follower networks of @Dilde97512368's followers, and found a total of 3204 accounts with the same mix of profile pics (anime, cats, GAN-generated faces, and repeated images), all created January 25th or later.
Almost all of the accounts in this network either A) follow a three-digit number of accounts and have few or no followers or B) have a three-digit number of followers and follow few or no accounts. Very few of them follow accounts outside of the network.
972 of the accounts in this network use GAN-generated face pics, with the interesting twist that nearly all are female and all have blank backgrounds.
Here's an animated demonstration of blending all 972 pics together. As is the case with unmodified GAN-generated face pics, the major facial features are in the same location on each image, with the eye placement being particularly well-aligned.
257 of the accounts in this network use anime pics, many of which bear artifacts suggesting that they (like the fact pics) were GAN-generated (note the vestigial head on @Ceylan34886674 and the distortion/lack of nose and mouth on @KarakulahVedia).
We previously ran across GAN-generated anime pics via 4chan, and wrote a rudimentary detector. It flagged 165 of the 257 anime pics used by this network (64.2%) as GAN-generated.
307 of the accounts in this network use cat pics as their avatars. Like the face and anime pics, these contain artifacts indicating that they are likely GAN-generated. The "fur" on both @Gzdem17160119 and @Glben64660310's profile pics melts into the background, for example.
An additional 997 of these accounts use profile pics that are duplicated across multiple members of the network (sometimes dozens). 219 of these appear to be logos of some sort, some of them containing Turkish phone numbers. Finally, 671 members of the network have default pics.
One last observation on the images: we've seen a network using a mixture of GAN-generated faces, cat pics, and anime profile pics once before, in the form of the now-defunct Thousand Followers network.
What do these accounts do, other than follow each other? 218 have never tweeted; the remainder reply to other tweets, tweet original tweets containing images (mostly porn/escort service ads), and tweet repetitive tweets (again, mostly porn/adult services stuff).
This network replies en masse to specific tweets, with dozens or hundreds of members replying to the same tweet. They reply to two styles of tweet:
• porn tweets, including the network's own content
• online sports betting tweets
Footnote: here's a link to all of our existing research on the use GAN-generated pics on Twitter (and occasionally elsewhere).
How does one detect renamed Twitter accounts and find the previous names? There's no surefire way to do it, but here are four methods that sometimes work.
The first method is to do a Twitter search for old replies to the account in question. Use a search of this form to find replies prior to a given date and make sure to use "Latest" rather than "Top" results.
The previous name(s) will show up (sometimes alongside the current name) at the beginning of replies to the account's old tweets. This method doesn't always work, but it seems that when it does work, it works even if the tweets being replied to have been deleted.
Answer: @Mippcivzla's tweets (and those of a few other large Venezuelan accounts) are being retweeted by a network of 2454 accounts, all allegedly using the Twitter Android app. These accounts post almost no original content (99.6% percent of their tweets are retweets).
These accounts have all retweeted hundreds of tweets or more, but have liked few or none. The content they boost reflects this - 98.6% of the tweets they retweeted got more retweets than likes. (Based on a set of 5M random tweets, only ~0.8% of tweets get more RTs than likes.)
The 9 accounts promoting monsterfundrise(dot)com discussed in this previous thread have been shut down by Twitter, but 15 new ones have taken their place. As before, their tweets appear be being astroturfed, garnering far more retweets than likes.
We downloaded the set of accounts amplifying the monsterfundrise tweets, and noticed that many of the other tweets they retweeted (particularly recent tweets from Punjab, Pakistan governor @ChMSarwar) also received more retweets than likes.
(some background info on the presence of more retweets than likes being a sign of astroturfing - average ratio is more than twice as many likes as retweets)
This botnet consists of 53 accounts, all created on either January 1st or January 3rd, 2021. None has ever liked a tweet or followed an account, and almost all of them have no followers.
These accounts tweet almost exclusively via TweetDeck, with the exception of a small number of early tweets allegedly sent via the Twitter website. They fire off tweetstorms in tandem, and their schedules are all variations on a couple of basic patterns.
Twitter permanently banned @gatewaypundit yesterday. With it gone, here's what the retweet network for thegatewaypundit(dot)com looks like: two main clusters (English and Japanese), and the main nodes are @mei98862477, @kirstiealley, and @CassandraRules.
Japanese-language accounts have been turning up prominently in US-centric right wing Twitter content for a while. Back in January 2020, 23.3% of tweets linking now-defunct QAnon site qmap(dot)pub were in Japanese.
Around the time of the November 2020 US election, Sidney Powell, Lin Wood, and Rudy Giuliani all experienced a large influx of Japanese-language followers. #KrakenWorldTour2020
What do these eighteen blue-check accounts have in common (other than the profile graphics)? Answer: they're all old accounts that were recently renamed and redecorated, giving the appearance that they are official Baja California Sur government accounts.
Some examples, accompanied by screenshots of each account's former appearance courtesy of Internet archive:
• @Servicios122 (ID 49774815) was formerly @jess_noons
• @prod121 (ID 142177657) was formerly @AmyOguntala
• @secfin12 (ID 349157648) was formerly @lewaron
Here are the 18 renamed verified accounts that are currently presenting themselves as affiliated with the government of Baja California Sur, Mexico. Most appear to have originally belonged to random US residents, and it is unclear at what point they gained their blue checkmarks.