There's some weird stuff going on with the engagement on this tweet from @ilkersenock, including multiple retweets from accounts with GAN-generated faces. Let's take a look. #SundayShenaniGANs
The accounts with GAN-generated faces that retweeted the @ilkersenock tweet in question are part of a network of 18 accounts created on February 17th, 2023 with GAN-generated faces. All of their content thus far is retweets of @ilkersenock tweeted via the Twitter Web App.
Unmodified GAN-generated faces (at least, those in widespread use) have the distinctive trait that the major facial feature (particularly the eyes) are in the same pixel position on each image. This becomes obvious when one blends several such "faces" together.
Most of the accounts in this network have several hundred followers, but their followers are no more real than their artificially-generated "faces": almost all are empty accounts created in between November 2022 and March 2023.
The fake followers following the 18 GAN-faced accounts are from a massive fake follower network that's been turning up all over the place lately:
Come for the fake video, stay for the fake followers. It turns out that @ThePatriotOasis, an account that went viral earlier this week with a Biden deepfake video, recently gained approximately 8000 batch-created followers with zero tweets and zero likes.
If you get scammed and tweet about it, you might get a reply or three offering "help" from spammy accounts promoting the services of alleged "Certified Network Defender" @WariorInc.
(Tip: spammy Twitter accounts can't actually help you get your money back.)
At the time of this writing, at least 18 accounts (including two Twitter Blue verified accounts) are spamming replies advertising @WariorInc's "services". Although most of these accounts have been around for years, almost all of their content is replies from February/March 2023.
These accounts' replies are extremely repetitive, with replies recommending the "services" of @WariorInc being the most frequent. Some of the repeated replies mention "Security Analyst" @astra_cyberdesk or suspended account @apexhelpdesk rather than @WariorInc.
Tens of thousands of accounts have been created on Spoutible (a new social media platform powered by Bot Sentinel) over the last few weeks. What do all of them except one have in common? Answer: a Bot Sentinel score of 0% ("Normal").
As of early evening on Saturday, February 25th, 2023, there were 96208 publicly viewable Spoutible accounts, 96207 of which have Bot Sentinel scores of 0%. The sole exception is @/MyNameIs, which was one of the earliest accounts created.
The list of accounts and scores was obtained by programmatically triggering the functionality of the "LOAD MORE" button on the Spoutible People search with no filters, which appears to return all accounts in newest-first order (along with their Bot Sentinel scores).
For unknown reasons, someone has created 2 "America First" accounts with the same stolen photo: @Philipwalter74 and @FrankAd06414461. Both have the biography "No man is a man until he has been a soldier" and both have been following large US political accounts.
Both @Philipwalter74 and @FrankAd06414461 follow a similar set of large US political accounts. These accounts are a mix of popular influencers on both the left and right ends of the political spectrum.
Here's a Substack article on a shady account sales site (accs-market dot com) that @ZellaQuixote and I have been monitoring since August 2020. Needless to say, buying used social media accounts is generally not a good use of funds. conspirator0.substack.com/p/should-you-b…
Unsurprisingly, a lot of the accounts up for sale on websites like accs-market dot com have fake followers (at least, until both the accounts and their followers get banned for being fake).
It's not uncommon for the huge accounts that constantly go viral posting pretty pictures to get sold off on sites like accs-market dot com once they grow to a sufficient size:
Did you know that know that you must read to learn about the Silver Price Analysis? This network of spammy automated Twitter accounts says so, and if you can't trust a network of spammy automated Twitter accounts, who can you trust?
This network consists of 77 accounts created in either 2013, 2014, or 2022 with references to forex trading in their profiles and locations in Spain. These accounts have very similar tweet schedules and post the majority of their tweets via automation service IFTTT.
The content tweeted by this botnet is extremely repetitive, with many tweets duplicated verbatim by all 77 accounts in the network. The primary purpose of the botnet appears to be to promote forex trading website edge-forex(dot)info and its associated Telegram channel.